Financial data extraction into the analyst's own model, the day the company files, with every number linked to its page
Every covered name's model refreshed the day it files, every number tied to its page.
See one case, screen by screen ↓Financial data extraction, also called financial statement data extraction, is pulling reported numbers (revenue, segment results, KPIs, cash flow, debt) out of filings such as the 10-Q, 10-K, earnings releases and decks and into a model; KPI extraction is the same work for operating metrics such as orders, backlog and book-to-bill. Financial Data Extraction does it into the analyst's own Excel model: seven AI agents read every page, map each figure to the analyst's own line names, foot every subtotal and link each cell to its page. A research associate accepts each new column, a different senior analyst checks it and the analyst publishes.
AI in asset management: a buyer's guide to agentic AI for investment managers →
Updated
When a covered company files, a research associate opens the 10-Q, the press release and the earnings deck and types thirty or more numbers into the analyst's own Excel model, under the analyst's own line names. Then come the checks: that segments add up to consolidated revenue, that thousands did not become millions, that year-to-date cash flow was not entered as the quarter, that last year's figures were not quietly recast. At Quillfield Capital this season that is 2,036 expected filings for 1,186 covered issuers, most of them inside a few weeks, after 4 PM and before the morning meeting.
Data vendors and their financial data extraction tools cover the largest listed names, but in their own standardized line items, and they stop at small caps, private and 144A issuers and local-language filers — 214 of Quillfield's 1,186 covered issuers are in no vendor dataset, so their numbers are typed by hand. The errors are silent: a corporate and eliminations line that was never in the model, a relabeled KPI that breaks a series, recast history overwritten. And when a portfolio manager asks where a number came from, someone reopens the PDF.
Estimated split of about two hours by hand per ticker per quarter, and about eight minutes of review with the solution.
Seven agents read, map, foot and check every filing into the analyst's own model; an associate accepts, a senior analyst checks, the analyst publishes.
Corvane Industrial Corp., a large accelerated filer with three segments — Flow Control, Motion Systems and Aerospace Components — files its Q3 2026 10-Q at 4:12 PM on Tuesday, October 27. Julian Marsh covers it for Quillfield Capital, and his morning meeting is at 9:30. Here is how the people work with the agents, screen by screen, in the working solution.
The Filing Watcher picks up the 10-Q from EDGAR, matches CIK and ISIN to Corvane and links the 4:05 PM release (8-K Exhibit 99.1) and the deck. The Table Reader reads 84 pages and 63 tables. Units & Periods turns thousands into millions and the deck's “$2.31 billion” into 2,310.0, and derives the quarter's cash flow from nine months: 301.7 − 182.8 = 118.9. The Mapper fills 31 of Julian's line names, Footing runs 46 checks, Restatement compares history with v13, and the Model Writer cross-checks it against Julian's estimates and writes the column for the associate.
“Seven agents, 40 seconds, every step on the record. They propose; people decide.”
Nine of Julian's 46 names filed since yesterday; eight models are refreshed and Bramwell Filtration is being read now. Eight decisions wait for Leah, the nearest due at 9:30 AM. Overnight findings each carry their page — Corvane's $1.7M segment gap on p.22, Altmark's two units on p.4 and p.9, Kisaragi's rounded English summary. She asks the agents what could miss its 9:30 deadline, and the Filing Watcher answers: Corvane and Altmark, escalated to Ines Carvalho at 9:00 AM if not accepted.
Filing Watcher: “Two models: Corvane Industrial (3 decisions) and Altmark Antriebstechnik (1 mapping). If they are not accepted by 9:00 AM I escalate to Ines Carvalho, per the SLA rule.”
The Q3 2026 column sits in Julian's own workbook, CVNI_JM_Model.xlsx, beside four quarters of history, his estimate and the difference. Revenue is $814.1M, 1.1% above his $805.0M; Adj. EBITDA ex-SBC is 177.0 against 171.5. A colored corner on each cell says whether it footed, needs her, was derived or converted, or is restated history. Beside the grid, the agents set out her next step.
“Two decisions need you before 9:30 AM, then you accept the column. Start with the footing: it changes total revenue.”
She clicks Adj. EBITDA ex-SBC. The lineage shows the filing (8-K Exhibit 99.1, filed 4:05 PM), page 9 with the reconciliation of non-GAAP measures, the table row printed 177.0, no conversion, and the mapping rule learned from 214 of Julian's decisions. The check ties 136.9 + 27.1 + 9.8 + 3.2 = 177.0. The vendor dataset shows 173.8 because it excludes restructuring — Julian's definition is kept.
Lineage: filing → page → table row → units and period → mapping → checks → cell.
The three segments add to $812.4M (318.6 + 276.9 + 216.9); consolidated net sales are $814.1M. The Footing agent searched all 48 pages and found “Corporate, other and eliminations” of $1,711 thousand in Note 14, “Segment information”, on page 22 — a line Julian's model never had. Q3 2025 to Q2 2026 were short by 1.2, 1.9, 1.4 and 1.6 for the same reason. It proposes a new row, back-filled for four quarters.
Footing: “I found the $1.7M … I will not plug it.” One click opens page 22 with the line highlighted.
Leah adds “Corp & elims rev” and total revenue foots in all five columns. Then the Mapper's question, at 0.91: the deck's “Aftermarket revenue” is now “Aftermarket & services revenue” of $289.0M, including $11.2M of field services previously in equipment revenue (deck p.6, footnote 2). Taking $289.0M as reported would break Julian's series. She keeps it comparable at 277.8 with a new “Field services rev” row, then accepts the column: 33 values, 46 of 46 checks footing, her name recorded as the maker.
The two choices on screen: “Keep comparable · 277.8, plus a new row ‘Field services rev’ 11.2” or “Take as reported · 289.0, mark a series break”.
Altmark Antriebstechnik's Quartalsmitteilung reports in Tsd. € on page 4 and Mio. € on page 9, with comma decimals; both are converted and revenue ties at €412.4M. “EBIT bereinigt” is a new label: its definition on page 11 excludes restructuring and purchase price allocation, matching Julian's Adj. EBIT at €42.6M. At 0.89, below the 0.92 threshold, the Mapper asks instead of mapping. For Kisaragi Precision's kessan tanshin, Q2 is derived from the half: ¥48,612M − ¥23,904M = ¥24,708M.
“Kisaragi's English summary rounds revenue to ¥48.6 billion; the agents used the Japanese original, ¥48,612M.”
Julian's names show eight exceptions — the three Corvane decisions now done, five still open — and 143 are open across all teams, by type: unmapped lines, failed footings, restatements, units and currency, low scan confidence, period derivations. Brightwater Pumps, a small cap outside vendor datasets, shows a new $1,384K tariff surcharge line the Mapper proposes to keep in revenue with a memo row. Norland Aerostructures' scanned 144A noteholder report read at 0.83, 0.61 and 0.82 — held for a person with the crops.
Table Reader: “I never guess a digit: type what you see and the Footing agent re-checks it against the covenant table on p.9.”
Precision Bearings moved from Motion Systems to Flow Control, and the 8-K of October 6 recast four quarters: Q3 2025 Flow Control revenue goes from 287.4 to 300.2 and Motion Systems from 262.0 to 249.2. Ines confirms the 16 cells match Exhibit 99.2, that the new row is the reconciling line on p.22 and not a plug, that the Aftermarket decision keeps the series comparable, and that there were no overrides. Leah could not have checked her own column.
“You accepted this column, so you cannot check it. A different person checks: Ines Carvalