[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"blog-blogEn-stock-research-tools-that-cite-sec-filings":3,"blog-sibling-stock-research-tools-that-cite-sec-filings-blogEl":486,"blog-all-blogEn":908},{"id":4,"title":5,"author":6,"body":7,"date":457,"description":458,"extension":459,"faq":460,"featured":473,"mentionsTickers":473,"meta":474,"metaTitle":475,"navigation":476,"ogImage":27,"path":477,"readMinutes":478,"seo":479,"stem":480,"tags":481,"translationKey":484,"__hash__":485},"blogEn\u002Fblog\u002Fen\u002Fstock-research-tools-that-cite-sec-filings.md","Stock research tools that cite SEC filings: what \"cites the filing\" actually means, tool by tool","Nikolaos Drongitis",{"type":8,"value":9,"toc":437},"minimark",[10,28,32,35,40,53,59,70,73,77,80,106,109,112,116,119,124,127,130,133,136,140,143,146,149,153,156,159,162,166,169,172,175,179,182,185,189,192,195,199,202,205,209,212,215,219,222,233,240,243,247,421,425,428,431,434],[11,12,20],"div",{"className":13},[14,15,16,17,18,19],"not-prose","mb-8","rounded-xl","overflow-hidden","border","border-surface-3",[21,22],"img",{"alt":23,"className":24,"src":27},"Three ways a stock research tool can point at an SEC filing: name it, link it, or read the number from it",[25,26],"w-full","block","\u002Fblog\u002Fog-cite-sec-filings.svg",[29,30,31],"p",{},"\"Grounded in SEC filings.\" \"Every figure cited to its source.\" \"Click through to the filing.\" Almost every research tool built in the last two years says something like this, and the phrases sound interchangeable. They are not. Three very different mechanisms hide behind the same sentence, and which one a tool uses decides whether its numbers are checkable, or merely sound checkable.",[29,33,34],{},"This is a comparison of eight tools, plus the free baseline they all sit on, with every claim taken from the vendor's own website in September 2026. Ploutos AI is one of the eight, and the same rules apply to it as to the rest, including the part where it says what it does not do.",[36,37,39],"h2",{"id":38},"three-things-cites-the-filing-can-mean","Three things \"cites the filing\" can mean",[29,41,42,46,47,52],{},[43,44,45],"strong",{},"It names the filing."," The answer says \"according to the latest 10-K, revenue grew 12%.\" There is no link. The language model may well have read the 10-K, or may have read a news article about it, or may have produced the most plausible sentence. You cannot tell which, and the ",[48,49,51],"a",{"href":50},"\u002Fblog\u002Fwhy-ai-stock-tools-hallucinate","failure mode is well documented",".",[29,54,55,58],{},[43,56,57],{},"It links to the filing."," The same sentence carries a link: to the document, to a section, or in the best case to the exact passage. Now you can check. But the number in the sentence was still read by the model out of prose or out of an HTML table, and the model can misread a table just as a person can. The link tells you where to look, not that the figure is right.",[29,60,61,64,65,69],{},[43,62,63],{},"It reads the number from the filing's structured data."," Every 10-K and 10-Q comes with an XBRL layer: each figure on the financial statements carries a machine-readable tag, the period it belongs to, and the unit. A tool that pulls ",[66,67,68],"code",{},"Revenues"," for fiscal 2025 from that layer has a number the company itself tagged, and the language model never typed it. The model reasons about the figure; it does not produce it. This is the strongest form, and it is also the one vendors describe least clearly, because it is plumbing rather than a feature.",[29,71,72],{},"A fourth question sits underneath all three: is the number from the filing at all, or from a data vendor who read the filing and standardised it? Standardised data is not wrong, it is often more comparable across companies, but \"from SEC filings\" is then true of the vendor, not of the tool.",[36,74,76],{"id":75},"the-free-baseline-sec-edgar-itself","The free baseline: SEC EDGAR itself",[29,78,79],{},"Everything below is built on the SEC's own systems, and they are free.",[81,82,83,90,100],"ul",{},[84,85,86,89],"li",{},[43,87,88],{},"Full-text search"," covers every filing submitted electronically since 2001, including exhibits. Boolean and phrase queries only; the SEC states plainly that natural-language search is not supported.",[84,91,92,95,96,99],{},[43,93,94],{},"The XBRL APIs."," The ",[66,97,98],{},"companyfacts"," endpoint returns every tagged figure a company has ever reported, from 10-K, 10-Q, 8-K, 20-F, 40-F and 6-K forms, in one JSON file. No authentication, no key, updated within a minute of the filing, with a documented limit of 10 requests per second and a required User-Agent header.",[84,101,102,105],{},[43,103,104],{},"Financial Statement Data Sets",", quarterly bulk files of the same numbers, from 2009 to today, \"as filed.\"",[29,107,108],{},"Two things the free layer does not cover: Form 4 insider transactions and 13F institutional holdings are not in the XBRL APIs. They are separate feeds with their own structure, which is why tools that offer them had to do extra work.",[29,110,111],{},"If you want to check a number yourself and have an afternoon, this is where you go. Every tool below is selling you the afternoon.",[36,113,115],{"id":114},"the-eight-tools","The eight tools",[29,117,118],{},"Prices and plans are left out on purpose: they change faster than an article does, and every vendor's pricing page is one click away. The mechanism is what matters.",[120,121,123],"h3",{"id":122},"alphasense-with-bamsec","AlphaSense, with BamSEC",[29,125,126],{},"AlphaSense is the institutional platform: 500 million documents, filings from 68,000 companies worldwide, broker research, transcripts. Its citation claim is the most specific in this list: \"every AI-generated insight links directly to the exact sentence in the source document.\" That is level two at its best, sentence-level, and their developer API shows inline markers that resolve to the source.",[29,128,129],{},"Numbers come from a different route. Since October 2025 the platform carries financial data built on Canalyst models, \"4,500+ ready-to-use models,\" and every figure keeps an audit link to the original source. Whether a number inside a chat answer came from a model field or from text is not stated.",[29,131,132],{},"BamSEC, which AlphaSense acquired through Tegus in 2024, is the opposite kind of tool: no language model at all. It is a reader for 10-Ks, 10-Qs, proxies and transcripts, with tables that \"come straight from the filings with no modifications,\" one-click Excel export, and human-made links to a selected passage or table.",[29,134,135],{},"AlphaSense is sold to hedge funds, banks and corporates. Its help centre lists Greek among 37 languages for filtering non-English documents, but the interface is English.",[120,137,139],{"id":138},"fiscalai-formerly-finchat","Fiscal.ai, formerly FinChat",[29,141,142],{},"Fiscal.ai renamed from FinChat after a 10 million dollar Series A and now positions itself as \"financial data infrastructure\" with a research terminal on top. Its claim: \"Source to Filing. Click-through auditability for every figure. Delivered in Image, PDF and URLs.\" A number links to the filing page it was taken from.",[29,144,145],{},"Where the number came from is stated more clearly than most: for US, Canadian, UK and EU companies plus ADRs the data is Fiscal.ai's own extraction \"with human analyst verification,\" updated within minutes of earnings; for other geographies it is S&P Capital IQ. XBRL is not mentioned anywhere on the site. AI summaries got \"citation support\" in May 2026; the granularity of a chat citation is not described.",[29,147,148],{},"Coverage is 100,000 companies globally. Filing click-through and as-reported statements are on the top plan only. English primarily.",[120,150,152],{"id":151},"equibles-and-alvis","Equibles and ALVIS",[29,154,155],{},"Equibles is the smallest company here and the one whose methodology page answers the structured-versus-text question directly: \"Fundamentals come from tagged XBRL, the company's own numbers.\" Narrative items, KPIs and non-GAAP bridges are extracted from the written filing, checked by \"an independent verifier,\" and carry the verbatim source quote. Each figure carries the form, filing date and period it came from.",[29,157,158],{},"It is US-listed only, including foreign issuers that file 20-F and 6-K. The corpus covers 10-K, 10-Q, 8-K, 13F, Forms 3, 4 and 5, 13D and 13G. ALVIS is the chat layer, \"a citation for every figure.\" The MCP server is open source and connects to Claude and ChatGPT.",[29,160,161],{},"Its own terms are unusually candid: AI responses \"can be incomplete, outdated, or incorrect even when they include citations.\"",[120,163,165],{"id":164},"stockportfoliopro","StockPortfolio.pro",[29,167,168],{},"StockPortfolio.pro promises \"every figure cited to its SEC filing\" and \"no estimates.\" The metrics are computed deterministically from filed statements, \"no AI judgement,\" rebuilt nightly. It reads 10-K, 10-Q, 8-K, 13F and Form 4.",[29,170,171],{},"The citation is at the filing level, not the figure or section level. In practice a stock page says \"Source: Apple Inc. SEC filings (10-K),\" and the public claim ledger links most rows to an EDGAR search page for the ticker rather than to a document. Whether extraction is XBRL or text is not stated. Coverage is US companies reporting in dollars, with limited ADR support.",[29,173,174],{},"English, plus a Simplified Chinese interface.",[120,176,178],{"id":177},"last10k","Last10K",[29,180,181],{},"Last10K is a filing reader with summaries, not a chat tool. Each filing page links back to the document on sec.gov, offers a table of contents by section, and a year-over-year \"track changes\" view showing what the company added, removed or reworded. Highlights are tagged by theme but do not cite item numbers or pages.",[29,183,184],{},"There is no AI answer feature, so there is nothing to hallucinate; the trade-off is that the reading is still yours. English.",[120,186,188],{"id":187},"simply-wall-st","Simply Wall St",[29,190,191],{},"Simply Wall St is the most popular retail platform on this list, seven million registered users, and it does not cite filings. \"All of our data comes from S&P Global Market Intelligence,\" whose upstream sources include EDGAR. Figures carry the S&P attribution, and the help centre tells you to verify against the regulator's filings yourself.",[29,193,194],{},"That is not a criticism of the data; it is a description of the mechanism. Numbers are vendor-standardised, presented as trailing twelve months, and the analysis model is public on GitHub. Coverage is 120,000 stocks in 90 markets. English, Japanese, German, Spanish and Korean.",[120,196,198],{"id":197},"tikr","TIKR",[29,200,201],{},"TIKR sits on S&P Capital IQ and Morningstar, with Financial Modeling Prep as a secondary set. The support pages say it directly: line items \"can be recategorized by CapitalIQ for better comparability.\" Filings and ownership data are available as tools inside the terminal; 13F tracking goes \"beyond just 13F filings\" to global disclosures from 10,000 investors.",[29,203,204],{},"Its AI product, Research Hub, is a beta on the Ultimate plan covering \"a curated set of 100 names,\" with the disclaimer that it \"uses auto-generated content which can have mistakes.\" No citation claim is made. The plan without global coverage is US-only. English.",[120,206,208],{"id":207},"fintool","Fintool",[29,210,211],{},"Fintool belongs in this list for what it claimed and for what happened to it. Through early 2026 it advertised page-level citations, \"Apple Inc. Form 10-K (2023), Page 23, verified by 3\u002F3 agents,\" an XBRL parser feeding a language model, and coverage of 8,000 US companies with filings indexed within a minute. It was sold to investment professionals.",[29,213,214],{},"In April 2026 the founder announced that the team and IP had joined Microsoft. As of September 2026 every page on fintool.com redirects to Microsoft 365. There is nothing to sign up for.",[120,216,218],{"id":217},"ploutos-ai","Ploutos AI",[29,220,221],{},"Ploutos AI is ours, so here is the same set of facts in the same order.",[29,223,224,225,227,228,232],{},"Numbers come from the SEC's structured XBRL data, the ",[66,226,98],{}," layer described above, not from the model. The model is handed the figures and reasons about them; when the data is not there it ",[48,229,231],{"href":230},"\u002Fblog\u002Fwhy-we-refuse-to-analyze-a-stock","stops rather than guesses",". Insider transactions from Form 4, material events from 8-K filings and the quality flags they trigger link to the exact filing on EDGAR. Derived figures do not link anywhere: a fair value estimate, a ten-year average or a trailing sum is built from several filings and no single document states it, so putting a link on it would be decoration. The method is stated instead.",[29,234,235,236,52],{},"After the analysis is written, a separate Devil's Advocate pass argues the bear case in writing, and it is published next to the bull case. The ",[48,237,239],{"href":238},"\u002Fblog\u002Fhow-ploutos-analyzes-a-stock","pipeline is described in full",[29,241,242],{},"Coverage is US-listed stocks only, NYSE, NASDAQ and AMEX. No earnings-call transcripts, no broker research, no non-US listings. Output in English or Greek. Research, not investment advice.",[36,244,246],{"id":245},"side-by-side","Side by side",[248,249,250,269],"table",{},[251,252,253],"thead",{},[254,255,256,260,263,266],"tr",{},[257,258,259],"th",{},"Tool",[257,261,262],{},"Where the numbers come from",[257,264,265],{},"What a citation points to",[257,267,268],{},"Coverage",[270,271,272,287,301,315,329,343,356,368,381,394,407],"tbody",{},[254,273,274,278,281,284],{},[275,276,277],"td",{},"SEC EDGAR",[275,279,280],{},"The filing's XBRL, as filed",[275,282,283],{},"The document itself",[275,285,286],{},"All SEC filers",[254,288,289,292,295,298],{},[275,290,291],{},"AlphaSense",[275,293,294],{},"Canalyst models plus text",[275,296,297],{},"The exact sentence",[275,299,300],{},"68,000 companies, global",[254,302,303,306,309,312],{},[275,304,305],{},"BamSEC",[275,307,308],{},"Tables straight from filings",[275,310,311],{},"A passage or table you pick",[275,313,314],{},"SEC filers",[254,316,317,320,323,326],{},[275,318,319],{},"Fiscal.ai",[275,321,322],{},"Own extraction, human-checked; S&P Capital IQ outside US\u002FCA\u002FUK\u002FEU",[275,324,325],{},"The filing page, on Max",[275,327,328],{},"100,000 companies, global",[254,330,331,334,337,340],{},[275,332,333],{},"Equibles",[275,335,336],{},"Tagged XBRL; verified text extraction with quote",[275,338,339],{},"Form, date, period and quote per figure",[275,341,342],{},"US-listed",[254,344,345,347,350,353],{},[275,346,165],{},[275,348,349],{},"Computed from filed statements, method unstated",[275,351,352],{},"The filing, or an EDGAR search",[275,354,355],{},"US, USD reporters",[254,357,358,360,363,366],{},[275,359,178],{},[275,361,362],{},"The filing, read by you",[275,364,365],{},"The document on sec.gov",[275,367,314],{},[254,369,370,372,375,378],{},[275,371,188],{},[275,373,374],{},"S&P Global Market Intelligence",[275,376,377],{},"Nothing; verify yourself",[275,379,380],{},"120,000 stocks, global",[254,382,383,385,388,391],{},[275,384,198],{},[275,386,387],{},"S&P Capital IQ, Morningstar, FMP",[275,389,390],{},"None claimed",[275,392,393],{},"100,000 stocks, global",[254,395,396,398,401,404],{},[275,397,208],{},[275,399,400],{},"XBRL parser plus text, historically",[275,402,403],{},"Filing and page, historically",[275,405,406],{},"8,000 US companies, historically",[254,408,409,411,416,419],{},[275,410,218],{},[275,412,413,414],{},"SEC XBRL ",[66,415,98],{},[275,417,418],{},"Exact filing for Form 4, 8-K, quality flags; none for derived figures",[275,420,342],{},[36,422,424],{"id":423},"how-to-read-this-table","How to read this table",[29,426,427],{},"If your question is \"can I check this number,\" the column that matters is the third one, and the honest answer for most tools is \"you can check the document, not the number.\" Sentence-level links, which AlphaSense claims and Equibles approximates with quotes, are the exception.",[29,429,430],{},"If your question is \"did a language model type this number,\" the column that matters is the second one. Equibles says XBRL. Ploutos AI says XBRL. Fiscal.ai says human-verified extraction. Fintool said XBRL and no longer exists. Everyone else is either a reader, a vendor feed, or does not say.",[29,432,433],{},"If your question is \"which one should I use,\" this article does not answer it, on purpose. The mechanisms are different because the jobs are different. A reader like Last10K or BamSEC is for people who will do the reading. A vendor-fed terminal like TIKR or Simply Wall St is for people who want comparable numbers across a hundred markets and are fine trusting S&P to have read the filing. The citation-first tools, Equibles, Fiscal.ai, StockPortfolio.pro and Ploutos AI, are for people who want the number and the way to check it in the same place, and they differ mainly in geography and language.",[29,435,436],{},"Whatever you use, ask it the question this article is built on: when you show me a number, did you read it, or did you look it up? A tool that cannot answer that is naming the filing, not citing it.",{"title":438,"searchDepth":439,"depth":439,"links":440},"",3,[441,443,444,455,456],{"id":38,"depth":442,"text":39},2,{"id":75,"depth":442,"text":76},{"id":114,"depth":442,"text":115,"children":445},[446,447,448,449,450,451,452,453,454],{"id":122,"depth":439,"text":123},{"id":138,"depth":439,"text":139},{"id":151,"depth":439,"text":152},{"id":164,"depth":439,"text":165},{"id":177,"depth":439,"text":178},{"id":187,"depth":439,"text":188},{"id":197,"depth":439,"text":198},{"id":207,"depth":439,"text":208},{"id":217,"depth":439,"text":218},{"id":245,"depth":442,"text":246},{"id":423,"depth":442,"text":424},"2026-09-07","Eight tools say their numbers come from SEC filings. They mean three different things. Where each one's figures actually come from, what a citation points to, and what each covers, checked on the vendors' own pages in September 2026.","md",[461,464,467,470],{"q":462,"a":463},"What does it mean when a stock research tool says it cites SEC filings?","One of three things. It names the filing in the text, it links to the filing or to a passage in it, or it reads the number from the filing's structured XBRL data so the language model never types the figure. Most tools do the first, several do the second, few say clearly whether they do the third.",{"q":465,"a":466},"Which tools link every figure back to the filing?","Fiscal.ai and Equibles state this on their own pages, and AlphaSense links each AI sentence to the source passage. StockPortfolio.pro links at the filing level. Ploutos AI links insider transactions, 8-K events and quality flags to the exact filing, but not derived figures such as fair value.",{"q":468,"a":469},"Is there a free way to get the numbers straight from SEC filings?","Yes. The SEC's own XBRL API returns every figure a company has tagged in its 10-K and 10-Q filings, free, without an API key, updated within a minute of the filing. It is what several of these tools are built on.",{"q":471,"a":472},"Do any of these tools work in Greek?","Ploutos AI produces its analysis in Greek and English. Simply Wall St has Japanese, German, Spanish and Korean. The others are English only.",false,{},"Stock research tools that cite SEC filings",true,"\u002Fblog\u002Fstock-research-tools-that-cite-sec-filings",11,{"title":5,"description":458},"blog\u002Fen\u002Fstock-research-tools-that-cite-sec-filings",[482,483],"product","methodology","stock-research-tools-that-cite-sec-filings","89cOub1tHxUT3hVKI5OpuH7zmtc3tCGo2Liteq7j_Lc",{"id":487,"title":488,"author":6,"body":489,"date":457,"description":887,"extension":459,"faq":888,"featured":473,"mentionsTickers":473,"meta":901,"metaTitle":902,"navigation":476,"ogImage":27,"path":903,"readMinutes":478,"seo":904,"stem":905,"tags":906,"translationKey":484,"__hash__":907},"blogEl\u002Fblog\u002Fel\u002Fstock-research-tools-that-cite-sec-filings.md","Εργαλεία ανάλυσης μετοχών που παραπέμπουν σε καταθέσεις SEC: τι σημαίνει πραγματικά «παραπέμπει», εργαλείο προς εργαλείο",{"type":8,"value":490,"toc":870},[491,498,501,504,508,518,524,533,536,540,543,566,569,572,576,579,583,586,589,592,595,599,602,605,608,612,615,618,621,623,626,629,632,634,637,640,642,645,648,650,653,656,658,661,664,666,669,680,687,690,694,854,858,861,864,867],[11,492,494],{"className":493},[14,15,16,17,18,19],[21,495],{"alt":496,"className":497,"src":27},"Τρεις τρόποι με τους οποίους ένα εργαλείο ανάλυσης μπορεί να δείξει μια κατάθεση SEC: να την ονομάσει, να δώσει link, ή να διαβάσει το νούμερο από αυτήν",[25,26],[29,499,500],{},"«Βασισμένο σε καταθέσεις SEC.» «Κάθε νούμερο με παραπομπή στην πηγή του.» «Κάνε κλικ και δες την κατάθεση.» Σχεδόν κάθε εργαλείο έρευνας που χτίστηκε τα τελευταία δύο χρόνια λέει κάτι τέτοιο, και οι φράσεις ακούγονται ίδιες. Δεν είναι. Πίσω από την ίδια πρόταση κρύβονται τρεις πολύ διαφορετικοί μηχανισμοί, και το ποιον χρησιμοποιεί ένα εργαλείο καθορίζει αν τα νούμερά του ελέγχονται, ή απλώς ακούγονται ελέγξιμα.",[29,502,503],{},"Αυτή είναι μια σύγκριση οκτώ εργαλείων, συν τη δωρεάν βάση πάνω στην οποία πατάνε όλα, με κάθε ισχυρισμό παρμένο από την ιστοσελίδα της ίδιας της εταιρείας τον Σεπτέμβριο του 2026. Το Ploutos AI είναι ένα από τα οκτώ, και ισχύουν γι' αυτό οι ίδιοι κανόνες όπως για τα υπόλοιπα, μαζί με το κομμάτι όπου λέει τι δεν κάνει.",[36,505,507],{"id":506},"τρία-πράγματα-που-μπορεί-να-σημαίνει-το-παραπέμπει-στην-κατάθεση","Τρία πράγματα που μπορεί να σημαίνει το «παραπέμπει στην κατάθεση»",[29,509,510,513,514,52],{},[43,511,512],{},"Ονομάζει την κατάθεση."," Η απάντηση λέει «σύμφωνα με το τελευταίο 10-K, τα έσοδα αυξήθηκαν 12%». Δεν υπάρχει link. Το γλωσσικό μοντέλο μπορεί όντως να διάβασε το 10-K, μπορεί να διάβασε ένα άρθρο για αυτό, μπορεί και να παρήγαγε την πιο εύλογη πρόταση. Δεν μπορείς να ξέρεις ποιο από τα τρία, και ",[48,515,517],{"href":516},"\u002Fel\u002Fblog\u002Fwhy-ai-stock-tools-hallucinate","το πρόβλημα είναι καλά τεκμηριωμένο",[29,519,520,523],{},[43,521,522],{},"Δίνει link στην κατάθεση."," Η ίδια πρόταση έχει link: στο έγγραφο, σε μια ενότητα, ή στην καλύτερη περίπτωση στο ακριβές απόσπασμα. Τώρα μπορείς να ελέγξεις. Αλλά το νούμερο στην πρόταση το διάβασε πάλι το μοντέλο από πεζό κείμενο ή από έναν πίνακα HTML, και το μοντέλο μπορεί να διαβάσει λάθος έναν πίνακα όπως και ένας άνθρωπος. Το link σού λέει πού να κοιτάξεις, όχι ότι ο αριθμός είναι σωστός.",[29,525,526,529,530,532],{},[43,527,528],{},"Διαβάζει το νούμερο από τα δομημένα δεδομένα της κατάθεσης."," Κάθε 10-K και 10-Q έρχεται με ένα στρώμα XBRL: κάθε μέγεθος των οικονομικών καταστάσεων έχει μια ετικέτα αναγνώσιμη από μηχανή, την περίοδο στην οποία ανήκει, και τη μονάδα. Ένα εργαλείο που τραβάει το ",[66,531,68],{}," της χρήσης 2025 από αυτό το στρώμα έχει έναν αριθμό που σήμανε η ίδια η εταιρεία, και το γλωσσικό μοντέλο δεν τον πληκτρολόγησε ποτέ. Το μοντέλο σκέφτεται πάνω στον αριθμό, δεν τον παράγει. Αυτή είναι η ισχυρότερη μορφή, και είναι επίσης αυτή που οι εταιρείες περιγράφουν λιγότερο καθαρά, γιατί είναι υδραυλικά και όχι χαρακτηριστικό.",[29,534,535],{},"Μια τέταρτη ερώτηση κάθεται κάτω από τις τρεις: το νούμερο είναι καθόλου από την κατάθεση, ή από έναν πάροχο δεδομένων που διάβασε την κατάθεση και το τυποποίησε; Τα τυποποιημένα δεδομένα δεν είναι λάθος, συχνά είναι πιο συγκρίσιμα μεταξύ εταιρειών, αλλά τότε το «από καταθέσεις SEC» ισχύει για τον πάροχο, όχι για το εργαλείο.",[36,537,539],{"id":538},"η-δωρεάν-βάση-το-ίδιο-το-sec-edgar","Η δωρεάν βάση: το ίδιο το SEC EDGAR",[29,541,542],{},"Όλα τα παρακάτω είναι χτισμένα πάνω στα συστήματα της ίδιας της SEC, και είναι δωρεάν.",[81,544,545,551,560],{},[84,546,547,550],{},[43,548,549],{},"Η αναζήτηση πλήρους κειμένου"," καλύπτει κάθε κατάθεση που υποβλήθηκε ηλεκτρονικά από το 2001, μαζί με τα παραρτήματα. Μόνο λογικοί τελεστές και φράσεις. Η SEC δηλώνει ξεκάθαρα ότι αναζήτηση σε φυσική γλώσσα δεν υποστηρίζεται.",[84,552,553,556,557,559],{},[43,554,555],{},"Τα APIs XBRL."," Το endpoint ",[66,558,98],{}," επιστρέφει κάθε σημασμένο νούμερο που έχει αναφέρει ποτέ μια εταιρεία, από έντυπα 10-K, 10-Q, 8-K, 20-F, 40-F και 6-K, σε ένα αρχείο JSON. Χωρίς ταυτοποίηση, χωρίς κλειδί, ενημερωμένο μέσα σε ένα λεπτό από την κατάθεση, με τεκμηριωμένο όριο 10 αιτήματα το δευτερόλεπτο και υποχρεωτική κεφαλίδα User-Agent.",[84,561,562,565],{},[43,563,564],{},"Τα Financial Statement Data Sets",", τριμηνιαία μαζικά αρχεία με τα ίδια νούμερα, από το 2009 μέχρι σήμερα, «όπως κατατέθηκαν».",[29,567,568],{},"Δύο πράγματα που το δωρεάν στρώμα δεν καλύπτει: οι συναλλαγές insiders του Form 4 και οι θεσμικές θέσεις του 13F δεν είναι στα APIs XBRL. Είναι ξεχωριστές ροές με δική τους δομή, και γι' αυτό τα εργαλεία που τις προσφέρουν χρειάστηκε να κάνουν επιπλέον δουλειά.",[29,570,571],{},"Αν θες να ελέγξεις ένα νούμερο μόνος σου και έχεις ένα απόγευμα, εδώ πας. Κάθε εργαλείο παρακάτω σού πουλάει το απόγευμα.",[36,573,575],{"id":574},"τα-οκτώ-εργαλεία","Τα οκτώ εργαλεία",[29,577,578],{},"Οι τιμές και τα πλάνα λείπουν επίτηδες: αλλάζουν πιο γρήγορα από ένα άρθρο, και η σελίδα τιμολόγησης κάθε εταιρείας είναι ένα κλικ μακριά. Ο μηχανισμός είναι αυτό που μετράει.",[120,580,582],{"id":581},"alphasense-με-το-bamsec","AlphaSense, με το BamSEC",[29,584,585],{},"Το AlphaSense είναι η θεσμική πλατφόρμα: 500 εκατομμύρια έγγραφα, καταθέσεις από 68.000 εταιρείες παγκοσμίως, εκθέσεις χρηματιστηριακών, απομαγνητοφωνήσεις. Ο ισχυρισμός του για τις παραπομπές είναι ο πιο συγκεκριμένος της λίστας: «κάθε συμπέρασμα που παράγει το AI συνδέεται απευθείας με την ακριβή πρόταση στο έγγραφο πηγής». Αυτό είναι το δεύτερο επίπεδο στην καλύτερη μορφή του, σε επίπεδο πρότασης, και το API για προγραμματιστές δείχνει ενσωματωμένους δείκτες που οδηγούν στην πηγή.",[29,587,588],{},"Τα νούμερα έρχονται από άλλη διαδρομή. Από τον Οκτώβριο του 2025 η πλατφόρμα έχει οικονομικά δεδομένα χτισμένα πάνω σε μοντέλα της Canalyst, «4.500+ έτοιμα μοντέλα», και κάθε μέγεθος κρατάει link ελέγχου προς την αρχική πηγή. Αν ένα νούμερο μέσα σε μια απάντηση chat ήρθε από πεδίο μοντέλου ή από κείμενο, δεν δηλώνεται.",[29,590,591],{},"Το BamSEC, που το AlphaSense απέκτησε μέσω της Tegus το 2024, είναι το αντίθετο είδος εργαλείου: κανένα γλωσσικό μοντέλο. Είναι αναγνώστης για 10-K, 10-Q, proxies και απομαγνητοφωνήσεις, με πίνακες που «έρχονται κατευθείαν από τις καταθέσεις χωρίς τροποποιήσεις», εξαγωγή σε Excel με ένα κλικ, και links φτιαγμένα από ανθρώπους σε επιλεγμένο απόσπασμα ή πίνακα.",[29,593,594],{},"Το AlphaSense πωλείται σε hedge funds, τράπεζες και μεγάλες εταιρείες. Το κέντρο βοήθειάς του αναφέρει τα ελληνικά ανάμεσα σε 37 γλώσσες για φιλτράρισμα μη αγγλικών εγγράφων, αλλά το περιβάλλον είναι αγγλικό.",[120,596,598],{"id":597},"fiscalai-πρώην-finchat","Fiscal.ai, πρώην FinChat",[29,600,601],{},"Το Fiscal.ai μετονομάστηκε από FinChat μετά από Series A 10 εκατομμυρίων δολαρίων και τοποθετείται πλέον ως «υποδομή οικονομικών δεδομένων» με ένα terminal έρευνας από πάνω. Ο ισχυρισμός του: «Πηγή προς κατάθεση. Ελεγξιμότητα με κλικ για κάθε νούμερο. Σε εικόνα, PDF και URL.» Ένα νούμερο συνδέεται με τη σελίδα της κατάθεσης από την οποία πάρθηκε.",[29,603,604],{},"Το από πού ήρθε το νούμερο δηλώνεται πιο καθαρά από τα περισσότερα: για εταιρείες ΗΠΑ, Καναδά, Ηνωμένου Βασιλείου και ΕΕ συν τα ADR, τα δεδομένα είναι δική τους εξαγωγή «με επαλήθευση από αναλυτή», ενημερωμένα μέσα σε λεπτά από τα αποτελέσματα. Για τις υπόλοιπες γεωγραφίες είναι S&P Capital IQ. Το XBRL δεν αναφέρεται πουθενά στο site. Οι περιλήψεις AI απέκτησαν «υποστήριξη παραπομπών» τον Μάιο του 2026. Η ανάλυση μιας παραπομπής σε chat δεν περιγράφεται.",[29,606,607],{},"Η κάλυψη είναι 100.000 εταιρείες παγκοσμίως. Το κλικ προς την κατάθεση και οι καταστάσεις όπως αναφέρθηκαν είναι μόνο στο ανώτερο πλάνο. Κυρίως αγγλικά.",[120,609,611],{"id":610},"equibles-και-alvis","Equibles και ALVIS",[29,613,614],{},"Το Equibles είναι η μικρότερη εταιρεία εδώ και αυτή που η σελίδα μεθοδολογίας της απαντάει ευθέως στην ερώτηση δομημένα-ή-κείμενο: «Τα θεμελιώδη έρχονται από σημασμένο XBRL, τα νούμερα της ίδιας της εταιρείας.» Τα αφηγηματικά στοιχεία, οι δείκτες KPI και οι γέφυρες non-GAAP εξάγονται από τη γραπτή κατάθεση, ελέγχονται από «ανεξάρτητο επαληθευτή», και κουβαλούν το αυτούσιο απόσπασμα της πηγής. Κάθε νούμερο φέρει το έντυπο, την ημερομηνία κατάθεσης και την περίοδο από την οποία προήλθε.",[29,616,617],{},"Είναι μόνο για εισηγμένες στις ΗΠΑ, μαζί με ξένους εκδότες που καταθέτουν 20-F και 6-K. Το σώμα εγγράφων καλύπτει 10-K, 10-Q, 8-K, 13F, Forms 3, 4 και 5, 13D και 13G. Το ALVIS είναι το στρώμα chat, «μια παραπομπή για κάθε νούμερο». Ο MCP server είναι ανοιχτού κώδικα και συνδέεται με Claude και ChatGPT.",[29,619,620],{},"Οι όροι χρήσης του είναι ασυνήθιστα ειλικρινείς: οι απαντήσεις AI «μπορεί να είναι ελλιπείς, παρωχημένες ή λανθασμένες ακόμα και όταν περιλαμβάνουν παραπομπές».",[120,622,165],{"id":164},[29,624,625],{},"Το StockPortfolio.pro υπόσχεται «κάθε νούμερο με παραπομπή στην κατάθεσή του στην SEC» και «χωρίς εκτιμήσεις». Οι δείκτες υπολογίζονται ντετερμινιστικά από τις κατατεθειμένες καταστάσεις, «χωρίς κρίση AI», και ξαναχτίζονται κάθε βράδυ. Διαβάζει 10-K, 10-Q, 8-K, 13F και Form 4.",[29,627,628],{},"Η παραπομπή είναι σε επίπεδο κατάθεσης, όχι νούμερου ή ενότητας. Στην πράξη μια σελίδα μετοχής λέει «Πηγή: καταθέσεις SEC της Apple Inc. (10-K)», και το δημόσιο μητρώο ισχυρισμών συνδέει τις περισσότερες γραμμές με μια σελίδα αναζήτησης EDGAR για το ticker και όχι με έγγραφο. Αν η εξαγωγή είναι XBRL ή κείμενο δεν δηλώνεται. Η κάλυψη είναι εταιρείες ΗΠΑ που αναφέρουν σε δολάρια, με περιορισμένη υποστήριξη ADR.",[29,630,631],{},"Αγγλικά, συν περιβάλλον σε απλοποιημένα κινεζικά.",[120,633,178],{"id":177},[29,635,636],{},"Το Last10K είναι αναγνώστης καταθέσεων με περιλήψεις, όχι εργαλείο chat. Κάθε σελίδα κατάθεσης συνδέεται πίσω με το έγγραφο στο sec.gov, προσφέρει πίνακα περιεχομένων ανά ενότητα, και μια όψη «παρακολούθηση αλλαγών» χρόνο με χρόνο που δείχνει τι πρόσθεσε, αφαίρεσε ή ξαναδιατύπωσε η εταιρεία. Τα σημεία ενδιαφέροντος έχουν ετικέτες ανά θέμα αλλά δεν παραπέμπουν σε αριθμό ενότητας ή σελίδα.",[29,638,639],{},"Δεν υπάρχει λειτουργία απαντήσεων AI, οπότε δεν υπάρχει τίποτα να επινοηθεί. Το αντάλλαγμα είναι ότι το διάβασμα παραμένει δικό σου. Αγγλικά.",[120,641,188],{"id":187},[29,643,644],{},"Το Simply Wall St είναι η δημοφιλέστερη πλατφόρμα λιανικής της λίστας, επτά εκατομμύρια εγγεγραμμένοι χρήστες, και δεν παραπέμπει σε καταθέσεις. «Όλα τα δεδομένα μας προέρχονται από την S&P Global Market Intelligence», της οποίας οι πηγές περιλαμβάνουν το EDGAR. Τα νούμερα φέρουν την αναφορά στην S&P, και το κέντρο βοήθειας σού λέει να επαληθεύσεις μόνος σου με τις καταθέσεις στην εποπτική αρχή.",[29,646,647],{},"Αυτό δεν είναι κριτική στα δεδομένα. Είναι περιγραφή του μηχανισμού. Τα νούμερα είναι τυποποιημένα από τον πάροχο, παρουσιάζονται ως κυλιόμενο δωδεκάμηνο, και το μοντέλο ανάλυσης είναι δημόσιο στο GitHub. Η κάλυψη είναι 120.000 μετοχές σε 90 αγορές. Αγγλικά, ιαπωνικά, γερμανικά, ισπανικά και κορεατικά.",[120,649,198],{"id":197},[29,651,652],{},"Το TIKR πατάει σε S&P Capital IQ και Morningstar, με το Financial Modeling Prep ως δευτερεύον σύνολο. Οι σελίδες υποστήριξης το λένε ευθέως: οι γραμμές «μπορεί να ανακατηγοριοποιηθούν από την CapitalIQ για καλύτερη συγκρισιμότητα». Καταθέσεις και δεδομένα ιδιοκτησίας υπάρχουν ως εργαλεία μέσα στο terminal. Η παρακολούθηση 13F πάει «πέρα από τις καταθέσεις 13F» σε παγκόσμιες γνωστοποιήσεις 10.000 επενδυτών.",[29,654,655],{},"Το προϊόν AI του, το Research Hub, είναι beta στο πλάνο Ultimate και καλύπτει «ένα επιλεγμένο σύνολο 100 ονομάτων», με τη δήλωση ότι «χρησιμοποιεί αυτόματα παραγόμενο περιεχόμενο που μπορεί να έχει λάθη». Δεν γίνεται κανένας ισχυρισμός για παραπομπές. Το πλάνο χωρίς παγκόσμια κάλυψη είναι μόνο ΗΠΑ. Αγγλικά.",[120,657,208],{"id":207},[29,659,660],{},"Το Fintool ανήκει στη λίστα για αυτό που ισχυριζόταν και για αυτό που του συνέβη. Μέχρι τις αρχές του 2026 διαφήμιζε παραπομπές σε επίπεδο σελίδας, «Apple Inc. Form 10-K (2023), σελίδα 23, επαληθευμένο από 3\u002F3 agents», έναν αναλυτή XBRL που τροφοδοτούσε γλωσσικό μοντέλο, και κάλυψη 8.000 εταιρειών ΗΠΑ με καταθέσεις ευρετηριασμένες μέσα σε ένα λεπτό. Πωλούνταν σε επαγγελματίες επενδυτές.",[29,662,663],{},"Τον Απρίλιο του 2026 ο ιδρυτής ανακοίνωσε ότι η ομάδα και η πνευματική ιδιοκτησία πέρασαν στη Microsoft. Τον Σεπτέμβριο του 2026 κάθε σελίδα του fintool.com ανακατευθύνει στο Microsoft 365. Δεν υπάρχει πού να εγγραφείς.",[120,665,218],{"id":217},[29,667,668],{},"Το Ploutos AI είναι δικό μας, οπότε εδώ είναι το ίδιο σύνολο γεγονότων με την ίδια σειρά.",[29,670,671,672,674,675,679],{},"Τα νούμερα έρχονται από τα δομημένα δεδομένα XBRL της SEC, το στρώμα ",[66,673,98],{}," που περιγράφηκε παραπάνω, όχι από το μοντέλο. Στο μοντέλο δίνονται τα μεγέθη και σκέφτεται πάνω τους. Όταν τα δεδομένα δεν υπάρχουν, ",[48,676,678],{"href":677},"\u002Fel\u002Fblog\u002Fwhy-we-refuse-to-analyze-a-stock","σταματάει αντί να μαντέψει",". Οι συναλλαγές insiders από το Form 4, τα ουσιώδη γεγονότα από καταθέσεις 8-K και τα σήματα ποιότητας που πυροδοτούν συνδέονται με τη συγκεκριμένη κατάθεση στο EDGAR. Τα παράγωγα μεγέθη δεν συνδέονται πουθενά: μια εκτίμηση εύλογης αξίας, ένας δεκαετής μέσος όρος ή ένα κυλιόμενο άθροισμα χτίζονται από πολλές καταθέσεις και κανένα μεμονωμένο έγγραφο δεν τα δηλώνει, οπότε ένα link πάνω τους θα ήταν διακόσμηση. Αντί γι' αυτό δηλώνεται η μέθοδος.",[29,681,682,683,52],{},"Αφού γραφτεί η ανάλυση, ένα ξεχωριστό πέρασμα Δικηγόρου του Διαβόλου επιχειρηματολογεί γραπτά για την αρνητική πλευρά, και δημοσιεύεται δίπλα στη θετική. Η ",[48,684,686],{"href":685},"\u002Fel\u002Fblog\u002Fhow-ploutos-analyzes-a-stock","διαδικασία περιγράφεται πλήρως",[29,688,689],{},"Η κάλυψη είναι μόνο μετοχές εισηγμένες στις ΗΠΑ, NYSE, NASDAQ και AMEX. Χωρίς απομαγνητοφωνήσεις τηλεδιασκέψεων αποτελεσμάτων, χωρίς εκθέσεις χρηματιστηριακών, χωρίς μη αμερικανικές εισηγμένες. Έξοδος στα ελληνικά ή στα αγγλικά. Έρευνα, όχι επενδυτική συμβουλή.",[36,691,693],{"id":692},"δίπλα-δίπλα","Δίπλα δίπλα",[248,695,696,712],{},[251,697,698],{},[254,699,700,703,706,709],{},[257,701,702],{},"Εργαλείο",[257,704,705],{},"Από πού έρχονται τα νούμερα",[257,707,708],{},"Πού δείχνει η παραπομπή",[257,710,711],{},"Κάλυψη",[270,713,714,727,740,753,766,779,792,804,816,828,841],{},[254,715,716,718,721,724],{},[275,717,277],{},[275,719,720],{},"Το XBRL της κατάθεσης, όπως κατατέθηκε",[275,722,723],{},"Το ίδιο το έγγραφο",[275,725,726],{},"Όλοι όσοι καταθέτουν στην SEC",[254,728,729,731,734,737],{},[275,730,291],{},[275,732,733],{},"Μοντέλα Canalyst συν κείμενο",[275,735,736],{},"Η ακριβής πρόταση",[275,738,739],{},"68.000 εταιρείες, παγκόσμια",[254,741,742,744,747,750],{},[275,743,305],{},[275,745,746],{},"Πίνακες κατευθείαν από καταθέσεις",[275,748,749],{},"Απόσπασμα ή πίνακας που διαλέγεις",[275,751,752],{},"Όσοι καταθέτουν στην SEC",[254,754,755,757,760,763],{},[275,756,319],{},[275,758,759],{},"Δική τους εξαγωγή με ανθρώπινο έλεγχο. S&P Capital IQ εκτός ΗΠΑ\u002FΚαναδά\u002FΗΒ\u002FΕΕ",[275,761,762],{},"Η σελίδα της κατάθεσης, στο Max",[275,764,765],{},"100.000 εταιρείες, παγκόσμια",[254,767,768,770,773,776],{},[275,769,333],{},[275,771,772],{},"Σημασμένο XBRL. Επαληθευμένη εξαγωγή κειμένου με απόσπασμα",[275,774,775],{},"Έντυπο, ημερομηνία, περίοδος και απόσπασμα ανά νούμερο",[275,777,778],{},"Εισηγμένες ΗΠΑ",[254,780,781,783,786,789],{},[275,782,165],{},[275,784,785],{},"Υπολογισμός από κατατεθειμένες καταστάσεις, μέθοδος αδήλωτη",[275,787,788],{},"Η κατάθεση, ή αναζήτηση EDGAR",[275,790,791],{},"ΗΠΑ, αναφορά σε δολάρια",[254,793,794,796,799,802],{},[275,795,178],{},[275,797,798],{},"Η κατάθεση, τη διαβάζεις εσύ",[275,800,801],{},"Το έγγραφο στο sec.gov",[275,803,752],{},[254,805,806,808,810,813],{},[275,807,188],{},[275,809,374],{},[275,811,812],{},"Πουθενά. Επαληθεύεις μόνος σου",[275,814,815],{},"120.000 μετοχές, παγκόσμια",[254,817,818,820,822,825],{},[275,819,198],{},[275,821,387],{},[275,823,824],{},"Δεν ισχυρίζεται κάτι",[275,826,827],{},"100.000 μετοχές, παγκόσμια",[254,829,830,832,835,838],{},[275,831,208],{},[275,833,834],{},"Αναλυτής XBRL συν κείμενο, ιστορικά",[275,836,837],{},"Κατάθεση και σελίδα, ιστορικά",[275,839,840],{},"8.000 εταιρείες ΗΠΑ, ιστορικά",[254,842,843,845,849,852],{},[275,844,218],{},[275,846,413,847],{},[66,848,98],{},[275,850,851],{},"Συγκεκριμένη κατάθεση για Form 4, 8-K, σήματα ποιότητας. Καμία για παράγωγα μεγέθη",[275,853,778],{},[36,855,857],{"id":856},"πώς-να-διαβάσεις-τον-πίνακα","Πώς να διαβάσεις τον πίνακα",[29,859,860],{},"Αν η ερώτησή σου είναι «μπορώ να ελέγξω αυτό το νούμερο», η στήλη που μετράει είναι η τρίτη, και η ειλικρινής απάντηση για τα περισσότερα εργαλεία είναι «μπορείς να ελέγξεις το έγγραφο, όχι το νούμερο». Τα links σε επίπεδο πρότασης, που ισχυρίζεται το AlphaSense και προσεγγίζει το Equibles με αποσπάσματα, είναι η εξαίρεση.",[29,862,863],{},"Αν η ερώτησή σου είναι «πληκτρολόγησε αυτό το νούμερο ένα γλωσσικό μοντέλο», η στήλη που μετράει είναι η δεύτερη. Το Equibles λέει XBRL. Το Ploutos AI λέει XBRL. Το Fiscal.ai λέει εξαγωγή με ανθρώπινη επαλήθευση. Το Fintool έλεγε XBRL και δεν υπάρχει πια. Όλα τα υπόλοιπα είναι είτε αναγνώστες, είτε ροή παρόχου, είτε δεν λένε.",[29,865,866],{},"Αν η ερώτησή σου είναι «ποιο να χρησιμοποιήσω», αυτό το άρθρο δεν απαντάει, επίτηδες. Οι μηχανισμοί διαφέρουν επειδή διαφέρουν οι δουλειές. Ένας αναγνώστης όπως το Last10K ή το BamSEC είναι για ανθρώπους που θα κάνουν το διάβασμα. Ένα terminal με δεδομένα παρόχου όπως το TIKR ή το Simply Wall St είναι για ανθρώπους που θέλουν συγκρίσιμα νούμερα σε εκατό αγορές και δεν τους πειράζει να εμπιστευτούν ότι η S&P διάβασε την κατάθεση. Τα εργαλεία με πρώτη προτεραιότητα την παραπομπή, Equibles, Fiscal.ai, StockPortfolio.pro και Ploutos AI, είναι για ανθρώπους που θέλουν το νούμερο και τον τρόπο να το ελέγξουν στο ίδιο μέρος, και διαφέρουν κυρίως σε γεωγραφία και γλώσσα.",[29,868,869],{},"Ό,τι κι αν χρησιμοποιείς, κάν' του την ερώτηση πάνω στην οποία χτίστηκε αυτό το άρθρο: όταν μου δείχνεις ένα νούμερο, το διάβασες ή το αναζήτησες; Ένα εργαλείο που δεν μπορεί να απαντήσει σε αυτό ονομάζει την κατάθεση, δεν παραπέμπει σε αυτήν.",{"title":438,"searchDepth":439,"depth":439,"links":871},[872,873,874,885,886],{"id":506,"depth":442,"text":507},{"id":538,"depth":442,"text":539},{"id":574,"depth":442,"text":575,"children":875},[876,877,878,879,880,881,882,883,884],{"id":581,"depth":439,"text":582},{"id":597,"depth":439,"text":598},{"id":610,"depth":439,"text":611},{"id":164,"depth":439,"text":165},{"id":177,"depth":439,"text":178},{"id":187,"depth":439,"text":188},{"id":197,"depth":439,"text":198},{"id":207,"depth":439,"text":208},{"id":217,"depth":439,"text":218},{"id":692,"depth":442,"text":693},{"id":856,"depth":442,"text":857},"Οκτώ εργαλεία λένε ότι τα νούμερά τους βγαίνουν από καταθέσεις στην SEC. Εννοούν τρία διαφορετικά πράγματα. Από πού προέρχονται πραγματικά τα νούμερα του καθενός, πού δείχνει η παραπομπή, και τι καλύπτει, ελεγμένα στις σελίδες των ίδιων των εταιρειών τον Σεπτέμβριο του 2026.",[889,892,895,898],{"q":890,"a":891},"Τι σημαίνει όταν ένα εργαλείο ανάλυσης μετοχών λέει ότι παραπέμπει σε καταθέσεις SEC;","Ένα από τρία πράγματα. Ονομάζει την κατάθεση μέσα στο κείμενο, δίνει link στην κατάθεση ή σε ένα απόσπασμά της, ή διαβάζει το νούμερο από τα δομημένα δεδομένα XBRL της κατάθεσης ώστε το γλωσσικό μοντέλο να μην πληκτρολογεί ποτέ τον αριθμό. Τα περισσότερα κάνουν το πρώτο, αρκετά το δεύτερο, λίγα λένε καθαρά αν κάνουν το τρίτο.",{"q":893,"a":894},"Ποια εργαλεία συνδέουν κάθε νούμερο με την κατάθεση;","Το Fiscal.ai και το Equibles το δηλώνουν στις σελίδες τους, και το AlphaSense συνδέει κάθε πρόταση του AI με το απόσπασμα της πηγής. Το StockPortfolio.pro συνδέει σε επίπεδο κατάθεσης. Το Ploutos AI συνδέει συναλλαγές insiders, γεγονότα 8-K και σήματα ποιότητας με τη συγκεκριμένη κατάθεση, όχι όμως παράγωγα μεγέθη όπως η εύλογη αξία.",{"q":896,"a":897},"Υπάρχει δωρεάν τρόπος να πάρω τα νούμερα κατευθείαν από τις καταθέσεις SEC;","Ναι. Το API XBRL της ίδιας της SEC επιστρέφει κάθε νούμερο που έχει σημάνει μια εταιρεία στα 10-K και 10-Q της, δωρεάν, χωρίς κλειδί, ενημερωμένο μέσα σε ένα λεπτό από την κατάθεση. Πάνω σε αυτό είναι χτισμένα αρκετά από τα εργαλεία.",{"q":899,"a":900},"Δουλεύει κάποιο από αυτά τα εργαλεία στα ελληνικά;","Το Ploutos AI παράγει την ανάλυσή του στα ελληνικά και στα αγγλικά. Το Simply Wall St έχει ιαπωνικά, γερμανικά, ισπανικά και κορεατικά. Τα υπόλοιπα είναι μόνο στα αγγλικά.",{},"Εργαλεία ανάλυσης μετοχών που παραπέμπουν σε καταθέσεις SEC","\u002Fel\u002Fblog\u002Fstock-research-tools-that-cite-sec-filings",{"title":488,"description":887},"blog\u002Fel\u002Fstock-research-tools-that-cite-sec-filings",[482,483],"SSbQeBe5us0wmIL0hgzzgBvPS0q88pHpOc4pJz7bQ7Y",[909,1298,1529,1727,2146,2353,2559,2761,2904,3121,3421,3582,3799,3988,4196,4391],{"id":910,"title":911,"author":6,"body":912,"date":1269,"description":1270,"extension":459,"faq":1271,"featured":473,"mentionsTickers":473,"meta":1287,"metaTitle":1288,"navigation":476,"ogImage":921,"path":1289,"readMinutes":1290,"seo":1291,"stem":1292,"tags":1293,"translationKey":1296,"__hash__":1297},"blogEn\u002Fblog\u002Fen\u002Fetf-overlap.md","ETF overlap: how much of your second fund did you already own?",{"type":8,"value":913,"toc":1261},[914,922,925,928,932,935,1002,1005,1009,1012,1015,1018,1038,1041,1045,1048,1131,1134,1142,1146,1149,1181,1184,1188,1191,1213,1216,1220,1223,1226,1239],[11,915,917],{"className":916},[14,15,16,17,18,19],[21,918],{"alt":919,"className":920,"src":921},"A Venn diagram of an S&P 500 ETF and an MSCI World ETF sharing 455 of the same stocks",[25,26],"\u002Fblog\u002Fog-etf-overlap.svg",[29,923,924],{},"You buy a global ETF for broad exposure. Later you add an S&P 500 ETF, because the US has been doing the heavy lifting and you want more of it. Two funds, two different names, two different indices. It feels like diversification.",[29,926,927],{},"Then you look inside them, and most of the second fund is the first one again.",[36,929,931],{"id":930},"how-much-do-an-sp-500-fund-and-a-global-fund-actually-share","How much do an S&P 500 fund and a global fund actually share?",[29,933,934],{},"Here is the measurement, taken on 14 August 2026 from the issuer's own published holdings files.",[248,936,937,949],{},[251,938,939],{},[254,940,941,943,946],{},[257,942],{},[257,944,945],{},"iShares Core S&P 500",[257,947,948],{},"iShares Core MSCI World",[270,950,951,962,973,987],{},[254,952,953,956,959],{},[275,954,955],{},"Tickers",[275,957,958],{},"SXR8, CSPX",[275,960,961],{},"IWDA, SWDA",[254,963,964,967,970],{},[275,965,966],{},"Holdings with a listed ticker",[275,968,969],{},"510",[275,971,972],{},"1,265",[254,974,975,978,983],{},[275,976,977],{},"Stocks held by both",[275,979,980],{},[43,981,982],{},"455",[275,984,985],{},[43,986,982],{},[254,988,989,992,997],{},[275,990,991],{},"Weight of those shared stocks",[275,993,994],{},[43,995,996],{},"98.8%",[275,998,999],{},[43,1000,1001],{},"69.1%",[29,1003,1004],{},"The two funds hold 455 of the same companies. That is not the interesting part. The interesting part is the last row, where the same 455 stocks account for almost all of one fund and roughly two thirds of the other.",[36,1006,1008],{"id":1007},"why-the-overlap-is-asymmetric","Why the overlap is asymmetric",[29,1010,1011],{},"An overlap figure is usually quoted as one number, and that hides what is actually going on.",[29,1013,1014],{},"The S&P 500 fund is nearly a subset of the global fund. Almost everything it owns, 98.8% by weight, is also sitting in the global fund. The global fund is the broader container: it holds those same US large caps and then roughly 800 more companies from Japan, the UK, Canada, and the rest of the developed world.",[29,1016,1017],{},"So the direction matters:",[81,1019,1020,1031],{},[84,1021,1022,1023,1026,1027,1030],{},"If you already hold the ",[43,1024,1025],{},"global"," fund and add the ",[43,1028,1029],{},"S&P 500"," fund, you are adding almost no new companies. You are increasing the weight of names you already owned.",[84,1032,1022,1033,1026,1035,1037],{},[43,1034,1029],{},[43,1036,1025],{}," fund, you are genuinely adding something, because roughly 31% of the global fund by weight is companies the S&P fund does not hold at all.",[29,1039,1040],{},"Same pair of funds. Completely different answer depending on which one came first.",[36,1042,1044],{"id":1043},"what-this-looks-like-at-the-individual-stock-level","What this looks like at the individual stock level",[29,1046,1047],{},"The shared names are not a long tail of small positions. They are the largest positions in both funds, at different weights.",[248,1049,1050,1063],{},[251,1051,1052],{},[254,1053,1054,1057,1060],{},[257,1055,1056],{},"Stock",[257,1058,1059],{},"In the World fund",[257,1061,1062],{},"In the S&P 500 fund",[270,1064,1065,1076,1087,1098,1109,1120],{},[254,1066,1067,1070,1073],{},[275,1068,1069],{},"NVIDIA",[275,1071,1072],{},"5.57%",[275,1074,1075],{},"8.13%",[254,1077,1078,1081,1084],{},[275,1079,1080],{},"Apple",[275,1082,1083],{},"4.78%",[275,1085,1086],{},"6.65%",[254,1088,1089,1092,1095],{},[275,1090,1091],{},"Microsoft",[275,1093,1094],{},"3.74%",[275,1096,1097],{},"5.48%",[254,1099,1100,1103,1106],{},[275,1101,1102],{},"Amazon",[275,1104,1105],{},"2.79%",[275,1107,1108],{},"3.92%",[254,1110,1111,1114,1117],{},[275,1112,1113],{},"Alphabet (Class A)",[275,1115,1116],{},"2.16%",[275,1118,1119],{},"3.02%",[254,1121,1122,1125,1128],{},[275,1123,1124],{},"Broadcom",[275,1126,1127],{},"2.02%",[275,1129,1130],{},"2.95%",[29,1132,1133],{},"Every one of these appears in both funds. If you hold both, your actual position in each company is a blend of the two weights, and it is higher than either row suggests on its own. Nobody sets out to put a specific percentage of their savings into one chip designer. It arrives as a by-product of owning two funds that both track the same market from different angles.",[29,1135,1136,1137,1141],{},"This is the same concentration question the ",[48,1138,1140],{"href":1139},"\u002Fblog\u002Fhow-to-analyze-an-etf","ETF composition guide"," covers for a single fund, except it compounds when funds are stacked.",[36,1143,1145],{"id":1144},"how-to-check-the-overlap-between-your-own-funds","How to check the overlap between your own funds",[29,1147,1148],{},"You do not need a tool or a subscription for this. The large issuers publish full daily holdings, free, on the product page for each fund.",[1150,1151,1152,1158,1164,1169,1175],"ol",{},[84,1153,1154,1157],{},[43,1155,1156],{},"Find the fund's product page"," on the issuer's site. Search the ticker or the ISIN, not the marketing name, because near identical names exist across share classes.",[84,1159,1160,1163],{},[43,1161,1162],{},"Download the holdings file."," It lists every position with its ticker, ISIN, weight, sector and country.",[84,1165,1166],{},[43,1167,1168],{},"Do the same for the second fund.",[84,1170,1171,1174],{},[43,1172,1173],{},"Match on ISIN rather than ticker."," The same company can carry different ticker strings on different exchanges, and ISIN is unique. Ticker matching is quicker and usually close enough, but it will miss a few dual-listed names.",[84,1176,1177,1180],{},[43,1178,1179],{},"Add up the weight of the matches, in each fund separately."," Two numbers, not one. That asymmetry is the answer.",[29,1182,1183],{},"Two things worth knowing before you interpret what you get. Weights move with prices, so these figures drift week to week, and the ones in this article are a snapshot of one day rather than a constant. And the weights in a holdings file will not always sum to exactly 100%, because funds also hold cash and instruments that do not carry an equity ticker.",[36,1185,1187],{"id":1186},"the-same-fund-under-two-names","The same fund under two names",[29,1189,1190],{},"One more trap, and it is a cheaper mistake to avoid than the overlap itself.",[29,1192,1193,1196,1197,1200,1201,1196,1204,1207,1208,1196,1210,1212],{},[66,1194,1195],{},"SXR8"," and ",[66,1198,1199],{},"CSPX"," in the table above are not two funds. They are the same iShares Core S&P 500 UCITS ETF, listed on different exchanges and quoted in different currencies. ",[66,1202,1203],{},"IWDA",[66,1205,1206],{},"SWDA"," are likewise one fund. A portfolio holding both ",[66,1209,1195],{},[66,1211,1199],{}," has no diversification between them whatsoever, because there is nothing between them.",[29,1214,1215],{},"Before comparing two funds, check the ISIN of each. If the ISIN matches, the overlap is 100% and no further analysis is needed.",[36,1217,1219],{"id":1218},"what-the-overlap-number-does-and-does-not-tell-you","What the overlap number does and does not tell you",[29,1221,1222],{},"It tells you what you own. That is genuinely useful, and most people holding several funds have never added it up.",[29,1224,1225],{},"It does not tell you what to do about it. Whether a concentrated position is appropriate depends on why each fund is in the portfolio, and that is not something an article can determine from the outside. What the measurement does is replace an assumption with a figure, so any decision that follows is made against the real composition instead of against two fund names that sound different.",[29,1227,1228,1229,1233,1234,1238],{},"If you want the same treatment applied to individual companies, the ",[48,1230,1232],{"href":1231},"\u002Fblog\u002Fhow-to-read-stock-fundamentals","fundamentals guide"," covers what to check inside a single business, and you can ",[48,1235,1237],{"href":1236},"\u002Fanalyze","run an analysis"," on any of the names above to see what sits behind the weight.",[29,1240,1241,1242,1246,1247,1251,1252,1256,1257,52],{},"The same measurement, run every week across the largest US equity ETFs, is published as a table on the ",[48,1243,1245],{"href":1244},"\u002Fetf-overlap","ETF overlap page",": which pairs share the most weight, and which stocks sit in the most funds. The US-listed counterparts of the two funds above have their own pages built from their SEC filings, ",[48,1248,1250],{"href":1249},"\u002Fetf\u002Fivv\u002Fholdings","IVV"," for the S&P 500 and ",[48,1253,1255],{"href":1254},"\u002Fetf\u002Furth\u002Fholdings","URTH"," for MSCI World, and the pair itself is measured at ",[48,1258,1260],{"href":1259},"\u002Fetf-overlap\u002Fivv-vs-urth","IVV vs URTH",{"title":438,"searchDepth":439,"depth":439,"links":1262},[1263,1264,1265,1266,1267,1268],{"id":930,"depth":442,"text":931},{"id":1007,"depth":442,"text":1008},{"id":1043,"depth":442,"text":1044},{"id":1144,"depth":442,"text":1145},{"id":1186,"depth":442,"text":1187},{"id":1218,"depth":442,"text":1219},"2026-08-14","Two popular UCITS ETFs share 455 of the same stocks. The measured overlap between an S&P 500 and an MSCI World fund, and how to check your own.",[1272,1275,1278,1281,1284],{"q":1273,"a":1274},"What is ETF overlap?","The portion of two funds that is invested in the same companies. Two funds with different names and different indices can still hold most of the same stocks, so buying both adds less than the two names suggest.",{"q":1276,"a":1277},"How much do an S&P 500 ETF and an MSCI World ETF overlap?","Measured on 14 August 2026, the iShares Core S&P 500 (SXR8\u002FCSPX) and the iShares Core MSCI World (IWDA\u002FSWDA) held 455 of the same stocks. Those shared names were 98.8% of the S&P fund and 69.1% of the World fund.",{"q":1279,"a":1280},"Why is ETF overlap asymmetric?","Because one fund can sit almost entirely inside a broader one. An S&P 500 fund is nearly a subset of a global developed-market fund, so almost all of the first is contained in the second, while the second still holds many companies the first does not.",{"q":1282,"a":1283},"How can I check the overlap between my own ETFs?","Download each fund's daily holdings file from the issuer's product page, then compare them by ticker or ISIN and add up the weight of the names that appear in both. The large issuers publish this data free and update it daily.",{"q":1285,"a":1286},"Does overlap mean I should not own both funds?","That is not a question a page can answer for you, because it depends on what each fund is doing in your portfolio. What the numbers do tell you is how much genuinely new exposure the second fund adds, so you are deciding with the actual composition in front of you rather than the names.",{},"ETF overlap: are your funds the same?","\u002Fblog\u002Fetf-overlap",7,{"title":911,"description":1270},"blog\u002Fen\u002Fetf-overlap",[1294,1295],"etf","basics","etf-overlap","FJcmyRxJmDDjD8aQG3_nHhNosBJT51732VBckSoGRWw",{"id":1299,"title":1300,"author":6,"body":1301,"date":1504,"description":1505,"extension":459,"faq":1506,"featured":473,"mentionsTickers":473,"meta":1519,"metaTitle":1520,"navigation":476,"ogImage":1310,"path":1521,"readMinutes":1522,"seo":1523,"stem":1524,"tags":1525,"translationKey":1527,"__hash__":1528},"blogEn\u002Fblog\u002Fen\u002Ffundamental-analysis-guide.md","Fundamental analysis of stocks: a guide to valuing a company",{"type":8,"value":1302,"toc":1494},[1303,1311,1318,1321,1325,1335,1342,1346,1349,1369,1372,1376,1383,1395,1399,1402,1408,1412,1420,1432,1436,1451,1458,1462,1469,1483,1487],[11,1304,1306],{"className":1305},[14,15,16,17,18,19],[21,1307],{"alt":1308,"className":1309,"src":1310},"Fundamental analysis in three steps: business, financials, valuation",[25,26],"\u002Fblog\u002Fog-fundamentals-guide.svg",[29,1312,1313,1314,1317],{},"A stock's price changes every second. The value of the business behind it changes far more slowly. ",[43,1315,1316],{},"Fundamental analysis"," is the attempt to answer the second, more important question: what is this business actually worth, regardless of what the market is saying today?",[29,1319,1320],{},"This guide explains what fundamental analysis is, how it differs from technical analysis, and the three steps in which you can evaluate a stock. It is the starting point; each step links to a deeper article if you want to dig in.",[36,1322,1324],{"id":1323},"what-fundamental-analysis-is-and-isnt","What fundamental analysis is (and isn't)",[29,1326,1327,1328,1331,1332,1334],{},"There are two broad schools in stock analysis. ",[43,1329,1330],{},"Technical analysis"," studies price and volume on charts, looking for patterns that supposedly predict the next move. ",[43,1333,1316],{}," does almost the opposite: it largely ignores price and looks at the business itself. How much it earns, how much cash it generates, how much debt it carries, how fast it grows, and whether it has something protecting it from competition.",[29,1336,1337,1338,1341],{},"The goal is not to guess tomorrow's price. It is to estimate a ",[43,1339,1340],{},"fair value"," for the company and compare it with what the market is asking. When the price is well below the estimated value, there is a margin of safety. When it is above, you may be paying up for optimism. The whole discipline of value investing lives in that gap between price and value.",[36,1343,1345],{"id":1344},"the-skeleton-three-questions","The skeleton: three questions",[29,1347,1348],{},"Every fundamental analysis, however complex it gets, essentially answers three questions in this order:",[1150,1350,1351,1357,1363],{},[84,1352,1353,1356],{},[43,1354,1355],{},"Is it a good business?"," (what it does, how it makes money, whether it has a competitive advantage)",[84,1358,1359,1362],{},[43,1360,1361],{},"Is it financially healthy?"," (earnings, cash flows, debt, growth)",[84,1364,1365,1368],{},[43,1366,1367],{},"Is it reasonably priced?"," (valuation and margin of safety)",[29,1370,1371],{},"If it fails one of the three, you usually do not need to proceed to the rest. Let's take them one at a time.",[36,1373,1375],{"id":1374},"step-1-understand-the-business","Step 1: Understand the business",[29,1377,1378,1379,1382],{},"Before you touch a number, you should be able to explain in one sentence ",[43,1380,1381],{},"what the company does and how it makes money",". If you cannot, you are not ready to evaluate it, no matter how much analysis you read.",[29,1384,1385,1386,1389,1390,1394],{},"Then you ask the most important qualitative question: ",[43,1387,1388],{},"what protects it?"," An excellent company with no defensive character attracts competitors who erode its profits. A durable competitive advantage, what is called an ",[48,1391,1393],{"href":1392},"\u002Fblog\u002Fwhat-is-an-economic-moat","economic moat",", is the reason a company can keep high returns for years instead of losing them.",[36,1396,1398],{"id":1397},"step-2-read-the-financials","Step 2: Read the financials",[29,1400,1401],{},"This is where the numbers come in, and the good news is you do not need to read all of them. Three financial statements tell almost the whole story: the income statement (revenue and earnings), the balance sheet (what it owns and owes), and the cash flow statement (the real cash coming in and out).",[29,1403,1404,1405,52],{},"From these come a few quality metrics that actually matter: profit margins, return on capital, debt relative to earnings, and revenue growth over time. The full, step-by-step way to read them without being an accountant is in ",[48,1406,1407],{"href":1231},"how to read a stock's fundamentals",[36,1409,1411],{"id":1410},"step-3-valuation-price-versus-value","Step 3: Valuation — price versus value",[29,1413,1414,1415,1419],{},"A good company is not automatically a good investment; it depends on the price you pay. Valuation is the attempt to put a number on fair value. The best-known method is ",[48,1416,1418],{"href":1417},"\u002Fblog\u002Fwhat-is-dcf-valuation","discounted cash flow (DCF)",", which estimates value from the company's future cash, but it is fragile and so is combined with others.",[29,1421,1422,1423,1426,1427,1431],{},"The key is not a precise number, but the ",[43,1424,1425],{},"margin of safety",": the distance between price and estimated value that protects you when some of your assumptions turn out wrong. But be careful: a cheap price does not automatically mean an opportunity. The difference between a genuinely ",[48,1428,1430],{"href":1429},"\u002Fblog\u002Fwhat-is-an-undervalued-stock","undervalued stock and a value trap"," is exactly whether the fundamentals justify the low price or warn about it.",[36,1433,1435],{"id":1434},"beyond-the-numbers-catalysts-smart-money-and-risks","Beyond the numbers: catalysts, smart money and risks",[29,1437,1438,1439,1442,1443,1446,1447,52],{},"Fundamental analysis does not end at the accounting. Two more lenses add context. The first is ",[43,1440,1441],{},"catalysts",": what could change the story (earnings, management changes, industry developments). The second is ",[43,1444,1445],{},"smart money",": what institutional investors and insiders hold, a signal that, read correctly, works as a prompt for research and not as a command. The way to use it without getting trapped is explained in ",[48,1448,1450],{"href":1449},"\u002Fblog\u002Fsmart-money-13f","smart money & 13F",[29,1452,1453,1454,1457],{},"And the most underrated lens: the ",[43,1455,1456],{},"bear case",". Before you conclude, deliberately ask what could go wrong. An analysis that has not considered how it would lose money is not an analysis, it is a wish list.",[36,1459,1461],{"id":1460},"how-much-time-and-how-not-to-give-up","How much time, and how not to give up",[29,1463,1464,1465,1468],{},"The reason most individual investors do not do fundamental analysis is not that it is conceptually hard, but that it is ",[43,1466,1467],{},"time-consuming",": gathering and interpreting data eats the hours, not the decision itself. That is exactly where most attempts stop.",[29,1470,1471,1472,1475,1476,1479,1480,1482],{},"That is the problem Ploutos solves: it compresses data gathering, a sector-aware valuation ensemble, smart-money signals and the Devil's Advocate into minutes, from verifiable official sources, and leaves the ",[43,1473,1474],{},"decision to you",". See ",[48,1477,1478],{"href":238},"how it analyzes a stock step by step",", or ",[48,1481,1237],{"href":1236}," on a company you care about.",[36,1484,1486],{"id":1485},"where-to-start","Where to start",[29,1488,1489,1490,1493],{},"Don't try to analyze everything. Pick ",[43,1491,1492],{},"one"," company you understand, give it an hour, and walk through the three questions: is it a good business, is it healthy, is it reasonably priced? Your first analysis will be slow and incomplete. Your tenth will be routine. That is how judgment is built, and judgment is the one thing no tool can replace.",{"title":438,"searchDepth":439,"depth":439,"links":1495},[1496,1497,1498,1499,1500,1501,1502,1503],{"id":1323,"depth":442,"text":1324},{"id":1344,"depth":442,"text":1345},{"id":1374,"depth":442,"text":1375},{"id":1397,"depth":442,"text":1398},{"id":1410,"depth":442,"text":1411},{"id":1434,"depth":442,"text":1435},{"id":1460,"depth":442,"text":1461},{"id":1485,"depth":442,"text":1486},"2026-06-19","What fundamental analysis is, how it differs from technical analysis, and the 3 steps to evaluate a stock: the business, the financials, and the valuation.",[1507,1510,1513,1516],{"q":1508,"a":1509},"What is fundamental analysis?","It is the method of evaluating a stock through the underlying business: earnings, cash flows, debt, growth and competitive advantage. The goal is to estimate the company's fair value and compare it with the market price.",{"q":1511,"a":1512},"What is the difference between fundamental and technical analysis?","Technical analysis studies price and charts to predict short-term movement. Fundamental analysis largely ignores price and asks what the business is actually worth. They are tools for different questions and time horizons.",{"q":1514,"a":1515},"How long does fundamental analysis of a stock take?","A careful analysis can take anywhere from an hour to several days, depending on depth. Most of the time goes into gathering and interpreting data, not the decision itself.",{"q":1517,"a":1518},"Do I need accounting knowledge?","Nothing specialised. You need to understand a few basic concepts (revenue, earnings, cash flow, debt, a few ratios). The guide below and the linked articles explain them in plain language.",{},"Fundamental analysis of stocks: a guide","\u002Fblog\u002Ffundamental-analysis-guide",10,{"title":1300,"description":1505},"blog\u002Fen\u002Ffundamental-analysis-guide",[1526,1295],"investing","fundamental-analysis-guide","SDK3dkS5KL9pBIrtgJ6BYWrYrjNBxIMonItFuxGmSIg",{"id":1530,"title":1531,"author":6,"body":1532,"date":1707,"description":1708,"extension":459,"faq":1709,"featured":473,"mentionsTickers":473,"meta":1719,"metaTitle":1720,"navigation":476,"ogImage":1541,"path":1721,"readMinutes":1290,"seo":1722,"stem":1723,"tags":1724,"translationKey":1725,"__hash__":1726},"blogEn\u002Fblog\u002Fen\u002Fhow-long-to-analyze-a-stock.md","How long does it take to analyze a stock (and where the time actually goes)",{"type":8,"value":1533,"toc":1699},[1534,1542,1545,1552,1556,1559,1566,1573,1577,1584,1590,1596,1600,1603,1635,1638,1642,1648,1654,1657,1661,1672,1679,1689,1693,1696],[11,1535,1537],{"className":1536},[14,15,16,17,18,19],[21,1538],{"alt":1539,"className":1540,"src":1541},"A time spectrum from minutes to weeks for analyzing a stock",[25,26],"\u002Fblog\u002Fog-research-time.svg",[29,1543,1544],{},"Ask ten investors how long it takes them to analyze a stock and you will get ten different answers. The most honest one usually sounds like this: \"anywhere from a few minutes to several weeks, depending.\" And the whole point hides in that \"depending\".",[29,1546,1547,1548,1551],{},"The question is not how much time, but ",[43,1549,1550],{},"where"," the time goes. Because once you understand where it goes, you can cut the part that is merely tedious and keep the part that actually matters.",[36,1553,1555],{"id":1554},"why-the-time-varies-so-much","Why the time varies so much",[29,1557,1558],{},"Two things decide whether an analysis takes you ten minutes or ten days.",[29,1560,1561,1562,1565],{},"The first is ",[43,1563,1564],{},"how well you already know the industry",". A software company that looks like five others you have worked through is fast. A biotech, an insurer, or a shipping company, if you have never touched the sector, takes time just to grasp what it actually sells and how it makes money.",[29,1567,1568,1569,1572],{},"The second is ",[43,1570,1571],{},"what decision you are trying to make",". A quick \"is this even worth a closer look?\" is minutes. A \"I am putting in a meaningful amount and holding for years\" justifies weeks. Not every stock needs the same depth, and treating them all the same is the surest way to waste time.",[36,1574,1576],{"id":1575},"where-the-time-actually-goes","Where the time actually goes",[29,1578,1579,1580,1583],{},"Here is the surprise: the time does ",[43,1581,1582],{},"not"," go into the numbers. Revenue, earnings, debt, margins, you find all of that in minutes. The real time goes into two things that no quick lookup solves.",[29,1585,1561,1586,1589],{},[43,1587,1588],{},"understanding exactly what the company does and where it makes money",". It is easy to buy something you do not understand, then sell it on the first dip because you have no conviction about what you hold. Plenty of people have sold good stocks far too early for exactly this reason, not because they got the entry wrong, but because they had no clear picture of the business to hold through the noise.",[29,1591,1568,1592,1595],{},[43,1593,1594],{},"checking whether the numbers are real",". Some companies massage things a little everywhere, in how they capitalise R&D, in depreciation, in adjusted earnings that conveniently make the ugly parts disappear. Spotting that takes time and experience, and it is exactly the part most people skip.",[36,1597,1599],{"id":1598},"the-5-point-quick-scan","The 5-point quick scan",[29,1601,1602],{},"Before you dig deep, a fast filter saves hours. The goal is not to decide, but to quickly figure out whether it is worth continuing. Look at five things:",[81,1604,1605,1611,1617,1623,1629],{},[84,1606,1607,1610],{},[43,1608,1609],{},"Profitability."," Does it earn consistent operating profit, or does it live on promises and narrative?",[84,1612,1613,1616],{},[43,1614,1615],{},"Debt."," Can it survive a bad year, or is it so leveraged that one wrong turn sends it into the wall?",[84,1618,1619,1622],{},[43,1620,1621],{},"Cash flow."," Do the earnings turn into actual cash, or do they sit on paper as receivables and accounting entries?",[84,1624,1625,1628],{},[43,1626,1627],{},"Valuation."," Are you paying a reasonable price for what you get, or pricing in a perfect future that has to play out exactly?",[84,1630,1631,1634],{},[43,1632,1633],{},"Quality of the numbers."," Are there signs something is being inflated? This is where ratios like Piotroski (fundamental quality), Altman Z (bankruptcy risk), and Beneish (possible manipulation) help.",[29,1636,1637],{},"If any of these flags a problem, you stop right there and save the weeks. That is the point of the quick scan, to let you say \"no\" fast, so you can spend your time where it is worth it.",[36,1639,1641],{"id":1640},"the-genuinely-hard-part","The genuinely hard part",[29,1643,1644,1645],{},"If it passes the filter, the hard part begins, and it is almost always the same: ",[43,1646,1647],{},"understanding the business well enough to know what you hold.",[29,1649,1650,1651,1653],{},"It is not the math. The math is time-consuming but not hard. The hard part is answering questions that have no ready-made number: why do customers pick this company over the next one? How easily could someone copy it? What has to happen for today's price to make sense? That \"how hard is it to copy\" is essentially the company's ",[48,1652,1393],{"href":1392},", and it is one of the most decisive pieces of the picture.",[29,1655,1656],{},"When you have clear answers to these, a 20% drop is an opportunity or a signal, not a reason to panic. When you do not, every move in the price scares you, because you cannot tell whether what you own changed or the market just had a bad day.",[36,1658,1660],{"id":1659},"how-to-save-time-without-losing-depth","How to save time without losing depth",[29,1662,1663,1664,1667,1668,1671],{},"The key is not to read less. It is to automate the ",[43,1665,1666],{},"mechanical"," part, gathering and checking the data, and keep your time for the part that needs ",[43,1669,1670],{},"judgment",", understanding the business and the risk scenarios.",[29,1673,1674,1675,1678],{},"That is where tools that read the official filings for you and surface a few understandable figures with the sources help, instead of leaving you to dig through hundreds of pages. With one condition: the numbers have to be real. A general-purpose language model often ",[48,1676,1677],{"href":50},"invents figures with total confidence",", so a tool is only worth it if its data is pulled from a verifiable source, not from the model's \"memory\".",[29,1680,1681,1682,1685,1686,1688],{},"That is exactly the logic behind Ploutos AI: the mechanical part happens automatically and with citations, and the judgment part stays with you, with the \"why\" alongside the \"what could go wrong\". If you want to see how the stages fit together, there is a ",[48,1683,1684],{"href":238},"detailed walkthrough of the pipeline",", or you can just ",[48,1687,1237],{"href":1236}," on a stock you know well and compare.",[36,1690,1692],{"id":1691},"what-this-means-for-you","What this means for you",[29,1694,1695],{},"The right question is not \"how much time does it take\". It is \"where is my time worth spending\". A few minutes to quickly cut anything that does not hold up, and the bulk of your time where the real decision hides: understanding the business and confirming the numbers are real.",[29,1697,1698],{},"If you have those two clear, analysis stops being a chore and becomes conviction. And conviction is what keeps you in a good position when the market tests you.",{"title":438,"searchDepth":439,"depth":439,"links":1700},[1701,1702,1703,1704,1705,1706],{"id":1554,"depth":442,"text":1555},{"id":1575,"depth":442,"text":1576},{"id":1598,"depth":442,"text":1599},{"id":1640,"depth":442,"text":1641},{"id":1659,"depth":442,"text":1660},{"id":1691,"depth":442,"text":1692},"2026-06-09","From a few minutes to several weeks. See where the time actually goes when you analyze a stock, a 5-point quick scan, and how to save time without losing depth.",[1710,1713,1716],{"q":1711,"a":1712},"How long does it take to analyze a stock?","Anywhere from a few minutes to several weeks, depending on how well you already know the industry and what decision you are trying to make.",{"q":1714,"a":1715},"Where does the time actually go?","Not into the numbers. It goes into understanding exactly what the company does and where it makes money, and confirming the numbers are real.",{"q":1717,"a":1718},"How do I save time without losing depth?","Automate the mechanical part (gathering and checking data) and keep your time for the part that needs judgment.",{},"How long does it take to analyze a stock","\u002Fblog\u002Fhow-long-to-analyze-a-stock",{"title":1531,"description":1708},"blog\u002Fen\u002Fhow-long-to-analyze-a-stock",[1526,1295],"how-long-to-analyze-a-stock","c4zrw7h3eAQT93qQAVEZsgggochCnzfgCUjDKZKX2e8",{"id":1728,"title":1729,"author":6,"body":1730,"date":2123,"description":2124,"extension":459,"faq":2125,"featured":476,"mentionsTickers":473,"meta":2138,"metaTitle":2139,"navigation":476,"ogImage":1739,"path":238,"readMinutes":1290,"seo":2140,"stem":2141,"tags":2142,"translationKey":2144,"__hash__":2145},"blogEn\u002Fblog\u002Fen\u002Fhow-ploutos-analyzes-a-stock.md","How Ploutos AI analyzes a stock in 4 stages (and one Devil's Advocate)",{"type":8,"value":1731,"toc":2113},[1732,1740,1743,1746,1750,1753,1767,1770,1774,1777,1791,1885,1888,1892,1895,1906,1909,1913,1920,1964,1967,1971,1974,2007,2010,2014,2017,2024,2056,2059,2063,2066,2086,2090,2093,2110],[11,1733,1735],{"className":1734},[14,15,16,17,18,19],[21,1736],{"alt":1737,"className":1738,"src":1739},"Diagram of the Ploutos AI 5-stage analysis pipeline",[25,26],"\u002Fblog\u002Fog-pipeline.svg",[29,1741,1742],{},"Most tools that promise \"AI stock analysis\" do roughly the same thing. They paste your ticker into a large language model and hand you back whatever comes out. It looks smart. It reads like a research report. And often, when you look closely, the numbers are made up, the cited sources don't exist, and the conclusion contradicts a sentence written three paragraphs earlier.",[29,1744,1745],{},"Ploutos AI is built around the opposite assumption: a language model alone is not a research analyst. It needs structure, real data, and (most importantly) something to push back against its own conclusion. That is why every analysis you run goes through the same five stages, in the same order, with the same discipline. This article walks through what is happening behind the spinner.",[36,1747,1749],{"id":1748},"why-a-pipeline-instead-of-one-big-prompt","Why a pipeline instead of one big prompt",[29,1751,1752],{},"A single LLM call has two failure modes that are hard to fix from inside the prompt:",[1150,1754,1755,1761],{},[84,1756,1757,1760],{},[43,1758,1759],{},"It hallucinates specifics",": a P\u002FE ratio that does not match reality, an insider transaction that never happened, a sector rotation that ended last quarter.",[84,1762,1763,1766],{},[43,1764,1765],{},"It confirms its own thesis",": once it has written \"the case for buy\" in paragraph one, paragraph four will rarely call it wrong.",[29,1768,1769],{},"Splitting the work into stages, with each stage given a narrow job and each grounded in actual data fetched live from the public-domain stack we ship today (SEC EDGAR XBRL for fundamentals, FRED for macro and FX, FINRA for short interest, Marketaux for news), addresses the first problem. A separate, hostile critique pass at the end addresses the second. Below are the five stages in order.",[36,1771,1773],{"id":1772},"stage-1-screen-macro-regime-fundamentals","Stage 1: Screen (macro regime + fundamentals)",[29,1775,1776],{},"The first thing the agent does is fire two calls in parallel:",[81,1778,1779,1785],{},[84,1780,1781,1784],{},[43,1782,1783],{},"Macro regime check",": current VIX level, yield curve, market trend, central bank stance. This shapes how every later decision is weighted (a stock that scores well in a bull market scores differently in a stagflation regime).",[84,1786,1787,1790],{},[43,1788,1789],{},"Fundamentals pull",": for each ticker you submitted, the agent retrieves the data needed to score it against 10 sector-aware quality criteria:",[248,1792,1793,1803],{},[251,1794,1795],{},[254,1796,1797,1800],{},[257,1798,1799],{},"Criterion",[257,1801,1802],{},"What it checks",[270,1804,1805,1813,1821,1829,1837,1845,1853,1861,1869,1877],{},[254,1806,1807,1810],{},[275,1808,1809],{},"P\u002FE vs sector",[275,1811,1812],{},"Is the valuation reasonable relative to peers?",[254,1814,1815,1818],{},[275,1816,1817],{},"Revenue growth (5y CAGR)",[275,1819,1820],{},"Is the top line actually growing?",[254,1822,1823,1826],{},[275,1824,1825],{},"Gross margin vs sector",[275,1827,1828],{},"Does the business have pricing power?",[254,1830,1831,1834],{},[275,1832,1833],{},"EPS positive and growing",[275,1835,1836],{},"Profitability and trajectory",[254,1838,1839,1842],{},[275,1840,1841],{},"Assets\u002FLiabilities > 1",[275,1843,1844],{},"Solvent balance sheet",[254,1846,1847,1850],{},[275,1848,1849],{},"ROIC vs sector",[275,1851,1852],{},"Capital efficiency",[254,1854,1855,1858],{},[275,1856,1857],{},"Free cash flow trend",[275,1859,1860],{},"Is real cash being generated?",[254,1862,1863,1866],{},[275,1864,1865],{},"Debt coverage over 10y",[275,1867,1868],{},"Can it service its debt across cycles?",[254,1870,1871,1874],{},[275,1872,1873],{},"Buybacks (share count)",[275,1875,1876],{},"Returning capital or diluting?",[254,1878,1879,1882],{},[275,1880,1881],{},"FCF\u002FDividend ratio",[275,1883,1884],{},"Is the dividend covered?",[29,1886,1887],{},"Each criterion passed or failed contributes to a score from 0 to 10. This score is not a verdict; it is a filter. It tells the next stage which tickers are worth deep research and which are not.",[36,1889,1891],{"id":1890},"stage-2-context-sector-news-macro-overlay","Stage 2: Context (sector news + macro overlay)",[29,1893,1894],{},"A high fundamental score does not mean much if the sector is being repriced this week. Before any deep work, the agent uses a real-time web search to surface recent sector-wide events:",[81,1896,1897,1900,1903],{},[84,1898,1899],{},"Regulatory changes affecting the industry",[84,1901,1902],{},"Interest rate impacts on cyclicals vs defensives",[84,1904,1905],{},"Industry-specific macro trends (chip cycle, energy supply, retail spending)",[29,1907,1908],{},"If something material has shifted, the agent re-ranks the candidates from Stage 1. A perfectly-fundamented company in a sector that just lost its biggest customer is not the same opportunity it was last month.",[36,1910,1912],{"id":1911},"stage-3-deep-research-parallel-data-pull-per-ticker","Stage 3: Deep research (parallel data pull per ticker)",[29,1914,1915,1916,1919],{},"This is where the heavy lifting happens. For each ticker that made it through, the agent fires seven independent data calls ",[43,1917,1918],{},"in parallel",":",[1150,1921,1922,1928,1934,1940,1946,1952,1958],{},[84,1923,1924,1927],{},[43,1925,1926],{},"Valuation models",": a sector-aware ensemble, DCF (or a growth-DCF variant for pre-profit fast-growers), sector multiples, EPV (Earnings Power Value), the Graham Number, and specialists for utilities, digital-asset-heavy balance sheets, and pre-revenue biotech, with confidence ranges.",[84,1929,1930,1933],{},[43,1931,1932],{},"News sentiment",": bullish vs bearish tilt relative to the sector, plus buzz level.",[84,1935,1936,1939],{},[43,1937,1938],{},"Earnings surprises",": beats, misses, and revisions over the last four quarters (execution quality).",[84,1941,1942,1945],{},[43,1943,1944],{},"Insider transactions",": what executives are actually doing with their personal money over the last 6 months.",[84,1947,1948,1951],{},[43,1949,1950],{},"Recent SEC filings",": material events from 8-K filings (M&A, restatements, cyber incidents, executive changes).",[84,1953,1954,1957],{},[43,1955,1956],{},"Institutional holdings",": what tracked smart-money investors are accumulating or selling.",[84,1959,1960,1963],{},[43,1961,1962],{},"Short interest",": bear thesis check; high short interest with weak fundamentals is a red flag.",[29,1965,1966],{},"These come back, and the model synthesises a thesis covering: which fundamentals are working and which are not, where the price sits versus fair value, whether insiders and institutions agree with the thesis, and what material events might be coming.",[36,1968,1970],{"id":1969},"stage-4-synthesis-structured-output","Stage 4: Synthesis (structured output)",[29,1972,1973],{},"The synthesis becomes a structured output you would recognise as a PickCard in the app:",[81,1975,1976,1983,1989,1995,2001],{},[84,1977,1978,1979,1982],{},"A ",[43,1980,1981],{},"model rating",": one of six tiers from Strong Outperform to Strong Underperform. This is the relative language a sell-side analyst would use to describe a stock against its peer companies. It is the model's analytical view at a point in time, not a personal recommendation to buy or sell, and it does not take your circumstances into account.",[84,1984,1985,1988],{},[43,1986,1987],{},"Margin of Safety",": the percentage gap between the current market price and the midpoint of the model's fair-value range. A discipline metric showing how much room the price has before reaching the model's estimated fair value, not a price target.",[84,1990,1991,1994],{},[43,1992,1993],{},"Conditions to watch",": observable factors the model highlights, for example macroeconomic conditions that would strengthen the thesis or price ranges where its valuation models would give the position more headroom. These are reference points the user can monitor, not instructions to transact.",[84,1996,1997,2000],{},[43,1998,1999],{},"Key risk and catalyst",": what could invalidate the thesis, what could accelerate it.",[84,2002,2003,2006],{},[43,2004,2005],{},"Insider, institutional, and short-interest data points",": one-line summaries distilled from the underlying filings.",[29,2008,2009],{},"Every user-submitted ticker gets one entry in this output, even those that score poorly. The agent is not allowed to silently drop a ticker. If the model rating is \"Underperform\", that is what the model says.",[36,2011,2013],{"id":2012},"stage-5-devils-advocate-the-part-most-tools-skip","Stage 5: Devil's Advocate (the part most tools skip)",[29,2015,2016],{},"This is the stage that, in my opinion, makes the difference between research and recommendation.",[29,2018,2019,2020,2023],{},"After the verdict is formed, a separate model pass (a more capable one) gets the picks and one instruction: ",[43,2021,2022],{},"find what we got wrong",". For each pick it returns:",[81,2025,2026,2032,2038,2044,2050],{},[84,2027,2028,2031],{},[43,2029,2030],{},"Robustness score"," (1 to 10): how resilient is this thesis to challenge?",[84,2033,2034,2037],{},[43,2035,2036],{},"Weakest assumption",": the single assumption most likely to be wrong.",[84,2039,2040,2043],{},[43,2041,2042],{},"Overlooked risk",": a real risk the main analyst did NOT flag.",[84,2045,2046,2049],{},[43,2047,2048],{},"Bear case summary",": how does this thesis go to zero or underperform by 30%+?",[84,2051,2052,2055],{},[43,2053,2054],{},"Invalidates thesis if",": a concrete observable signal that would force exiting the position.",[29,2057,2058],{},"For positive verdicts this matters most; it is the structured antidote to AI confirmation bias. For negative verdicts the critique is collapsed by default in the UI, because piling more bad news onto an already-negative thesis is just noise.",[36,2060,2062],{"id":2061},"what-you-actually-see-at-the-end","What you actually see at the end",[29,2064,2065],{},"The whole pipeline takes about 30 to 50 seconds for a single ticker in a warm path. What lands on your screen is:",[81,2067,2068,2074,2080],{},[84,2069,1978,2070,2073],{},[43,2071,2072],{},"PickCard"," with the model rating, margin-of-safety metric, conditions to watch, all the data points, and (for positive ratings) an expanded Devil's Advocate critique.",[84,2075,1978,2076,2079],{},[43,2077,2078],{},"Top Picks tab"," summarising the ratings across all submitted tickers.",[84,2081,1978,2082,2085],{},[43,2083,2084],{},"Chat"," where you can ask the agent follow-up questions and it has the full pipeline output as context. \"What if rates stay above 5%?\" \"How does this compare to similar companies?\" \"What changed in their last earnings call?\"",[36,2087,2089],{"id":2088},"why-this-matters-for-value-investing","Why this matters for value investing",[29,2091,2092],{},"Value investing is not about predicting next quarter. It is about discipline: knowing what you own, knowing what could go wrong, refusing to confuse a story for a thesis. Every stage in the pipeline is built around that discipline:",[81,2094,2095,2098,2101,2104,2107],{},[84,2096,2097],{},"Stage 1 says: do not waste time on companies that fail the basic quality screen.",[84,2099,2100],{},"Stage 2 says: the present matters, do not analyse last year's company.",[84,2102,2103],{},"Stage 3 says: ground every claim in actual data.",[84,2105,2106],{},"Stage 4 says: be honest about what you see, even if the conclusion is uncomfortable.",[84,2108,2109],{},"Stage 5 says: assume you might be wrong, and look for the evidence that would prove it.",[29,2111,2112],{},"You still make the final decision. Ploutos AI is a research tool, not an advisor. It compresses what would take hours of manual data-gathering and cross-referencing into about a minute. Whether the model's view matches your own conviction is up to you.",{"title":438,"searchDepth":439,"depth":439,"links":2114},[2115,2116,2117,2118,2119,2120,2121,2122],{"id":1748,"depth":442,"text":1749},{"id":1772,"depth":442,"text":1773},{"id":1890,"depth":442,"text":1891},{"id":1911,"depth":442,"text":1912},{"id":1969,"depth":442,"text":1970},{"id":2012,"depth":442,"text":2013},{"id":2061,"depth":442,"text":2062},{"id":2088,"depth":442,"text":2089},"2026-05-24","From screening to bear-case stress test, the full pipeline that runs every time you submit a ticker, explained without jargon.",[2126,2129,2132,2135],{"q":2127,"a":2128},"How does Ploutos AI analyze a stock?","In stages: screening, gathering data from filings, synthesising a structured thesis, and a separate Devil's Advocate pass that looks for what could go wrong.",{"q":2130,"a":2131},"Where does the data come from?","From official sources (SEC EDGAR filings, market data), with a source on every claim, not from a language model's 'memory'.",{"q":2133,"a":2134},"How long does an analysis take?","Usually a few minutes. The mechanical part (gathering and checking data) is automated.",{"q":2136,"a":2137},"Is it investment advice?","No. It's research and education, not personalised advice. You are solely responsible for your decisions.",{},"How Ploutos AI analyzes a stock",{"title":1729,"description":2124},"blog\u002Fen\u002Fhow-ploutos-analyzes-a-stock",[482,2143,483],"how-to","how-ploutos-analyzes-a-stock","rWJQ-MzS4J7R5TVv7Y-KsA3DEFO6rlAp2aSmAZ10ovg",{"id":2147,"title":2148,"author":6,"body":2149,"date":2327,"description":2328,"extension":459,"faq":2329,"featured":473,"mentionsTickers":473,"meta":2345,"metaTitle":2346,"navigation":476,"ogImage":2158,"path":1139,"readMinutes":2347,"seo":2348,"stem":2349,"tags":2350,"translationKey":2351,"__hash__":2352},"blogEn\u002Fblog\u002Fen\u002Fhow-to-analyze-an-etf.md","How to analyze an ETF: what's actually inside a fund",{"type":8,"value":2150,"toc":2319},[2151,2159,2162,2167,2185,2189,2192,2195,2199,2205,2211,2217,2223,2238,2242,2254,2257,2277,2281,2284,2288,2291,2294,2298,2309],[11,2152,2154],{"className":2153},[14,15,16,17,18,19],[21,2155],{"alt":2156,"className":2157,"src":2158},"An ETF's composition broken down by position",[25,26],"\u002Fblog\u002Fog-etf-holdings.svg",[29,2160,2161],{},"An ETF (Exchange-Traded Fund) isn't one investment, it's a basket of dozens to thousands of stocks or bonds, packaged under a single ticker you buy like a stock. Its name (\"Technology,\" \"Growth,\" \"Dividend\") gives you the general idea, but not what you actually own: how concentrated it is in a handful of positions, which sectors really dominate, or how high-quality the underlying companies are. This article shows you how to check that yourself, before you put money in.",[29,2163,2164],{},[43,2165,2166],{},"In short:",[81,2168,2169,2172,2179,2182],{},[84,2170,2171],{},"An ETF's name is marketing, not analysis: two funds with a similar title can have completely different compositions.",[84,2173,2174,2175,2178],{},"The source of truth is the ",[43,2176,2177],{},"N-PORT filing"," every fund files monthly with the SEC, not the provider's fact sheet.",[84,2180,2181],{},"Worth checking before the name or the past performance chart: concentration, sector tilt, expense ratio, and the average quality of the holdings.",[84,2183,2184],{},"ETFs and individual stocks aren't rivals, they solve different problems.",[36,2186,2188],{"id":2187},"why-a-funds-name-isnt-enough","Why a fund's name isn't enough",[29,2190,2191],{},"Two ETFs with nearly the same name can behave completely differently. A cap-weighted \"technology\" ETF might have a third of its portfolio in two or three mega-cap names, while another, equal-weighted, spreads the risk evenly across hundreds of names. A \"dividend\" ETF might emphasize high yield today, while another emphasizes decades of steady dividend growth, two very different risk profiles under a similar label.",[29,2193,2194],{},"That's not deception, it's just that an ETF's name is designed for marketing, not for due diligence. The only way to know what you're actually holding is to look at its composition directly.",[36,2196,2198],{"id":2197},"what-to-check-before-you-buy-an-etf","What to check before you buy an ETF",[29,2200,2201,2204],{},[43,2202,2203],{},"Holdings composition."," The full list of positions with their exact weights, not just the top-10 a fact sheet usually shows. A fund with 200 positions can have 60% of its value in the first 15.",[29,2206,2207,2210],{},[43,2208,2209],{},"Concentration."," How much of the fund sits in its few largest positions. High concentration isn't automatically bad, but it means the ETF's return depends heavily on a handful of names, not the whole \"basket\" its name implies.",[29,2212,2213,2216],{},[43,2214,2215],{},"Sector and style tilt."," Which sectors actually dominate, and whether the fund leans toward value, growth, or quality companies. A \"broad market\" ETF can today carry heavy exposure to one sector simply because that sector has grown disproportionately.",[29,2218,2219,2222],{},[43,2220,2221],{},"Expense ratio."," The annual management cost, quietly deducted from the return every year. Over a decade or more, the gap between 0.05% and 0.75% annually compounds into a meaningful amount.",[29,2224,2225,2228,2229,2233,2234,2237],{},[43,2226,2227],{},"Quality of the underlying companies."," Beyond ",[2230,2231,2232],"em",{},"what"," the ETF holds, there's ",[2230,2235,2236],{},"how good"," those companies are fundamentally: profitability, debt, cash flow. An ETF is, in effect, the weighted average quality of its members, even though it's rarely presented that way.",[36,2239,2241],{"id":2240},"where-to-find-the-real-numbers-the-n-port-filing","Where to find the real numbers: the N-PORT filing",[29,2243,2244,2245,2247,2248,2253],{},"Every US-registered fund, ETFs included, is required to file a monthly ",[43,2246,2177],{}," with the SEC: the full, detailed composition of its portfolio, public and free via ",[48,2249,277],{"href":2250,"rel":2251},"https:\u002F\u002Fwww.sec.gov\u002Fedgar\u002Fsearch\u002F",[2252],"nofollow",". It's the official source, one step ahead of the provider's fact sheet or marketing page, which often shows rounded figures, stale data, or only the top 10 positions.",[29,2255,2256],{},"The gap matters most for lesser-known or more thematic ETFs, where the actual composition can diverge noticeably from what you'd assume just reading the name and the short description.",[29,2258,2259,2260,2264,2265,1196,2269,2273,2274,2276],{},"If you would rather not open the filing yourself, the same N-PORT data is laid out per fund on our holdings pages, for example ",[48,2261,2263],{"href":2262},"\u002Fetf\u002Fspy\u002Fholdings","SPY",", ",[48,2266,2268],{"href":2267},"\u002Fetf\u002Fqqq\u002Fholdings","QQQ",[48,2270,2272],{"href":2271},"\u002Fetf\u002Fvti\u002Fholdings","VTI",": the reported positions, the sector split, what changed since the previous filing, and the funds most like it. The ",[48,2275,1245],{"href":1244}," puts two to five of them side by side.",[36,2278,2280],{"id":2279},"a-simple-example","A simple example",[29,2282,2283],{},"Take a hypothetical \"Semiconductors\" ETF. On paper, it looks like diversified exposure to an entire industry. In practice, because the semiconductor industry is dominated by a handful of very large companies, its N-PORT filing might show the top three positions covering 35-40% of the fund. You wouldn't see that from the name, only by looking at the actual composition. That's not necessarily a problem, it's just something worth knowing before you decide how it fits alongside the rest of your portfolio.",[36,2285,2287],{"id":2286},"etfs-vs-individual-stocks-which-do-you-pick","ETFs vs individual stocks: which do you pick?",[29,2289,2290],{},"They're not really rivals. For many investors, a broad ETF is a perfectly sensible choice: diversification with minimal effort, low cost, no need to track individual companies. Picking individual stocks is for people who want to do that deliberately, whether for part of their portfolio or out of genuine interest, and want to do it with discipline rather than tips or guesswork.",[29,2292,2293],{},"Even if you invest mainly through ETFs, it's worth knowing what you hold: an ETF is just a way of owning many companies at once, not something fundamentally different from them.",[36,2295,2297],{"id":2296},"how-ploutos-ai-does-it","How Ploutos AI does it",[29,2299,2300,2301,2304,2305,2308],{},"Instead of showing you a rounded top-10 or leaving you to guess, Ploutos AI pulls an ETF's real composition directly from its most recent N-PORT filing with the SEC, then runs every holding through the same fundamental analysis it would apply if you'd typed it in individually: profitability, debt, cash flow, valuation. The result is a ",[43,2302,2303],{},"weighted score"," (how high-quality the average holding is, weighted by position size), a ",[43,2306,2307],{},"style profile"," (value, quality, growth), sector concentration, and expense ratio, all in one report.",[29,2310,2311,2312,2314,2315,52],{},"If you want to see what that looks like on a real fund, ",[48,2313,1237],{"href":1236}," on any ETF, or check which ETFs stand out right now under the same methodology on ",[48,2316,2318],{"href":2317},"\u002Fmarket-radar","Market Radar",{"title":438,"searchDepth":439,"depth":439,"links":2320},[2321,2322,2323,2324,2325,2326],{"id":2187,"depth":442,"text":2188},{"id":2197,"depth":442,"text":2198},{"id":2240,"depth":442,"text":2241},{"id":2279,"depth":442,"text":2280},{"id":2286,"depth":442,"text":2287},{"id":2296,"depth":442,"text":2297},"2026-07-09","A fund's name doesn't tell you what you actually own. How to check its composition, concentration, and holding quality before you buy.",[2330,2333,2336,2339,2342],{"q":2331,"a":2332},"How can I see what an ETF actually holds?","From the fund's most recent N-PORT filing with the SEC (EDGAR), which is public and free. It lists the full portfolio with exact weights, not the rounded top-10 you get from a fact sheet.",{"q":2334,"a":2335},"What is an N-PORT filing?","A monthly report every US-registered fund, ETFs included, is required to file with the SEC, disclosing the fund's full portfolio composition. It's the official source, one step ahead of the provider's marketing material.",{"q":2337,"a":2338},"What does \"concentration\" mean for an ETF?","How much of the fund sits in its few largest positions. Two ETFs with a similar theme can have very different concentration, one may effectively be a bet on 3-4 names, while another is genuinely spread out.",{"q":2340,"a":2341},"ETFs or individual stocks, which is better?","They're not competing. A broad ETF is a sensible choice for low-effort diversification. Picking individual stocks is for people who want to do that deliberately, with discipline, not tips.",{"q":2343,"a":2344},"How often does an ETF's composition change?","Depends on its strategy. A passive index fund changes rarely (mostly at index rebalancing). An actively managed or thematic ETF can turn over its holdings much more often.",{},"How to analyze an ETF",9,{"title":2148,"description":2328},"blog\u002Fen\u002Fhow-to-analyze-an-etf",[1294,1295],"how-to-analyze-an-etf","ID8V27FVUZASwFJ-A4RHiWnH5ehevwuCSqZCmJEBc0M",{"id":2354,"title":2355,"author":6,"body":2356,"date":2533,"description":2534,"extension":459,"faq":2535,"featured":473,"mentionsTickers":473,"meta":2551,"metaTitle":2552,"navigation":476,"ogImage":2365,"path":1231,"readMinutes":2553,"seo":2554,"stem":2555,"tags":2556,"translationKey":2557,"__hash__":2558},"blogEn\u002Fblog\u002Fen\u002Fhow-to-read-stock-fundamentals.md","How to read a stock's fundamentals (without being an accountant)",{"type":8,"value":2357,"toc":2525},[2358,2366,2373,2377,2380,2400,2403,2405,2408,2439,2442,2446,2449,2481,2484,2488,2495,2502,2506,2513,2520,2522],[11,2359,2361],{"className":2360},[14,15,16,17,18,19],[21,2362],{"alt":2363,"className":2364,"src":2365},"The three financial statements: income statement, balance sheet, cash flow",[25,26],"\u002Fblog\u002Fog-fundamentals.svg",[29,2367,2368,2369,2372],{},"A company's fundamentals scare most beginners. Filings are huge, full of jargon and numbers. The good news: ",[43,2370,2371],{},"you don't have to read all of it."," You need a few things, in the right order.",[36,2374,2376],{"id":2375},"the-3-statements-one-sentence-each","The 3 statements, one sentence each",[29,2378,2379],{},"Every company publishes three core financial statements. You don't need accounting, you need to know what each one asks:",[81,2381,2382,2388,2394],{},[84,2383,2384,2387],{},[43,2385,2386],{},"Income statement:"," does the company make a profit? It shows revenue, costs, and what's left at the end.",[84,2389,2390,2393],{},[43,2391,2392],{},"Balance sheet:"," what does it own and owe? It shows assets, debt, and equity at a point in time.",[84,2395,2396,2399],{},[43,2397,2398],{},"Cash flow statement:"," do the earnings turn into real cash? It shows the money that actually comes in and goes out.",[29,2401,2402],{},"The third is the most underrated. A company can show profit on paper but generate no cash, and that's where a lot of problems hide.",[36,2404,1599],{"id":1598},[29,2406,2407],{},"Before you dig deep, this filter saves you hours. Look at five things:",[1150,2409,2410,2415,2420,2425,2434],{},[84,2411,2412,2414],{},[43,2413,1609],{}," Does it earn consistent operating profit, or live on narrative? Look at the 3-5 year trend, not one year.",[84,2416,2417,2419],{},[43,2418,1615],{}," Can it survive a bad year? Compare debt to operating profit (e.g. debt-to-EBITDA). Too much debt = fragile.",[84,2421,2422,2424],{},[43,2423,1621],{}," Is free cash flow positive and close to reported earnings? If earnings are far bigger than cash, find out why.",[84,2426,2427,2429,2430,2433],{},[43,2428,1627],{}," Are you paying reasonably or pricing in a perfect future? This is where ",[48,2431,2432],{"href":1429},"fair value and margin of safety"," help.",[84,2435,2436,2438],{},[43,2437,1633],{}," Are there signs something is being inflated? (More below.)",[29,2440,2441],{},"If any of these flags a problem, you stop there. You don't need to read 200 pages on a company with collapsing margins.",[36,2443,2445],{"id":2444},"three-quality-scores-that-summarise-a-lot","Three quality scores that summarise a lot",[29,2447,2448],{},"Instead of weighing dozens of figures, three classic scores give a quick read:",[81,2450,2451,2461,2471],{},[84,2452,2453,2456,2457,2460],{},[43,2454,2455],{},"Piotroski F-Score (0-9):"," uses 9 simple checks to show whether fundamentals are ",[43,2458,2459],{},"improving or deteriorating",". An 8-9 is a strong quality signal.",[84,2462,2463,2466,2467,2470],{},[43,2464,2465],{},"Altman Z-Score:"," estimates ",[43,2468,2469],{},"bankruptcy risk",". It tells you whether the balance sheet is in a safe or dangerous zone.",[84,2472,2473,2476,2477,2480],{},[43,2474,2475],{},"Beneish M-Score:"," surfaces the likelihood of ",[43,2478,2479],{},"earnings 'cooking'",". Not proof of fraud, but a signal to look more closely.",[29,2482,2483],{},"They're not magic, but a clearly bad score on these saves a lot of wasted time.",[36,2485,2487],{"id":2486},"the-most-insidious-red-flag","The most insidious red flag",[29,2489,2490,2491,2494],{},"The most dangerous sign in a company isn't one big hole that shouts. It's when ",[43,2492,2493],{},"no red flag shows",", because the 'cooking' has been spread a little everywhere: in how they capitalise R&D, in depreciation, in stock compensation, in adjusted or non-GAAP earnings that conveniently make the ugly parts disappear.",[29,2496,2497,2498,2501],{},"A practical rule: always go back to the ",[43,2499,2500],{},"GAAP"," numbers (the official ones), not the adjusted ones the company serves you. Non-GAAP gives a 'clean' story, and then you have to roll it back to GAAP to see what was removed or added.",[36,2503,2505],{"id":2504},"where-tools-take-it-from-here","Where tools take it from here",[29,2507,2508,2509,2512],{},"The mechanical part, gathering and checking all of this, is time-consuming but automatable. With one condition: the numbers have to be real. A general-purpose language model often ",[48,2510,2511],{"href":50},"invents figures with confidence",", so a tool is only worth it if its data comes from the official filings, not from its \"memory\".",[29,2514,2515,2516,2519],{},"If you want to see these checks run automatically in one structured analysis, with the sources in front, you can ",[48,2517,2518],{"href":1236},"run one"," on a stock you know well.",[36,2521,1692],{"id":1691},[29,2523,2524],{},"You don't need to become an accountant. You need to know what each statement asks, run the 5-point quick scan, and not blindly trust adjusted numbers. With that, in a few minutes you can separate what's worth digging into from what to leave.",{"title":438,"searchDepth":439,"depth":439,"links":2526},[2527,2528,2529,2530,2531,2532],{"id":2375,"depth":442,"text":2376},{"id":1598,"depth":442,"text":1599},{"id":2444,"depth":442,"text":2445},{"id":2486,"depth":442,"text":2487},{"id":2504,"depth":442,"text":2505},{"id":1691,"depth":442,"text":1692},"2026-06-10","Filings are huge, but you don't have to read all of it. The 3 statements, a 5-point quick scan, and 3 quality scores in plain English.",[2536,2539,2542,2545,2548],{"q":2537,"a":2538},"What are a stock's fundamentals?","They are the financials of the business behind the stock: revenue, earnings, debt, cash flow, margins. Fundamental analysis looks at the health and value of the company, as opposed to technical analysis, which looks at the price and the chart.",{"q":2540,"a":2541},"What are the 3 core financial statements?","The income statement (does the company make a profit?), the balance sheet (what does it own and owe?), and the cash flow statement (do the earnings turn into real cash?). Together they give the full picture.",{"q":2543,"a":2544},"Do I have to read the whole 10-K?","No. A 10-K is hundreds of pages, but for a first read a few figures are enough: profitability, debt, cash flow, valuation, and the quality of the numbers. If any of those flags a problem, you stop there.",{"q":2546,"a":2547},"What is the Piotroski score?","A 0-9 score that uses 9 simple checks to show whether a company's fundamentals are improving or deteriorating (profitability, leverage, efficiency). An 8-9 is a quality signal, a 2-3 a warning.",{"q":2549,"a":2550},"What is the most insidious red flag?","When it looks like there is none. Spread-thin 'cooking', a little in R&D capitalisation, a little in depreciation, a little in adjusted\u002Fnon-GAAP earnings, is hard to filter because no single figure screams.",{},"How to read stock fundamentals (for beginners)",8,{"title":2355,"description":2534},"blog\u002Fen\u002Fhow-to-read-stock-fundamentals",[1526,1295],"how-to-read-stock-fundamentals","x5C-U-5zQLjtOu3SGVK5Mp1QSmOgMxNn35-KMwx54fM",{"id":2560,"title":2561,"author":6,"body":2562,"date":2737,"description":2738,"extension":459,"faq":2739,"featured":473,"mentionsTickers":473,"meta":2752,"metaTitle":2753,"navigation":476,"ogImage":2571,"path":2754,"readMinutes":2553,"seo":2755,"stem":2756,"tags":2757,"translationKey":2759,"__hash__":2760},"blogEn\u002Fblog\u002Fen\u002Fhow-to-tell-if-a-stock-is-overvalued.md","How to tell if a stock is overvalued: the signs that actually matter",{"type":8,"value":2563,"toc":2729},[2564,2572,2578,2581,2585,2606,2613,2617,2620,2626,2633,2637,2644,2655,2659,2662,2672,2676,2679,2705,2709,2726],[11,2565,2567],{"className":2566},[14,15,16,17,18,19],[21,2568],{"alt":2569,"className":2570,"src":2571},"Market price sitting above a company's intrinsic value",[25,26],"\u002Fblog\u002Fog-overvalued.svg",[29,2573,2574,2575,52],{},"It is tempting to judge a stock by its price tag. A share at 800 feels expensive; a share at 4 feels cheap. But the number on the screen says almost nothing about whether a stock is overvalued. A company at 800 can be a bargain and one at 4 can be wildly expensive. Overvaluation is not about the price — it is about the price ",[2230,2576,2577],{},"relative to what the business is worth",[29,2579,2580],{},"This article explains what overvalued actually means, the signs that genuinely matter, and the trap that catches even careful investors: confusing a great company with a great price.",[36,2582,2584],{"id":2583},"overvalued-means-price-above-value-not-price-above-zero","Overvalued means price above value, not price above zero",[29,2586,2587,2588,2591,2592,2595,2596,2599,2600,2603,2604,52],{},"Every business is, in the end, a machine for producing cash over time. The ",[2230,2589,2590],{},"intrinsic value"," of a stock is what that future cash is reasonably worth today (see our piece on ",[48,2593,2594],{"href":1417},"DCF valuation"," for how that translation works). A stock is ",[43,2597,2598],{},"overvalued"," when its market price sits meaningfully above that intrinsic value, and ",[43,2601,2602],{},"undervalued"," when it sits below it, leaving a ",[48,2605,1425],{"href":1429},[29,2607,2608,2609,2612],{},"So the price tag is the wrong starting point. The right question is always: ",[2230,2610,2611],{},"what is the business worth, and what am I being asked to pay for it?"," Everything below is a way of approaching that one question.",[36,2614,2616],{"id":2615},"sign-1-the-valuation-multiple-is-high-relative-to-the-growth-that-justifies-it","Sign 1: the valuation multiple is high relative to the growth that justifies it",[29,2618,2619],{},"The fastest sanity check is a valuation multiple — most commonly price-to-earnings (P\u002FE), or price-to-sales (P\u002FS) for companies that don't yet earn much. A multiple compresses \"price relative to what the company produces\" into a single number.",[29,2621,2622,2623,2625],{},"But a high multiple is ",[43,2624,1582],{}," the same as overvalued. A high P\u002FE simply means the market expects strong future growth. The real test is whether the business can plausibly deliver the growth the multiple already assumes. A company growing earnings 40% a year may fully deserve a P\u002FE that would be absurd for one growing at 3%. Overvaluation appears when the multiple demands a future the business is unlikely to produce.",[29,2627,2628,2629,2632],{},"A useful habit: instead of asking \"is this multiple high?\", ask \"",[43,2630,2631],{},"what would have to be true"," for this multiple to make sense?\" If the answer requires near-flawless execution for a decade, the expectations — not the company — are the risk.",[36,2634,2636],{"id":2635},"sign-2-the-price-already-has-years-of-good-news-baked-in","Sign 2: the price already has years of good news baked in",[29,2638,2639,2640,2643],{},"This is the heart of overvaluation. When a stock has run up on a strong story, the optimism becomes the price. From there, even ",[2230,2641,2642],{},"good"," results can disappoint, because the bar that was set is so high.",[29,2645,2646,2647,2650,2651,2654],{},"The mental move is to separate the ",[43,2648,2649],{},"business"," from the ",[43,2652,2653],{},"expectations priced into it",". Ask: how much growth, for how many years, at what margins, is this price assuming? When those assumptions start to sound heroic — a small company growing into a number larger than entire mature industries — you are likely looking at expectations that leave no room for error.",[36,2656,2658],{"id":2657},"sign-3-a-great-company-at-any-price","Sign 3: a great company at any price",[29,2660,2661],{},"The most expensive mistake in investing is believing that a wonderful business is automatically a wonderful investment. It is not. Price decides the return.",[29,2663,2664,2665,2667,2668,2671],{},"A dominant company with a wide ",[48,2666,1393],{"href":1392}," can keep growing, keep winning, keep doing everything right — and still be a poor ",[2230,2669,2670],{},"stock"," if you overpaid at the start. The company succeeds; the expectations baked into your entry price do not. \"Overvalued\" is simply the gap between an excellent business and a price that already assumes perfection.",[36,2673,2675],{"id":2674},"how-to-check-it-sensibly-without-pretending-to-be-precise","How to check it sensibly (without pretending to be precise)",[29,2677,2678],{},"Valuation is a range, not a verdict. A disciplined process does not produce a single magic \"fair price\" — it produces a reasonable range and a sense of how much optimism the current price requires. A few practical steps:",[81,2680,2681,2687,2693,2699],{},[84,2682,2683,2686],{},[43,2684,2685],{},"Compare price to value, never price to itself."," Estimate what the business is worth, then look at the gap. The share price in isolation is noise.",[84,2688,2689,2692],{},[43,2690,2691],{},"Stress-test the assumptions."," Change the growth rate and margins a little and watch what happens to the value. If the case only works in the best scenario, the price is fragile.",[84,2694,2695,2698],{},[43,2696,2697],{},"Look for the argument against."," Before trusting any bullish case, find the strongest reason it could be wrong. A view that has survived its own counter-argument is worth far more than one that hasn't.",[84,2700,2701,2704],{},[43,2702,2703],{},"Treat any \"target price\" with suspicion."," Targets depend on assumptions that may not play out; a small change moves them a lot.",[36,2706,2708],{"id":2707},"how-ploutos-approaches-this","How Ploutos approaches this",[29,2710,2711,2712,2715,2716,2719,2720,2722,2723,52],{},"Ploutos is built around exactly this discipline. Every analysis pairs the bullish reasoning with a deliberate ",[43,2713,2714],{},"Devil's Advocate"," — the strongest case ",[2230,2717,2718],{},"against"," the idea — so the optimism in a price is tested rather than taken for granted. It works from real filings, frames value as a range with its assumptions on the table, and is built to describe what the data shows, not to tell you what to buy. If you want to pressure-test the expectations inside a price yourself, you can ",[48,2721,1237],{"href":1236}," on a stock you know well, or read how it fits into the ",[48,2724,2725],{"href":238},"full pipeline",[29,2727,2728],{},"Overvaluation is never about the size of the price tag. It is about how much has to go right to justify it — and whether you are being paid to take that risk, or paying for the privilege.",{"title":438,"searchDepth":439,"depth":439,"links":2730},[2731,2732,2733,2734,2735,2736],{"id":2583,"depth":442,"text":2584},{"id":2615,"depth":442,"text":2616},{"id":2635,"depth":442,"text":2636},{"id":2657,"depth":442,"text":2658},{"id":2674,"depth":442,"text":2675},{"id":2707,"depth":442,"text":2708},"2026-06-23","A high price doesn't make a stock overvalued. Learn the real signs: valuation multiples, the growth already priced in, and when a great company is still a bad price.",[2740,2743,2746,2749],{"q":2741,"a":2742},"How can you tell if a stock is overvalued?","A stock is overvalued when its market price sits above the intrinsic value of the underlying business — that is, above what the company's future cash flows are reasonably worth today. A high share price on its own tells you nothing; the question is price relative to value, usually examined through valuation multiples and the growth those multiples already assume.",{"q":2744,"a":2745},"Does a high P\u002FE ratio mean a stock is overvalued?","Not by itself. A high price-to-earnings ratio means the market expects strong future growth. It is overvalued only if the business cannot plausibly deliver the growth the multiple implies. A fast-growing company can deserve a high P\u002FE; a slow-growing one with the same P\u002FE may not.",{"q":2747,"a":2748},"Can a great company still be an overvalued stock?","Yes. A great business and a great investment are not the same thing. If the price already reflects years of flawless execution, even excellent results can disappoint the expectations baked into the price. The company can keep winning while the stock still falls.",{"q":2750,"a":2751},"What is the opposite of an overvalued stock?","An undervalued stock, where the price sits below intrinsic value, leaving a margin of safety. The same discipline — comparing price to value rather than judging the price tag itself — applies in both directions.",{},"How to tell if a stock is overvalued","\u002Fblog\u002Fhow-to-tell-if-a-stock-is-overvalued",{"title":2561,"description":2738},"blog\u002Fen\u002Fhow-to-tell-if-a-stock-is-overvalued",[2758,1295],"valuation","how-to-tell-if-a-stock-is-overvalued","YsnA3sCdzngyYRr0SrbHc_aBCINJyu2HdXZ3ZQcmPyw",{"id":2762,"title":2763,"author":6,"body":2764,"date":2880,"description":2881,"extension":459,"faq":2882,"featured":473,"mentionsTickers":473,"meta":2895,"metaTitle":2896,"navigation":476,"ogImage":2773,"path":2897,"readMinutes":2898,"seo":2899,"stem":2900,"tags":2901,"translationKey":2902,"__hash__":2903},"blogEn\u002Fblog\u002Fen\u002Fhow-we-track-our-track-record.md","How we track our own track record (and why most pickers hide theirs)",{"type":8,"value":2765,"toc":2872},[2766,2774,2777,2780,2784,2791,2798,2801,2805,2808,2815,2819,2825,2828,2832,2835,2841,2845,2848,2851,2855],[11,2767,2769],{"className":2768},[14,15,16,17,18,19],[21,2770],{"alt":2771,"className":2772,"src":2773},"Tracked picks plotted against a market benchmark over time",[25,26],"\u002Fblog\u002Fog-track-record.svg",[29,2775,2776],{},"Watch finance content for a week and you will see a hundred confident calls and almost no scorecards. Every screenshot is a winner. Every \"I told you so\" arrives after the fact. The losing calls quietly evaporate, and the ones that worked get pinned to the top of the feed.",[29,2778,2779],{},"This is not usually fraud. It is something more ordinary and more corrosive: a thousand small choices about what to show and what to forget, all leaning in the same flattering direction. A research tool that wants to be trusted has to do the opposite, and it has to do it by design, not by good intentions.",[36,2781,2783],{"id":2782},"survivorship-bias-why-most-track-records-are-fiction","Survivorship bias: why most track records are fiction",[29,2785,2786,2787,2790],{},"The technical name for the trick is ",[43,2788,2789],{},"survivorship bias",": you judge a strategy only by the examples that survived, because the failures are no longer in front of you. A pundit who makes fifty predictions and reminds you of the five that landed looks like a genius. The other forty-five are not lies, they are just gone.",[29,2792,2793,2794,2797],{},"It has a close cousin, ",[43,2795,2796],{},"hindsight bias",": the temptation to describe a call as cleaner and more confident than it actually was at the time. \"I always said that sector would rebound\" is easy to write once it has rebounded.",[29,2799,2800],{},"Both biases share one root cause: there is no fixed, timestamped record of what was actually claimed, and when. Remove that record and any track record becomes a story you tell about the past. Keep it, rigorously, and the story has to match the receipts.",[36,2802,2804],{"id":2803},"we-timestamp-every-pick-at-decision-time","We timestamp every pick at decision time",[29,2806,2807],{},"The foundation is simple and unglamorous: the moment an analysis produces a verdict, that verdict is recorded, with its date, its rating, and the price at that instant. It is frozen. Nothing about it can be quietly edited later to look smarter.",[29,2809,2810,2811,2814],{},"That timestamp is what makes the whole thing honest. Performance is always measured from the price on the day the call was made, not from some flattering entry point chosen afterwards. There is no reaching back to start the clock at a convenient low. The clock starts when the analysis ran, and you can see the analysis that produced it, the same ",[48,2812,2813],{"href":238},"structured output and Devil's Advocate critique"," you saw on the day.",[36,2816,2818],{"id":2817},"benchmark-against-the-market-not-against-zero","Benchmark against the market, not against zero",[29,2820,2821,2822],{},"\"Up 12%\" sounds great until you learn the whole market was up 15% over the same window. A number on its own is not a result, the only honest scorecard is ",[2230,2823,2824],{},"relative to the alternative you actually had.",[29,2826,2827],{},"So every tracked pick is measured against a broad market benchmark over the same period. The question is never just \"did this go up,\" it is \"did this do better or worse than simply owning the index instead.\" That is a far harder bar to clear, and it is the only one that tells you whether the research added anything at all. Beating zero is luck in a bull market. Beating the market is the thing that has to be earned.",[36,2829,2831],{"id":2830},"we-show-the-losers-too","We show the losers too",[29,2833,2834],{},"A scorecard that only contains winners is not a scorecard. The track record surfaces the full distribution, the best calls and the worst ones side by side, not a curated highlight reel. If a thesis underperformed, it stays on the record underperforming.",[29,2836,2837,2838,52],{},"This is uncomfortable on purpose. A tool that hides its misses is optimising for how it looks, and the moment it does that it stops being useful to you. Seeing where the process was wrong is exactly how you calibrate how much weight to give it, and it is information a marketing page will never volunteer. The same discipline runs through the product itself: every submitted ticker gets an honest verdict even when that verdict is negative, and the analysis ",[48,2839,2840],{"href":230},"stops rather than guess when the data is too thin",[36,2842,2844],{"id":2843},"why-past-performance-still-is-not-a-promise","Why past performance still is not a promise",[29,2846,2847],{},"Here is the part that honesty requires us to say plainly, even though it undercuts the marketing: a good track record is not a guarantee of future results, and you should distrust anyone who implies otherwise.",[29,2849,2850],{},"Markets change regimes. A process that suited the last few years can struggle in the next. Sample sizes are smaller than they look, and luck and skill are genuinely hard to separate over short windows. A transparent track record is valuable not because it predicts the future, but because it tells you the truth about the past, which biases, sample size, benchmark, and all. That is the most any honest scorecard can offer, and it is a great deal more than most offer at all.",[36,2852,2854],{"id":2853},"see-it-for-yourself","See it for yourself",[29,2856,2857,2858,2862,2863,2867,2868,2871],{},"The point of tracking openly is that you do not have to take any of this on faith. You can look at the ",[48,2859,2861],{"href":2860},"\u002Ftrack-record","public track record"," and your own ",[48,2864,2866],{"href":2865},"\u002Fperformance","performance history",", timestamps, benchmark comparison, winners and losers included. Then read how the underlying analysis is produced in our ",[48,2869,2870],{"href":238},"walkthrough of the pipeline",", and decide for yourself how much weight the process has earned.",{"title":438,"searchDepth":439,"depth":439,"links":2873},[2874,2875,2876,2877,2878,2879],{"id":2782,"depth":442,"text":2783},{"id":2803,"depth":442,"text":2804},{"id":2817,"depth":442,"text":2818},{"id":2830,"depth":442,"text":2831},{"id":2843,"depth":442,"text":2844},{"id":2853,"depth":442,"text":2854},"2026-05-25","Most stock pickers quietly bury their misses. Here is how Ploutos AI records every pick at decision time and benchmarks it against the market, openly.",[2883,2886,2889,2892],{"q":2884,"a":2885},"How is the track record measured?","Each pick is tracked from its entry price, marked to market, and compared against the index (SPY) over the same period.",{"q":2887,"a":2888},"What is alpha?","How a pick did relative to the market (S&P 500). Positive alpha means it beat the index.",{"q":2890,"a":2891},"Why make it public?","Because the right way to earn trust isn't 'trust me', it's showing what actually happened, the good moves and the bad.",{"q":2893,"a":2894},"Do a few weeks of track record mean anything?","No. A forward track record needs months to mean something. A few weeks is an indication, not proof.",{},"How we track our own track record","\u002Fblog\u002Fhow-we-track-our-track-record",6,{"title":2763,"description":2881},"blog\u002Fen\u002Fhow-we-track-our-track-record",[482,483],"how-we-track-our-track-record","d-GvAvb3ckqTzvedE57eZwnEzw6uFY1EblPwGUSDp_s",{"id":2905,"title":2906,"author":6,"body":2907,"date":1504,"description":3100,"extension":459,"faq":3101,"featured":473,"mentionsTickers":473,"meta":3114,"metaTitle":3115,"navigation":476,"ogImage":2916,"path":1449,"readMinutes":2553,"seo":3116,"stem":3117,"tags":3118,"translationKey":3119,"__hash__":3120},"blogEn\u002Fblog\u002Fen\u002Fsmart-money-13f.md","Smart money: what 13F filings show, and how to read them properly",{"type":8,"value":2908,"toc":3092},[2909,2917,2924,2927,2931,2938,2941,2945,2948,2958,2968,2974,2981,2985,2996,2999,3010,3014,3017,3041,3044,3048,3055,3065,3074,3078,3081],[11,2910,2912],{"className":2911},[14,15,16,17,18,19],[21,2913],{"alt":2914,"className":2915,"src":2916},"Institutional (13F) and insider signals as a research starting point, not a command",[25,26],"\u002Fblog\u002Fog-smart-money.svg",[29,2918,2919,2920,2923],{},"\"Smart money.\" The phrase promises that if you can see what the big institutional investors are buying, you can simply follow them. It is an appealing idea, and largely wrong. The data is real and public, but what it means, and especially what it ",[2230,2921,2922],{},"doesn't"," mean, is more subtle than the title of a \"See what Buffett bought\" video suggests.",[29,2925,2926],{},"This article explains what 13F filings and insider signals really are, what you can and cannot conclude from them, and how to use them as a research tool, never as a command.",[36,2928,2930],{"id":2929},"what-13f-filings-are-and-why-they-exist","What 13F filings are (and why they exist)",[29,2932,2933,2934,2937],{},"In the US, every institutional manager with more than $100 million in US equities is required by the SEC to publish, each quarter, what they hold. That filing is the ",[43,2935,2936],{},"Form 13F",". It is the reason you can, entirely free and legally, see the portfolios of Warren Buffett's Berkshire Hathaway, Pershing Square, Baupost and dozens of others.",[29,2939,2940],{},"The purpose is transparency: the market should know where large institutional capital is accumulating. For the individual investor, it is a rare window into the moves of people with research teams and access you do not have. The problem is not the data. It is how most people read it.",[36,2942,2944],{"id":2943},"what-they-tell-you-and-what-they-dont","What they tell you, and what they DON'T",[29,2946,2947],{},"13F filings have three serious limits that \"see what X bought\" videos rarely mention:",[29,2949,2950,2953,2954,2957],{},[43,2951,2952],{},"1. They are delayed up to ~45 days."," A manager reports what they held at the ",[2230,2955,2956],{},"end"," of the quarter, but has up to 45 days to publish it. By the time you see it, the information can be three or four months old, and the manager may already have sold.",[29,2959,2960,2963,2964,2967],{},[43,2961,2962],{},"2. They show only long, US positions."," You do not see shorts, you do not see most options positions, you do not see cash, bonds or foreign stocks. You see half the picture. A fund can hold a stock long ",[2230,2965,2966],{},"and"," have hedged it with positions that simply never appear in the 13F.",[29,2969,2970,2973],{},[43,2971,2972],{},"3. Herd risk."," Ten famous funds holding the same stock does not mean they are right; it means they think alike. The most dangerous trade is often the one all the \"smart\" money has already made, because the market has already priced it in.",[29,2975,2976,2977,2980],{},"And the most important limit is conceptual: ",[43,2978,2979],{},"\"a big fund holds it\" is not a reason to buy."," You do not know the price they paid, how large the position is as a share of their portfolio, the time horizon, or the role it plays in their overall strategy. You are copying a move without knowing its reasoning, and the reasoning is the entire value.",[36,2982,2984],{"id":2983},"insiders-form-4-a-cleaner-signal","Insiders (Form 4): a cleaner signal?",[29,2986,2987,2988,2991,2992,2995],{},"There is a second kind of signal many consider more reliable: ",[43,2989,2990],{},"insiders",". When an officer (CEO, CFO, board member) buys or sells their own company's stock, they report it to the SEC on a ",[43,2993,2994],{},"Form 4",", usually within two business days, so it is far fresher than a 13F.",[29,2997,2998],{},"The logic: nobody knows a company better than the people running it. When they put their own money down buying in the open market (not through options grants), it is a vote of confidence with real cost behind it.",[29,3000,3001,3002,3005,3006,3009],{},"But here too, be careful: insiders ",[43,3003,3004],{},"sell"," for a thousand reasons (taxes, diversification, buying a house) that say nothing about the company, while they mostly ",[43,3007,3008],{},"buy"," for one. That is why insider buying, especially \"cluster buying\" by several officers at once, is more interesting than selling. It remains a signal, not proof.",[36,3011,3013],{"id":3012},"how-to-read-them-properly","How to read them properly",[29,3015,3016],{},"The right way to use smart-money data is not as a shopping list to copy, but as:",[81,3018,3019,3025,3031],{},[84,3020,3021,3024],{},[43,3022,3023],{},"An idea source."," \"Why did three value funds enter this sector this year?\" is a good prompt to start your own research.",[84,3026,3027,3030],{},[43,3028,3029],{},"Confirmation or challenge."," If you have already done the work and find that experienced, long-horizon investors hold the same name, that adds a little confidence. If you find everyone leaving, it is worth asking why.",[84,3032,3033,3036,3037,3040],{},[43,3034,3035],{},"Never a command."," The signal is the ",[2230,3038,3039],{},"start"," of research, not the end of it.",[29,3042,3043],{},"The move from \"Buffett holds it\" to \"I understand why, and I agree with the reasoning\" is the whole job. What you skip in that move is exactly your risk.",[36,3045,3047],{"id":3046},"how-ploutos-uses-it","How Ploutos uses it",[29,3049,3050,3051,3054],{},"At Ploutos, smart-money data is ",[43,3052,3053],{},"one lens among many",", not the driver. In every analysis we track whether any of 12 legendary investors holds the stock via 13F, and whether there is recent insider buying via Form 4, flagging explicitly when cluster buying occurs.",[29,3056,3057,3058,3061,3062,3064],{},"But, crucially, it sits ",[43,3059,3060],{},"next to"," the fundamentals, the valuation and, above all, the ",[43,3063,2714],{},", which actively builds the counter-argument: why the \"smart money\" might be wrong here, what could go wrong, what the bear case is. You will never see \"X holds it, so buy.\" You will see \"X holds it, here is what that means, and here is why it may not matter for your situation.\"",[29,3066,3067,3068,3071,3072,52],{},"To see how it fits the whole picture, read ",[48,3069,3070],{"href":238},"how Ploutos analyzes a stock",", or how we weigh quality through the ",[48,3073,1393],{"href":1392},[36,3075,3077],{"id":3076},"a-starting-point-not-a-shortcut","A starting point, not a shortcut",[29,3079,3080],{},"13F and insider signals are among the most underrated free tools for the individual investor, and at the same time among the most misused. Read correctly, they give you ideas and context. Read badly, they make you copy moves you do not understand, late and without the hedge that may accompany them.",[29,3082,3083,3084,3086,3087,3089,3090,52],{},"Their value is not to tell you what to buy. It is to show you where to ",[2230,3085,3039],{}," looking. You can see smart-money signals fold into a full analysis if you ",[48,3088,2518],{"href":1236},", or see stocks that stand out on the ",[48,3091,2318],{"href":2317},{"title":438,"searchDepth":439,"depth":439,"links":3093},[3094,3095,3096,3097,3098,3099],{"id":2929,"depth":442,"text":2930},{"id":2943,"depth":442,"text":2944},{"id":2983,"depth":442,"text":2984},{"id":3012,"depth":442,"text":3013},{"id":3046,"depth":442,"text":3047},{"id":3076,"depth":442,"text":3077},"What 13F filings are, what smart-money and insider signals actually tell you (and what they don't), and how to read them as a research starting point, not a buy command.",[3102,3105,3108,3111],{"q":3103,"a":3104},"What is a Form 13F?","It is the quarterly SEC filing that requires every institutional manager with more than $100 million in US equities to disclose what they hold. It is public and free, which is why you can see the portfolios of funds like Berkshire Hathaway.",{"q":3106,"a":3107},"Can I just buy whatever Buffett buys?","No. 13F filings are delayed up to 45 days, show only long US positions (not shorts or hedges), and never tell you the price paid or the time horizon. They are a signal to research, not a buy command.",{"q":3109,"a":3110},"How are insider filings (Form 4) different from 13F?","Form 4 filings are made by company officers (CEO\u002FCFO\u002Fboard) on trades in their own company's stock, usually within two business days, so they are far fresher. Insider buying, especially by several officers at once, is a more interesting signal than selling.",{"q":3112,"a":3113},"Is smart money reliable in the end?","It is useful as an idea source and as confirmation after your own research, not as proof. Watch the herd risk: if all the smart money already holds something, the market may have already priced it in.",{},"What does 13F 'smart money' really tell you?",{"title":2906,"description":3100},"blog\u002Fen\u002Fsmart-money-13f",[1526,1295],"smart-money-13f","UALqvPua3ySVpbHx2geN7AnIlUGRmk88N8zUmeOXXnA",{"id":4,"title":5,"author":6,"body":3122,"date":457,"description":458,"extension":459,"faq":3413,"featured":473,"mentionsTickers":473,"meta":3418,"metaTitle":475,"navigation":476,"ogImage":27,"path":477,"readMinutes":478,"seo":3419,"stem":480,"tags":3420,"translationKey":484,"__hash__":485},{"type":8,"value":3123,"toc":3396},[3124,3130,3132,3134,3136,3142,3146,3152,3154,3156,3158,3174,3176,3178,3180,3182,3184,3186,3188,3190,3192,3194,3196,3198,3200,3202,3204,3206,3208,3210,3212,3214,3216,3218,3220,3222,3224,3226,3228,3230,3232,3234,3236,3238,3240,3242,3244,3250,3254,3256,3258,3386,3388,3390,3392,3394],[11,3125,3127],{"className":3126},[14,15,16,17,18,19],[21,3128],{"alt":23,"className":3129,"src":27},[25,26],[29,3131,31],{},[29,3133,34],{},[36,3135,39],{"id":38},[29,3137,3138,46,3140,52],{},[43,3139,45],{},[48,3141,51],{"href":50},[29,3143,3144,58],{},[43,3145,57],{},[29,3147,3148,64,3150,69],{},[43,3149,63],{},[66,3151,68],{},[29,3153,72],{},[36,3155,76],{"id":75},[29,3157,79],{},[81,3159,3160,3164,3170],{},[84,3161,3162,89],{},[43,3163,88],{},[84,3165,3166,95,3168,99],{},[43,3167,94],{},[66,3169,98],{},[84,3171,3172,105],{},[43,3173,104],{},[29,3175,108],{},[29,3177,111],{},[36,3179,115],{"id":114},[29,3181,118],{},[120,3183,123],{"id":122},[29,3185,126],{},[29,3187,129],{},[29,3189,132],{},[29,3191,135],{},[120,3193,139],{"id":138},[29,3195,142],{},[29,3197,145],{},[29,3199,148],{},[120,3201,152],{"id":151},[29,3203,155],{},[29,3205,158],{},[29,3207,161],{},[120,3209,165],{"id":164},[29,3211,168],{},[29,3213,171],{},[29,3215,174],{},[120,3217,178],{"id":177},[29,3219,181],{},[29,3221,184],{},[120,3223,188],{"id":187},[29,3225,191],{},[29,3227,194],{},[120,3229,198],{"id":197},[29,3231,201],{},[29,3233,204],{},[120,3235,208],{"id":207},[29,3237,211],{},[29,3239,214],{},[120,3241,218],{"id":217},[29,3243,221],{},[29,3245,224,3246,227,3248,232],{},[66,3247,98],{},[48,3249,231],{"href":230},[29,3251,235,3252,52],{},[48,3253,239],{"href":238},[29,3255,242],{},[36,3257,246],{"id":245},[248,3259,3260,3272],{},[251,3261,3262],{},[254,3263,3264,3266,3268,3270],{},[257,3265,259],{},[257,3267,262],{},[257,3269,265],{},[257,3271,268],{},[270,3273,3274,3284,3294,3304,3314,3324,3334,3344,3354,3364,3374],{},[254,3275,3276,3278,3280,3282],{},[275,3277,277],{},[275,3279,280],{},[275,3281,283],{},[275,3283,286],{},[254,3285,3286,3288,3290,3292],{},[275,3287,291],{},[275,3289,294],{},[275,3291,297],{},[275,3293,300],{},[254,3295,3296,3298,3300,3302],{},[275,3297,305],{},[275,3299,308],{},[275,3301,311],{},[275,3303,314],{},[254,3305,3306,3308,3310,3312],{},[275,3307,319],{},[275,3309,322],{},[275,3311,325],{},[275,3313,328],{},[254,3315,3316,3318,3320,3322],{},[275,3317,333],{},[275,3319,336],{},[275,3321,339],{},[275,3323,342],{},[254,3325,3326,3328,3330,3332],{},[275,3327,165],{},[275,3329,349],{},[275,3331,352],{},[275,3333,355],{},[254,3335,3336,3338,3340,3342],{},[275,3337,178],{},[275,3339,362],{},[275,3341,365],{},[275,3343,314],{},[254,3345,3346,3348,3350,3352],{},[275,3347,188],{},[275,3349,374],{},[275,3351,377],{},[275,3353,380],{},[254,3355,3356,3358,3360,3362],{},[275,3357,198],{},[275,3359,387],{},[275,3361,390],{},[275,3363,393],{},[254,3365,3366,3368,3370,3372],{},[275,3367,208],{},[275,3369,400],{},[275,3371,403],{},[275,3373,406],{},[254,3375,3376,3378,3382,3384],{},[275,3377,218],{},[275,3379,413,3380],{},[66,3381,98],{},[275,3383,418],{},[275,3385,342],{},[36,3387,424],{"id":423},[29,3389,427],{},[29,3391,430],{},[29,3393,433],{},[29,3395,436],{},{"title":438,"searchDepth":439,"depth":439,"links":3397},[3398,3399,3400,3411,3412],{"id":38,"depth":442,"text":39},{"id":75,"depth":442,"text":76},{"id":114,"depth":442,"text":115,"children":3401},[3402,3403,3404,3405,3406,3407,3408,3409,3410],{"id":122,"depth":439,"text":123},{"id":138,"depth":439,"text":139},{"id":151,"depth":439,"text":152},{"id":164,"depth":439,"text":165},{"id":177,"depth":439,"text":178},{"id":187,"depth":439,"text":188},{"id":197,"depth":439,"text":198},{"id":207,"depth":439,"text":208},{"id":217,"depth":439,"text":218},{"id":245,"depth":442,"text":246},{"id":423,"depth":442,"text":424},[3414,3415,3416,3417],{"q":462,"a":463},{"q":465,"a":466},{"q":468,"a":469},{"q":471,"a":472},{},{"title":5,"description":458},[482,483],{"id":3422,"title":3423,"author":6,"body":3424,"date":3560,"description":3561,"extension":459,"faq":3562,"featured":473,"mentionsTickers":473,"meta":3575,"metaTitle":3564,"navigation":476,"ogImage":3433,"path":3576,"readMinutes":2898,"seo":3577,"stem":3578,"tags":3579,"translationKey":3580,"__hash__":3581},"blogEn\u002Fblog\u002Fen\u002Fwhat-is-a-stock-split.md","What is a stock split, and does it change what your shares are worth?",{"type":8,"value":3425,"toc":3553},[3426,3434,3437,3441,3444,3447,3451,3477,3481,3507,3511,3514,3525,3528,3532,3538,3545],[11,3427,3429],{"className":3428},[14,15,16,17,18,19],[21,3430],{"alt":3431,"className":3432,"src":3433},"A stock split explained: same pizza, more slices, total value is unchanged",[25,26],"\u002Fblog\u002Fog-stock-split.svg",[29,3435,3436],{},"You check your portfolio and the price of a stock you own just dropped 90% overnight. Your first instinct is panic. Then you notice your share count went up by the same proportion, and the actual value of your position did not move at all. What you just watched happen was a stock split, and it is one of the most misunderstood events in investing precisely because the headline number moves so dramatically while nothing of substance changes underneath it.",[36,3438,3440],{"id":3439},"what-actually-happens","What actually happens",[29,3442,3443],{},"A stock split is a company dividing each existing share into a larger number of new shares. A common example is a 10 for 1 split. If you owned 1 share worth $2,300 before the split, you own 10 shares worth $230 each right after it. Multiply it out and you get the same $2,300 either way.",[29,3445,3446],{},"Nothing about the company changed at the moment of the split. Not its revenue, not its profits, not its assets, not the percentage of the company you own. The only thing that changed is how that same ownership stake is sliced up on paper. Same pizza, more slices, same total value.",[36,3448,3450],{"id":3449},"what-does-not-change","What does not change",[81,3452,3453,3459,3465,3471],{},[84,3454,3455,3458],{},[43,3456,3457],{},"The value of your position."," Ten shares at $230 is worth exactly what one share at $2,300 was worth.",[84,3460,3461,3464],{},[43,3462,3463],{},"Your ownership percentage."," You still own the same fraction of the company as before.",[84,3466,3467,3470],{},[43,3468,3469],{},"The company's fundamentals."," Revenue, margins, debt, competitive position, none of it is touched by a split.",[84,3472,3473,3476],{},[43,3474,3475],{},"The underlying valuation."," If the stock was fairly valued, overvalued, or undervalued before the split, it is exactly as fairly valued, overvalued, or undervalued right after it, once you adjust the numbers for the new share count.",[36,3478,3480],{"id":3479},"what-does-change","What does change",[81,3482,3483,3489,3495,3501],{},[84,3484,3485,3488],{},[43,3486,3487],{},"The share count",", obviously, since that is the entire mechanism.",[84,3490,3491,3494],{},[43,3492,3493],{},"The price per share",", in exact inverse proportion to the share count.",[84,3496,3497,3500],{},[43,3498,3499],{},"Per-share metrics like earnings per share."," If a company earned $10 per share before a 10 for 1 split, it earns $1 per share right after, on the same total profit spread across ten times as many shares.",[84,3502,3503,3506],{},[43,3504,3505],{},"Accessibility and liquidity",", which is the real, practical reason companies do this. A $230 share price is easier for a retail investor to buy in round lots than a $2,300 one, and a lower per-share price tends to widen the pool of potential buyers and increase trading volume.",[36,3508,3510],{"id":3509},"why-companies-actually-split-their-stock","Why companies actually split their stock",[29,3512,3513],{},"There is no legal or financial requirement to split a stock, and plenty of well known companies never do it, keeping share prices in the thousands of dollars on purpose. The usual reasons a company chooses to split instead:",[81,3515,3516,3519,3522],{},[84,3517,3518],{},"Keeping the share price in a range that feels approachable to individual investors, since a high sticker price can (irrationally, but in practice) discourage smaller buyers.",[84,3520,3521],{},"Supporting liquidity and tighter bid ask spreads by increasing the number of shares available to trade.",[84,3523,3524],{},"Sometimes signaling management's confidence that the higher price will hold or keep climbing, though this is a soft signal at best, not a fundamental one.",[29,3526,3527],{},"None of these reasons involve the business becoming worth more. A split is a cosmetic, mechanical adjustment, not a value creation event.",[36,3529,3531],{"id":3530},"the-myth-worth-retiring","The myth worth retiring",[29,3533,3534,3535,3537],{},"The mistake people make most often is treating a split as if it made a stock cheaper, and therefore a better buy. It did not. A $2,300 stock that splits 10 for 1 and becomes a $230 stock is exactly as expensive, relative to the business behind it, as it was the day before. If it was trading above ",[48,3536,1340],{"href":1417}," at $2,300, the split version at $230 is trading above the equivalent, split adjusted fair value too. The lower sticker price does nothing to change that math, it just changes the denominator.",[29,3539,3540,3541,3544],{},"The only thing a split genuinely earns you is a lower barrier to buying a smaller dollar amount, and slightly better liquidity. If you are deciding whether a stock is a good investment, the split itself is not evidence either way, you still have to answer ",[48,3542,3543],{"href":2754},"the real question",": is the price, whatever it happens to be denominated in today, low or high relative to what the business is actually worth.",[29,3546,3547,3548,3550,3551,52],{},"One detail that matters more than it should: historical financial data needs to be adjusted for every split a company has ever done, or ratios like earnings per share and price to earnings end up compared against numbers from a different share count entirely, quietly corrupting the analysis. It is a small, boring detail, and it is exactly the kind of thing that is easy to get wrong by hand and easy to automate correctly. If you want to see how a full valuation and the rest of the research come together once splits, and everything else, are accounted for, you can ",[48,3549,1237],{"href":1236}," or read the ",[48,3552,2725],{"href":238},{"title":438,"searchDepth":439,"depth":439,"links":3554},[3555,3556,3557,3558,3559],{"id":3439,"depth":442,"text":3440},{"id":3449,"depth":442,"text":3450},{"id":3479,"depth":442,"text":3480},{"id":3509,"depth":442,"text":3510},{"id":3530,"depth":442,"text":3531},"2026-07-25","A stock split multiplies your share count and divides the price by the same amount. Here is what actually changes, what does not, and the one thing worth checking afterward.",[3563,3566,3569,3572],{"q":3564,"a":3565},"What is a stock split?","A company dividing each existing share into multiple new shares, without changing the total value of what shareholders own.",{"q":3567,"a":3568},"Does a stock split make a stock cheaper or more valuable?","Neither. The price per share drops in the same proportion the share count rises, so the total value of your position is unchanged.",{"q":3570,"a":3571},"Why do companies do stock splits?","Mostly to keep the per-share price approachable for individual investors and to support liquidity, not because anything about the business changed.",{"q":3573,"a":3574},"Does a split affect whether a stock looks cheap or expensive?","No. Per-share numbers like earnings and fair value scale down by the same factor as the price, so the underlying valuation is unchanged.",{},"\u002Fblog\u002Fwhat-is-a-stock-split",{"title":3423,"description":3561},"blog\u002Fen\u002Fwhat-is-a-stock-split",[1526,1295],"what-is-a-stock-split","-RB6xeCfouqKjjLNXn9Ja-UbUWYcR_F3xw5MZz2q3E8",{"id":3583,"title":3584,"author":6,"body":3585,"date":2880,"description":3779,"extension":459,"faq":3780,"featured":473,"mentionsTickers":473,"meta":3793,"metaTitle":3782,"navigation":476,"ogImage":3594,"path":1392,"readMinutes":2553,"seo":3794,"stem":3795,"tags":3796,"translationKey":3797,"__hash__":3798},"blogEn\u002Fblog\u002Fen\u002Fwhat-is-an-economic-moat.md","What is an economic moat, and how do you actually measure it?",{"type":8,"value":3586,"toc":3771},[3587,3595,3602,3609,3613,3620,3623,3627,3630,3662,3669,3673,3676,3687,3691,3694,3704,3710,3713,3717,3720,3743,3746,3750,3753,3763],[11,3588,3590],{"className":3589},[14,15,16,17,18,19],[21,3591],{"alt":3592,"className":3593,"src":3594},"An economic moat shown as a defended core with above-sector returns",[25,26],"\u002Fblog\u002Fog-moat.svg",[29,3596,3597,3598,3601],{},"Warren Buffett popularised one of the most useful images in investing. Picture a great business as a castle, and its profits as the treasure inside. Every profitable castle attracts attackers, competitors who want a share of that treasure. What protects it over time is the ",[43,3599,3600],{},"moat",": a structural advantage that makes the castle genuinely hard to attack.",[29,3603,3604,3605,3608],{},"A company without a moat can earn high returns for a while, but competition eventually arrives, undercuts its prices, and grinds those returns back down to ordinary. A company ",[2230,3606,3607],{},"with"," a moat can defend its profits for years, sometimes decades. For a long-term investor, almost nothing matters more, and almost nothing is harder to measure honestly.",[36,3610,3612],{"id":3611},"what-an-economic-moat-actually-is","What an economic moat actually is",[29,3614,3615,3616,3619],{},"An economic moat is a ",[2230,3617,3618],{},"durable"," competitive advantage, the operative word being durable. It is the reason a business can keep earning returns on its capital that are well above the cost of that capital, without competition immediately competing those returns away.",[29,3621,3622],{},"That phrase, returns above the cost of capital, is the whole game. Any company can grow by pouring in more money. The question that separates a great business from a merely large one is whether each dollar it invests comes back as meaningfully more than a dollar, year after year, even when rivals are trying to take its lunch. A moat is the structural reason the answer stays yes.",[36,3624,3626],{"id":3625},"the-main-types-of-moat","The main types of moat",[29,3628,3629],{},"Moats come in a handful of recognisable shapes:",[81,3631,3632,3638,3644,3650,3656],{},[84,3633,3634,3637],{},[43,3635,3636],{},"Network effects."," The product gets more valuable as more people use it. A marketplace or a payments network is worth more to each user precisely because everyone else is already there, which makes it extremely hard for a newcomer to break in.",[84,3639,3640,3643],{},[43,3641,3642],{},"Switching costs."," Once a customer is embedded, leaving is painful, expensive, or risky. Enterprise software that a whole company runs on is a classic example, the cost is not the subscription, it is the migration.",[84,3645,3646,3649],{},[43,3647,3648],{},"Cost advantage."," The business can produce the same thing more cheaply than anyone else, through scale, location, or a unique process, and can therefore undercut rivals and still profit.",[84,3651,3652,3655],{},[43,3653,3654],{},"Intangible assets."," Brands that let a company charge a premium, patents that lock out imitation, or regulatory licences that competitors cannot easily obtain.",[84,3657,3658,3661],{},[43,3659,3660],{},"Efficient scale."," A market just big enough for one or two players to serve profitably, where a third entrant would make the economics bad for everyone, so none arrive.",[29,3663,3664,3665,3668],{},"The names matter less than the test they all have to pass: does this advantage ",[2230,3666,3667],{},"persist"," when a well-funded competitor attacks it? A trendy product is not a moat. A patent that expires next year is a shrinking one.",[36,3670,3672],{"id":3671},"the-measurement-problem","The measurement problem",[29,3674,3675],{},"Here is the hard part. The descriptions above are qualitative, you can only really confirm a network effect or a switching cost by understanding the business deeply, reading its filings, and watching how it behaves over years. You cannot do that, by hand, across thousands of companies. So how does an automated process judge something this subtle?",[29,3677,3678,3679,3682,3683,3686],{},"The answer is to stop trying to read the moat directly and instead look for the ",[43,3680,3681],{},"fingerprint it leaves in the financial statements."," A real, durable advantage is not just a story, it shows up in the numbers, because a business that can defend its profits ",[2230,3684,3685],{},"actually keeps earning them."," That is something you can measure.",[36,3688,3690],{"id":3689},"the-financial-fingerprint-of-a-moat","The financial fingerprint of a moat",[29,3692,3693],{},"Two signals do most of the work.",[29,3695,3696,3699,3700,3703],{},[43,3697,3698],{},"High and persistent return on invested capital (ROIC)."," A single great year of ROIC proves nothing, anyone can get lucky once. What a moat produces is ROIC that stays well above the sector average ",[2230,3701,3702],{},"year after year after year",". The persistence is the signal. A business that earns far more on its capital than its peers, and keeps doing so across an entire cycle, is almost certainly being protected by something structural, even if you cannot name it from the outside. So we look not just at the level of ROIC but at how many of the recent years cleared the sector bar.",[29,3705,3706,3709],{},[43,3707,3708],{},"Premium gross margins."," Pricing power, the ability to charge more than rivals without losing customers, is the most visible symptom of a moat. It shows up as a gross margin meaningfully above the sector. A company with no advantage has to compete on price, which compresses margins toward the industry average. A company that holds a premium margin for years is telling you customers are paying up for something they cannot easily get elsewhere.",[29,3711,3712],{},"Combine the two and you get a defensible, repeatable classification, which is exactly how Ploutos AI grades it.",[36,3714,3716],{"id":3715},"wide-narrow-or-none","Wide, narrow, or none",[29,3718,3719],{},"The classification lands in three buckets:",[81,3721,3722,3731,3737],{},[84,3723,3724,3727,3728,3730],{},[43,3725,3726],{},"Wide moat."," ROIC has been far above the sector, with consistency across most of the observed years, ",[2230,3729,2966],{}," gross margins carry a clear premium. This is the rare, durable-advantage profile.",[84,3732,3733,3736],{},[43,3734,3735],{},"Narrow moat."," A real but more modest edge, returns above the sector with decent consistency, or clear pricing power with respectable returns. An advantage, but one a determined competitor could erode.",[84,3738,3739,3742],{},[43,3740,3741],{},"None."," The financial fingerprint of commodity-like economics: returns near or below the cost of capital, no margin premium. The business may still be a fine investment at the right price, but you should not pay up expecting a durable advantage that the numbers do not support.",[29,3744,3745],{},"No single year and no single metric earns a \"wide\" label. It takes a pattern that holds up over time, which is the point, a moat that only existed last quarter was never a moat.",[36,3747,3749],{"id":3748},"why-a-moat-changes-how-much-margin-of-safety-you-need","Why a moat changes how much margin of safety you need",[29,3751,3752],{},"This is where the concept becomes practical. A moat does not just describe a business, it should change the price you are willing to pay for it.",[29,3754,3755,3756,3759,3760,3762],{},"A wide-moat company can reinvest its profits at high returns for years, which means its intrinsic value tends to ",[2230,3757,3758],{},"grow"," over time. That durability justifies accepting a smaller discount to fair value, a smaller ",[48,3761,1425],{"href":1417},", because time is working for you. A no-moat business is the opposite: its advantage, if any, is fragile, its returns are more likely to fade, so you should demand a larger discount to compensate for that fragility before the price looks attractive.",[29,3764,3765,3766,3768,3769,52],{},"In other words, quality and price are not separate questions, they are the same question. A moat is one of the cleanest ways to ask it rigorously. If you want to see a moat classification produced alongside a full valuation and the rest of the research, you can ",[48,3767,1237],{"href":1236},", or read how moat, valuation, and the other signals come together in the ",[48,3770,2725],{"href":238},{"title":438,"searchDepth":439,"depth":439,"links":3772},[3773,3774,3775,3776,3777,3778],{"id":3611,"depth":442,"text":3612},{"id":3625,"depth":442,"text":3626},{"id":3671,"depth":442,"text":3672},{"id":3689,"depth":442,"text":3690},{"id":3715,"depth":442,"text":3716},{"id":3748,"depth":442,"text":3749},"An economic moat is a durable competitive advantage. Here is what one looks like, the main types, and how to measure it from a company's financial fingerprint.",[3781,3784,3787,3790],{"q":3782,"a":3783},"What is an economic moat?","The durable competitive advantage that protects a company's profits from competitors, like a moat around a castle.",{"q":3785,"a":3786},"What types of moat are there?","The main ones: network effects, switching costs, intangibles (brand, patents), cost advantage, and efficient scale.",{"q":3788,"a":3789},"How do you measure a moat in practice?","From durable high margins and ROIC over time, and from how hard it is for someone to copy the business.",{"q":3791,"a":3792},"Why does a moat matter to an investor?","A wide moat means more predictable and durable profits, which support a valuation over the long run.",{},{"title":3584,"description":3779},"blog\u002Fen\u002Fwhat-is-an-economic-moat",[1526,1295],"what-is-an-economic-moat","L1O_7c59al-sZRyK-mRSpEW1jM-VMvMobGNI64ByB-I",{"id":3800,"title":3801,"author":6,"body":3802,"date":2533,"description":3964,"extension":459,"faq":3965,"featured":473,"mentionsTickers":473,"meta":3981,"metaTitle":3982,"navigation":476,"ogImage":3811,"path":1429,"readMinutes":1290,"seo":3983,"stem":3984,"tags":3985,"translationKey":3986,"__hash__":3987},"blogEn\u002Fblog\u002Fen\u002Fwhat-is-an-undervalued-stock.md","What does 'undervalued' really mean: margin of safety and value traps",{"type":8,"value":3803,"toc":3956},[3804,3812,3815,3819,3829,3832,3836,3846,3849,3853,3863,3866,3880,3886,3890,3897,3900,3926,3932,3936,3939,3941,3951],[11,3805,3807],{"className":3806},[14,15,16,17,18,19],[21,3808],{"alt":3809,"className":3810,"src":3811},"Price below fair value, with the margin of safety in between",[25,26],"\u002Fblog\u002Fog-margin-of-safety.svg",[29,3813,3814],{},"One of the most misunderstood words in investing is \"undervalued\". Most people confuse it with \"cheap\", meaning a low number on the ticker. They are not the same, and the difference is the whole game.",[36,3816,3818],{"id":3817},"price-vs-value","Price vs value",[29,3820,3821,3824,3825,3828],{},[43,3822,3823],{},"Price"," is what you pay. ",[43,3826,3827],{},"Value"," is what you get. A stock is undervalued when its price is below the intrinsic value of the business, regardless of whether the price number is large or small.",[29,3830,3831],{},"A stock at 500 can be undervalued if the company is worth 800. A stock at 2 can be expensive if the company is worth 1. \"Cheap as a number\" tells you nothing on its own.",[36,3833,3835],{"id":3834},"how-you-approach-value","How you approach \"value\"",[29,3837,3838,3839,3842,3843,52],{},"Intrinsic value is not visible on the ticker, you estimate it. The most common way is a ",[43,3840,3841],{},"DCF (Discounted Cash Flow)",": you work out how much cash the company will generate in the future and discount it back to today. A DCF is not a crystal ball, it's a model, and it's only as good as its assumptions. For how it works and why analyst price targets so often miss, see the ",[48,3844,3845],{"href":1417},"guide to DCF valuation",[29,3847,3848],{},"Because every value estimate carries uncertainty, it isn't enough to buy \"at value\". You need a buffer.",[36,3850,3852],{"id":3851},"margin-of-safety-the-most-important-part","Margin of safety: the most important part",[29,3854,3855,3856,3858,3859,3862],{},"The ",[43,3857,1425],{}," is the idea that made Benjamin Graham and Warren Buffett famous: buy far enough ",[43,3860,3861],{},"below"," your estimate of value that, even if your estimate is wrong, you aren't ruined.",[29,3864,3865],{},"If you estimate a company is worth 100 per share and you buy at 70, you have a 30% margin. That protects you from two things:",[81,3867,3868,3874],{},[84,3869,3870,3873],{},[43,3871,3872],{},"Being wrong on the estimate."," Maybe the company is actually worth 85, not 100. Buying at 70, you still come out ahead.",[84,3875,3876,3879],{},[43,3877,3878],{},"Bad luck."," An unexpected bad year, a sector shock. The margin absorbs part of the hit.",[29,3881,3882,3883,3885],{},"The more uncertain the company (cyclical, no strong ",[48,3884,1393],{"href":1392},"), the bigger the margin you want.",[36,3887,3889],{"id":3888},"the-trap-value-traps","The trap: value traps",[29,3891,3892,3893,3896],{},"This is where most people lose. A stock that looks cheap (low P\u002FE, beaten-down price) is not automatically an opportunity. It can be a ",[43,3894,3895],{},"value trap",": cheap for a reason.",[29,3898,3899],{},"Signs you are looking at a value trap, not an opportunity:",[81,3901,3902,3908,3914,3920],{},[84,3903,3904,3907],{},[43,3905,3906],{},"A declining business."," Revenue or margins shrinking year after year.",[84,3909,3910,3913],{},[43,3911,3912],{},"Losing share"," to a faster competitor or a new technology.",[84,3915,3916,3919],{},[43,3917,3918],{},"Burning cash"," or carrying debt that is strangling it.",[84,3921,3922,3925],{},[43,3923,3924],{},"The \"cheap\" never closes."," The price keeps falling and the low P\u002FE just trails the falling earnings.",[29,3927,3928,3929,3931],{},"The difference between an opportunity and a trap is not in the price. It's in the ",[43,3930,2649],{},": is it temporarily misunderstood, or permanently broken?",[36,3933,3935],{"id":3934},"the-special-case-companies-with-no-earnings","The special case: companies with no earnings",[29,3937,3938],{},"When a company has no earnings yet (many tech, early-stage), P\u002FE doesn't help, there isn't one. There, valuation becomes purely forward-looking: you have to estimate when and how much it will become profitable, based on the business. It's the hardest case, and where most people either dramatically over- or under-estimate.",[36,3940,1692],{"id":1691},[29,3942,3943,3944,3946,3947,3950],{},"\"Undervalued\" does not mean \"cheap\". It means: you estimate the value of the business, you buy far enough below it, and you make sure the discount is due to the market misunderstanding something, not a real problem. The two tools are the ",[43,3945,1425],{}," (buying discipline) and ",[43,3948,3949],{},"understanding the business"," (so you don't fall into a value trap).",[29,3952,3953,3954,2519],{},"If you want to see fair value computed with DCF, margin of safety, and a read of the business in one structured analysis, you can ",[48,3955,2518],{"href":1236},{"title":438,"searchDepth":439,"depth":439,"links":3957},[3958,3959,3960,3961,3962,3963],{"id":3817,"depth":442,"text":3818},{"id":3834,"depth":442,"text":3835},{"id":3851,"depth":442,"text":3852},{"id":3888,"depth":442,"text":3889},{"id":3934,"depth":442,"text":3935},{"id":1691,"depth":442,"text":1692},"A low price doesn't mean undervalued. See what intrinsic value is, what margin of safety means, and how to tell a real opportunity from a value trap.",[3966,3969,3972,3975,3978],{"q":3967,"a":3968},"What does 'undervalued' mean?","An undervalued stock is one whose market price is below the intrinsic (real) value of the business. It has nothing to do with whether the price is a small number. A stock at 500 can be undervalued and one at 3 can be expensive.",{"q":3970,"a":3971},"What is margin of safety?","It is the safety buffer: you buy far enough below your estimate of value that, even if your estimate is somewhat wrong, you don't lose. If you estimate value at 100 and buy at 70, you have a 30% margin of safety.",{"q":3973,"a":3974},"How do I work out a stock's real value?","The most common method is a DCF: you discount the company's future cash flows back to today. It's an estimate, not an exact number, and it depends on your assumptions, which is exactly why you also need a margin of safety.",{"q":3976,"a":3977},"What is a value trap?","A stock that looks cheap (low P\u002FE, low price) but is cheap for a reason: the business is declining, losing share, or burning cash. The price keeps falling and the 'discount' never closes.",{"q":3979,"a":3980},"Does a low P\u002FE mean a stock is undervalued?","Not necessarily. A low P\u002FE can mean the market expects earnings to fall (value trap), or that the sector is cyclical. P\u002FE is a hint, not proof of undervaluation.",{},"What does undervalued mean (margin of safety)",{"title":3801,"description":3964},"blog\u002Fen\u002Fwhat-is-an-undervalued-stock",[2758,1295],"what-is-an-undervalued-stock","9l_kcKYhPhiy-YH1epZjvfb98EIGD4YjzRbLbwkHOjA",{"id":3989,"title":3990,"author":6,"body":3991,"date":2880,"description":4175,"extension":459,"faq":4176,"featured":473,"mentionsTickers":473,"meta":4189,"metaTitle":4190,"navigation":476,"ogImage":4000,"path":1417,"readMinutes":2553,"seo":4191,"stem":4192,"tags":4193,"translationKey":4194,"__hash__":4195},"blogEn\u002Fblog\u002Fen\u002Fwhat-is-dcf-valuation.md","What is DCF valuation, and why analyst price targets are often wrong",{"type":8,"value":3992,"toc":4167},[3993,4001,4008,4011,4015,4022,4029,4033,4039,4049,4059,4062,4066,4069,4072,4076,4083,4125,4139,4143,4146,4149,4153,4159],[11,3994,3996],{"className":3995},[14,15,16,17,18,19],[21,3997],{"alt":3998,"className":3999,"src":4000},"Future cash flows discounted back to a present value",[25,26],"\u002Fblog\u002Fog-dcf.svg",[29,4002,4003,4004,4007],{},"The price of a stock is what the market is asking for it today. The ",[2230,4005,4006],{},"value"," of a stock is a separate question: what is the underlying business actually worth? Those two numbers are often very different, and the entire discipline of value investing lives in the gap between them.",[29,4009,4010],{},"Discounted cash flow, or DCF, is the most fundamental tool for answering the second question. It is also one of the most misused. This article explains what a DCF really does, why a single DCF can be dangerously fragile, and why a careful process leans on it without trusting it alone.",[36,4012,4014],{"id":4013},"the-one-idea-behind-dcf-time-has-a-price","The one idea behind DCF: time has a price",[29,4016,4017,4018,4021],{},"Imagine someone offers you 1,000 euros. Would you rather have it today or in five years? Today, obviously, and not only because of impatience. Money today can be invested, it carries no risk of never arriving, and inflation erodes the value of money you receive later. So a euro in the future is worth ",[2230,4019,4020],{},"less"," than a euro now, and the further away it is, the less it is worth.",[29,4023,4024,4025,4028],{},"DCF takes that intuition and turns it into arithmetic. A business is, in the end, a machine for producing cash over time. If you can estimate the cash it will generate in the years ahead, and you can put a price on waiting, then you can translate all that future cash into a single number: what it is worth ",[2230,4026,4027],{},"today",". That number is the company's intrinsic value, and dividing it by the share count gives you a value per share to compare against the market price.",[36,4030,4032],{"id":4031},"how-a-dcf-actually-works-in-three-steps","How a DCF actually works, in three steps",[29,4034,4035,4038],{},[43,4036,4037],{},"Step 1: Project the cash flows."," Start from the company's free cash flow, the cash left over after it pays to run and maintain the business, and project it forward, usually about ten years, using a growth rate. A mature, stable business might be assumed to grow that cash flow at a low single-digit rate, a faster-growing company at something higher, typically tapering toward a modest long-term rate as it matures.",[29,4040,4041,4044,4045,4048],{},[43,4042,4043],{},"Step 2: Discount each year back to today."," This is the \"discounted\" part. Each future year's cash flow is divided by a ",[2230,4046,4047],{},"discount rate"," that reflects risk and the time value of money. A common shorthand: at a 10% discount rate, 1,000 euros a year from now is worth about 909 euros today, and 1,000 euros ten years out is worth only about 386. Riskier businesses get a higher discount rate, which shrinks their future cash more aggressively.",[29,4050,4051,4054,4055,4058],{},[43,4052,4053],{},"Step 3: Add a terminal value."," A company does not stop existing in year ten. To capture everything after the explicit forecast, you add a ",[2230,4056,4057],{},"terminal value",", an estimate of all the cash beyond the projection, assuming a slow, steady long-term growth rate. This is then discounted back like everything else.",[29,4060,4061],{},"Add up all the discounted yearly cash flows and the discounted terminal value, and you have the estimated intrinsic value of the whole business today.",[36,4063,4065],{"id":4064},"why-a-single-dcf-is-dangerously-fragile","Why a single DCF is dangerously fragile",[29,4067,4068],{},"Here is the catch that humbles everyone who builds one: a DCF is exquisitely sensitive to its inputs. Nudge the growth rate up by two points and lower the discount rate by one, and the \"fair value\" can swing 40% or more. The model produces a precise-looking number, but that precision is an illusion, it inherits all the uncertainty of the assumptions you fed it. Garbage in, confidently-formatted garbage out.",[29,4070,4071],{},"This is why a DCF in the wrong hands becomes a way to justify a conclusion you already had. Want the stock to look cheap? Bump the growth rate. Want it to look expensive? Raise the discount rate. The math is honest, but it will faithfully launder a biased assumption into an authoritative-looking target.",[36,4073,4075],{"id":4074},"why-we-use-a-sector-aware-ensemble-not-one-model","Why we use a sector-aware ensemble, not one model",[29,4077,4078,4079,4082],{},"The defence against that fragility is not a better single model, it is refusing to rely on any single model, and refusing to force every company through the same handful of numbers regardless of what kind of business it actually is. Ploutos AI estimates fair value with an ",[43,4080,4081],{},"ensemble of methods that adapts to the company",", each with different blind spots:",[81,4084,4085,4091,4097,4103,4113,4119],{},[84,4086,4087,4090],{},[43,4088,4089],{},"DCF",", the growth-aware estimate described above, for businesses with real free cash flow to project.",[84,4092,4093,4096],{},[43,4094,4095],{},"Growth DCF",", for young, fast-growing companies that aren't profitable yet: it projects revenue instead of cash flow and models the profit margin maturing gradually over a longer horizon, instead of refusing to value the company at all.",[84,4098,4099,4102],{},[43,4100,4101],{},"Sector multiples",", what comparable companies trade at on earnings, sales, and operating profit, adjusted for quality.",[84,4104,4105,4108,4109,4112],{},[43,4106,4107],{},"Earnings Power Value (EPV)",", a conservative anchor that assumes ",[2230,4110,4111],{},"zero"," growth and asks what the business is worth purely on its current earning power. If a DCF says a company is cheap but EPV strongly disagrees, that tension is itself a signal.",[84,4114,4115,4118],{},[43,4116,4117],{},"The Graham Number",", a classic book-value-and-earnings benchmark from Benjamin Graham.",[84,4120,4121,4124],{},[43,4122,4123],{},"Sector specialists",", for the businesses a generic ensemble handles badly: an H-model dividend discount for regulated utilities with a stable, policy-driven payout, a net-asset-value calculation for companies whose balance sheet is dominated by digital-asset holdings, and a conservative net-cash floor for pre-revenue biotech, where a full valuation isn't possible without clinical-trial data.",[29,4126,4127,4128,4131,4132,4135,4136,4138],{},"These are blended with weights that depend on the sector and the company's own financial profile, because book value matters more for a bank than for a software company, and a pre-profit growth story needs a different lens than a mature dividend payer, and the result comes with a ",[2230,4129,4130],{},"confidence level"," and a fair-value ",[2230,4133,4134],{},"range",", not a single false-precision point. When the applicable methods agree, confidence is high. When they scatter widely, that disagreement is reported rather than hidden. Quality also bends the inputs: a business with a durable ",[48,4137,1393],{"href":1392}," earns a lower discount rate and a richer multiple, because its advantage is more likely to persist.",[36,4140,4142],{"id":4141},"why-analyst-price-targets-are-often-unreliable","Why analyst price targets are often unreliable",[29,4144,4145],{},"Sell-side analyst price targets are familiar, widely quoted, and structurally compromised. They are produced inside institutions with business relationships to the companies they cover, they cluster tightly around each other because being wrong alone is more career-threatening than being wrong together, and they usually carry a short, twelve-month horizon that has more to do with momentum than with intrinsic worth. The result is a number that often tracks the recent stock price rather than leading it.",[29,4147,4148],{},"A transparent, assumption-explicit valuation has the opposite character. You can see every input, you can change the ones you disagree with, and the conservative anchors keep optimism in check. It will not be right every time, no valuation method is, but it is honest about its uncertainty, which is more than a single confident price target ever offers.",[36,4150,4152],{"id":4151},"what-a-fair-value-is-and-is-not","What a fair value is, and is not",[29,4154,4155,4156,4158],{},"A fair-value estimate is a reference point, not a prophecy. It says \"based on these assumptions, here is roughly what the business is worth today, and here is how far the market price sits from that.\" The distance between price and value is the ",[43,4157,1425],{},", the cushion that protects you when, not if, some of your assumptions turn out to be wrong.",[29,4160,4161,4162,4164,4165,52],{},"It does not tell you what the stock will do next quarter, and it is not a target you should expect the price to obediently reach. It is one disciplined input into your own judgement. If you want to see the sector-aware valuation run on a real company alongside the rest of the research, you can ",[48,4163,1237],{"href":1236},", or read how it fits into the ",[48,4166,2725],{"href":238},{"title":438,"searchDepth":439,"depth":439,"links":4168},[4169,4170,4171,4172,4173,4174],{"id":4013,"depth":442,"text":4014},{"id":4031,"depth":442,"text":4032},{"id":4064,"depth":442,"text":4065},{"id":4074,"depth":442,"text":4075},{"id":4141,"depth":442,"text":4142},{"id":4151,"depth":442,"text":4152},"Discounted cash flow, explained simply: valuing a company by its future cash, why a single DCF is fragile, and why analyst price targets are often unreliable.",[4177,4180,4183,4186],{"q":4178,"a":4179},"What is a DCF valuation?","DCF stands for Discounted Cash Flow: you estimate a company's future cash flows and discount them back to today to derive a fair value per share.",{"q":4181,"a":4182},"What is the discount rate (WACC)?","The rate that reflects the company's risk and cost of capital. Small changes in it move the valuation dramatically, which makes it one of the most sensitive assumptions.",{"q":4184,"a":4185},"Why do analyst price targets so often miss?","Because they depend on assumptions (growth, margins, discount rate, terminal value) that may not play out. A small change in an assumption moves the target a lot.",{"q":4187,"a":4188},"Is a DCF reliable?","It's a useful framework, not an exact number. It's only as good as its assumptions, which is why it's always paired with a margin of safety.",{},"What is DCF valuation?",{"title":3990,"description":4175},"blog\u002Fen\u002Fwhat-is-dcf-valuation",[2758,1295],"what-is-dcf-valuation","scyXeQ-ggAEXvIHqkd4jD78Gg9nlFr4JZ3M9TB4KZ9g",{"id":4197,"title":4198,"author":6,"body":4199,"date":2880,"description":4370,"extension":459,"faq":4371,"featured":473,"mentionsTickers":473,"meta":4384,"metaTitle":4385,"navigation":476,"ogImage":4208,"path":50,"readMinutes":1290,"seo":4386,"stem":4387,"tags":4388,"translationKey":4389,"__hash__":4390},"blogEn\u002Fblog\u002Fen\u002Fwhy-ai-stock-tools-hallucinate.md","Why AI stock tools hallucinate, and how we stop it",{"type":8,"value":4200,"toc":4362},[4201,4209,4212,4215,4219,4222,4249,4252,4256,4263,4274,4277,4291,4295,4302,4308,4314,4318,4325,4332,4335,4339,4342,4345,4347,4353],[11,4202,4204],{"className":4203},[14,15,16,17,18,19],[21,4205],{"alt":4206,"className":4207,"src":4208},"An ungrounded language-model answer next to a data-grounded pipeline",[25,26],"\u002Fblog\u002Fog-hallucinate.svg",[29,4210,4211],{},"Paste a ticker into a general-purpose chatbot and ask for an analysis, and you will get something that reads beautifully: a tidy P\u002FE ratio, a confident note about insiders buying last quarter, a clean paragraph on how the sector is rotating. The prose is fluent, the structure is professional, the tone is certain.",[29,4213,4214],{},"A lot of it may also be false. Not because the model is broken, but because it is doing exactly what it was built to do, and that job is not \"report facts.\"",[36,4216,4218],{"id":4217},"what-a-hallucination-looks-like-in-stock-analysis","What a hallucination looks like in stock analysis",[29,4220,4221],{},"In AI, a \"hallucination\" is a confident, fluent, plausible statement that simply is not true. In finance the failure is especially dangerous, because the output is full of exactly the specifics that make it look trustworthy:",[81,4223,4224,4230,4237,4243],{},[84,4225,1978,4226,4229],{},[43,4227,4228],{},"valuation multiple"," that is close to reality but quietly wrong, last year's P\u002FE, or a forward figure presented as trailing.",[84,4231,4232,4233,4236],{},"An ",[43,4234,4235],{},"insider transaction"," that never happened, or a real one with the direction flipped.",[84,4238,1978,4239,4242],{},[43,4240,4241],{},"catalyst"," lifted from an article that is two years old, described as if it were this week.",[84,4244,1978,4245,4248],{},[43,4246,4247],{},"cited source"," that does not exist, a report, a filing, an analyst note, fabricated wholesale because the sentence needed a citation to sound complete.",[29,4250,4251],{},"The unifying problem is that none of these are flagged as guesses. They sit in the same confident paragraph as the genuinely correct statements, and nothing in the text tells you which is which.",[36,4253,4255],{"id":4254},"why-it-happens-the-model-predicts-words-not-facts","Why it happens: the model predicts words, not facts",[29,4257,4258,4259,4262],{},"A large language model is, at its core, a very sophisticated next-word predictor. It has read an enormous amount of text and learned what ",[2230,4260,4261],{},"tends to come next",". When you ask it for a company's free cash flow, it does not look up the number. It generates the most statistically plausible continuation of your question, and a specific-looking number is more plausible than \"I am not sure.\"",[29,4264,4265,4266,4269,4270,4273],{},"That is the whole trap. The model is optimised to be ",[2230,4267,4268],{},"fluent",", and fluency rewards confident specifics. \"Its return on capital is around 14%\" reads better than \"I would need to check.\" So when the real figure is not reliably encoded in its training, the model does not stop, it produces a number that ",[2230,4271,4272],{},"fits the shape"," of an answer. For casual writing that is fine. For an investment decision it is a landmine.",[29,4275,4276],{},"Two structural weaknesses make it worse:",[1150,4278,4279,4285],{},[84,4280,4281,4284],{},[43,4282,4283],{},"Training data is frozen and fuzzy."," A model trained months ago has no idea what a company reported last week, and even older figures are blended across everything it ever read, not stored as a clean ledger.",[84,4286,4287,4290],{},[43,4288,4289],{},"The model wants to agree with itself."," Once it has written \"the bull case is strong\" in the opening, the rest of the answer tends to confirm that, not challenge it. This is confirmation bias, baked in.",[36,4292,4294],{"id":4293},"the-fix-part-one-do-not-ask-the-model-to-remember-make-it-fetch","The fix, part one: do not ask the model to remember, make it fetch",[29,4296,4297,4298,4301],{},"The single most important design decision in Ploutos AI is that ",[43,4299,4300],{},"the language model is never the source of a number."," Every figure that enters an analysis is fetched live, at the moment you run it, from a real source: fundamentals and prices, filings from SEC EDGAR for insider transactions and material events, news and sentiment feeds, and so on.",[29,4303,4304,4305,52],{},"The model's job is reframed from \"recall the facts\" to \"reason over facts I am handing you right now.\" That is a job language models are genuinely good at, weighing a return-on-capital figure against a sector average, noticing that free cash flow diverges from reported earnings, connecting an 8-K filing to a stated risk. The numbers are not its opinion, they are inputs it is not allowed to invent. And when the data simply is not there, we ",[48,4306,4307],{"href":230},"stop rather than fill the gap with a guess",[29,4309,4310,4311,52],{},"This is also why the work is split into stages rather than one giant prompt. Each stage has a narrow job grounded in specific data, which leaves far less room for the model to drift into invention. The full sequence is described in our ",[48,4312,4313],{"href":238},"walkthrough of the analysis pipeline",[36,4315,4317],{"id":4316},"the-fix-part-two-make-the-model-argue-against-itself","The fix, part two: make the model argue against itself",[29,4319,4320,4321,4324],{},"Grounding kills invented ",[2230,4322,4323],{},"facts",". It does not, on its own, kill the second problem, the model talking itself into its own conclusion. For that we add a deliberately adversarial step.",[29,4326,4327,4328,4331],{},"After a verdict is formed, a separate and more capable model pass receives the picks with a single instruction: ",[43,4329,4330],{},"find what we got wrong."," It is told to behave like a hostile short-seller. For each idea it has to produce the weakest assumption in the thesis, a risk the first pass did not flag, a concrete bear case, and the specific observable event that would prove the thesis wrong.",[29,4333,4334],{},"This matters most exactly where confirmation bias is most dangerous: on the ideas that scored well. A tool that only ever tells you why an idea is good is not doing research, it is doing marketing. Forcing a structured rebuttal is the antidote.",[36,4336,4338],{"id":4337},"why-just-cite-your-sources-is-not-enough","Why \"just cite your sources\" is not enough",[29,4340,4341],{},"A common half-measure is to ask the model to cite sources. It helps with appearances and almost nothing else, because a model that will invent a P\u002FE will just as happily invent the citation next to it. A fabricated footnote is not a safeguard, it is a second hallucination wearing a suit.",[29,4343,4344],{},"The only reliable fix is architectural: the facts must come from outside the model and be verifiable, and the reasoning must be stress-tested by something whose job is to disagree. Citations are a presentation layer. Grounding is a plumbing layer. They are not substitutes.",[36,4346,1692],{"id":1691},[29,4348,4349,4350],{},"When you read an AI stock analysis, the right question is not \"does this sound smart?\" Fluency is free, and it is exactly what a hallucination is made of. The right questions are: ",[2230,4351,4352],{},"where did each number come from, and what tried to prove this wrong?",[29,4354,4355,4356,4358,4359,52],{},"That is the bar we hold ourselves to. Numbers are fetched, not remembered. Conclusions are challenged, not just stated. And when the data is too thin to do either honestly, we say so. If you want to see the grounded pipeline produce a full analysis, you can ",[48,4357,2518],{"href":1236},", or read how the ",[48,4360,4361],{"href":238},"five stages fit together",{"title":438,"searchDepth":439,"depth":439,"links":4363},[4364,4365,4366,4367,4368,4369],{"id":4217,"depth":442,"text":4218},{"id":4254,"depth":442,"text":4255},{"id":4293,"depth":442,"text":4294},{"id":4316,"depth":442,"text":4317},{"id":4337,"depth":442,"text":4338},{"id":1691,"depth":442,"text":1692},"A language model will happily invent a P\u002FE ratio or an insider trade that never happened. Here is why it happens, and how Ploutos AI prevents it.",[4372,4375,4378,4381],{"q":4373,"a":4374},"Why do AI stock tools invent figures?","A language model predicts the most likely next word, not the fact. When it doesn't reliably 'remember' a number, it produces one that looks right instead of saying 'I don't know'.",{"q":4376,"a":4377},"How is it prevented?","Numbers are fetched live from the official filings (grounding), not from the model's memory, and a separate pass challenges the conclusion.",{"q":4379,"a":4380},"Isn't it enough to 'cite sources'?","No. A model that will invent a P\u002FE will just as happily invent the citation next to it. Grounding has to be architectural, not cosmetic.",{"q":4382,"a":4383},"How do I know if an AI analysis is trustworthy?","Ask: where did each number come from, and what tried to prove it wrong? If the numbers have no verifiable source, treat them with caution.",{},"Why AI stock tools hallucinate",{"title":4198,"description":4370},"blog\u002Fen\u002Fwhy-ai-stock-tools-hallucinate",[482,483],"why-ai-stock-tools-hallucinate","gfpUEMXnRkfS_7JK5TaGceoGhvPxMKEl2tyh2dNoFxk",{"id":4392,"title":4393,"author":6,"body":4394,"date":2880,"description":4545,"extension":459,"faq":4546,"featured":473,"mentionsTickers":473,"meta":4556,"metaTitle":4557,"navigation":476,"ogImage":4403,"path":230,"readMinutes":2898,"seo":4558,"stem":4559,"tags":4560,"translationKey":4561,"__hash__":4562},"blogEn\u002Fblog\u002Fen\u002Fwhy-we-refuse-to-analyze-a-stock.md","Why we sometimes refuse to analyze a stock (and tell you instead)",{"type":8,"value":4395,"toc":4537},[4396,4404,4407,4410,4414,4417,4424,4431,4438,4442,4448,4454,4460,4464,4467,4474,4481,4496,4500,4507,4510,4514,4517,4521,4528],[11,4397,4399],{"className":4398},[14,15,16,17,18,19],[21,4400],{"alt":4401,"className":4402,"src":4403},"Three data-coverage tiers: full, partial, and insufficient",[25,26],"\u002Fblog\u002Fog-data-coverage.svg",[29,4405,4406],{},"There is a failure mode that almost every \"AI stock analyzer\" shares, and almost none of them admit to: when the underlying data is thin, they answer anyway.",[29,4408,4409],{},"Ask one of these tools about a tiny micro-cap, a freshly-listed company, or a foreign listing with sparse filings, and you will still get a confident verdict, a fair value, a score out of ten. It looks exactly like the verdict you would get for a household-name large cap. The difference is that one is built on real fundamentals and the other is built on blanks. You cannot tell which from the output, and that is the problem.",[36,4411,4413],{"id":4412},"a-low-score-and-no-data-look-identical-but-they-are-not","A low score and \"no data\" look identical, but they are not",[29,4415,4416],{},"Our quality score rates a company against ten value-investing criteria: valuation versus its sector, revenue growth, gross margin, return on invested capital, free cash flow, balance-sheet strength, and so on. Each criterion either passes or fails, and the failures drag the score down.",[29,4418,4419,4420,4423],{},"Here is the subtle trap. If a company genuinely has a weak return on invested capital, that criterion fails and the score drops. But if we simply ",[2230,4421,4422],{},"do not have"," the return-on-invested-capital figure, the most naive thing a scoring system can do is treat the missing value as a fail and drop the score in exactly the same way.",[29,4425,4426,4427,4430],{},"The result is a stock that ",[2230,4428,4429],{},"looks"," like a mediocre business when the honest description is \"we do not have enough information to judge this business at all.\" A genuinely bad company and a company we know nothing about end up with the same low number. For a tool whose entire job is to help you tell good businesses from bad ones, that is not a small bug. It is a credibility problem.",[29,4432,4433,4434,4437],{},"So before anything else runs, we measure how much of the data we would need is actually present. We call this ",[43,4435,4436],{},"data coverage",", and it sorts every ticker you submit into one of three tiers.",[36,4439,4441],{"id":4440},"the-three-tiers-of-data-coverage","The three tiers of data coverage",[29,4443,4444,4447],{},[43,4445,4446],{},"Full."," We have the company's price, its share count, and most of the core fundamentals a valuation needs: earnings, margins, return on capital, cash flow, revenue, and a usable balance sheet. This is the normal case for established, well-covered companies. We proceed silently, exactly as you would expect.",[29,4449,4450,4453],{},[43,4451,4452],{},"Partial."," The essentials are there, but several core fundamentals are missing or the company is not yet in a state where a valuation model can produce a meaningful number, for example a business that is not yet profitable and has no positive free cash flow to discount. We can still run the analysis, but the verdict will rest on less. So we tell you that, show you the trade-offs, and let you decide whether to spend an analysis on it.",[29,4455,4456,4459],{},[43,4457,4458],{},"Insufficient."," The data is too thin to support a fundamentals verdict at all: no usable price or share count, a feed error, or only a handful of the core metrics present. This is the tier where most tools would quietly produce a number anyway. We do not. We stop and tell you what we found.",[36,4461,4463],{"id":4462},"what-enough-data-actually-means","What \"enough data\" actually means",[29,4465,4466],{},"Coverage is not a vague feeling, it is a count. We look for two things.",[29,4468,4469,4470,4473],{},"First, the ",[43,4471,4472],{},"critical inputs",": a current price and a share count. Without these, almost nothing downstream works, you cannot compute a per-share value or a margin of safety, so a verdict would be meaningless.",[29,4475,4476,4477,4480],{},"Second, the ",[43,4478,4479],{},"core fundamentals",", the eight figures that actually drive a value verdict: trailing earnings, gross margin, return on invested capital, free cash flow, revenue and its growth, the balance sheet, and the price-to-earnings ratio. The more of these are missing, the less any verdict can be trusted. We also check whether at least one valuation method can run at all, which needs either positive earnings or positive free cash flow to work with.",[29,4482,4483,4484,4487,4488,4491,4492,4495],{},"Importantly, this is a measurement of ",[2230,4485,4486],{},"presence",", not a judgement of ",[2230,4489,4490],{},"quality",". A company can have full coverage and still score poorly, that is a real, useful answer. What coverage protects against is the opposite case: a confident answer built on absence. If you want to see what the full pipeline does once the data clears this bar, we wrote a ",[48,4493,4494],{"href":238},"walkthrough of how Ploutos AI analyzes a stock"," end to end.",[36,4497,4499],{"id":4498},"why-we-check-this-before-charging-you-anything","Why we check this before charging you anything",[29,4501,4502,4503,4506],{},"The coverage check runs ",[2230,4504,4505],{},"before"," an analysis is counted against your plan, and that ordering is deliberate. If a ticker is too thin to analyze honestly, we would rather spend zero of your searches on it than hand you a hollow report and quietly tick the counter down.",[29,4508,4509],{},"There is a cost on our side too, every analysis runs a multi-stage pipeline that calls a language model and a stack of data services, and burning that on a ticker we cannot do justice to is wasteful for everyone. But the user-facing reason is the one that matters: a search you spend should buy you something worth having.",[36,4511,4513],{"id":4512},"what-happens-when-a-ticker-is-flagged","What happens when a ticker is flagged",[29,4515,4516],{},"When coverage comes back partial or insufficient, you get a short, honest prompt instead of a redirect into a misleading report. It tells you what we actually have, weighs the upside of proceeding against the downside, and gives you three choices: run it anyway with your eyes open, drop the thin tickers and analyze only the ones with enough data, or step back and pick a different name. If you do proceed on partial data, the resulting analysis carries that warning forward, so the verdict is never dressed up as more certain than it is.",[36,4518,4520],{"id":4519},"why-this-is-a-feature-not-a-limitation","Why this is a feature, not a limitation",[29,4522,4523,4524,4527],{},"It is tempting to think that a tool which always has an answer is more powerful than one that sometimes says \"not enough to go on.\" The opposite is true. The willingness to say ",[2230,4525,4526],{},"I do not know"," is exactly what separates a research process from a guessing machine. Value investing runs on the same principle, Warren Buffett's \"circle of competence\" is just a disciplined way of refusing to act where you lack the information to act well.",[29,4529,4530,4531,1479,4534,4536],{},"A verdict you can trust on the companies where the data is real is worth far more than a verdict on everything, because the second kind teaches you to ignore the warning labels. When Ploutos AI does give you a full analysis, you can know that it cleared this bar first. When it declines, that is information too. You can read about what we do with a full, clean dataset in our ",[48,4532,4533],{"href":238},"breakdown of how the analysis pipeline works",[48,4535,1237],{"href":1236}," and see the coverage check in action.",{"title":438,"searchDepth":439,"depth":439,"links":4538},[4539,4540,4541,4542,4543,4544],{"id":4412,"depth":442,"text":4413},{"id":4440,"depth":442,"text":4441},{"id":4462,"depth":442,"text":4463},{"id":4498,"depth":442,"text":4499},{"id":4512,"depth":442,"text":4513},{"id":4519,"depth":442,"text":4520},"Incomplete data produces a confident but misleading verdict. Here is how Ploutos AI checks data coverage before spending your analysis, and when it stops.",[4547,4550,4553],{"q":4548,"a":4549},"Why does it sometimes refuse to analyze a stock?","When the available data is too thin for a reliable analysis, we stop rather than fill the gap with guesswork.",{"q":4551,"a":4552},"What counts as 'thin data'?","For example incomplete filings, a very recent IPO, or a ticker outside our coverage.",{"q":4554,"a":4555},"Isn't it better to always give an answer?","No. A confident answer built on missing data is worse and more dangerous than an honest 'there isn't enough information'.",{},"Why we refuse to analyze a stock",{"title":4393,"description":4545},"blog\u002Fen\u002Fwhy-we-refuse-to-analyze-a-stock",[482,483],"why-we-refuse-to-analyze-a-stock","99t2CtEm0z1JEnPSsy7nOLo62yFNj16WXIFuaUadlRU"]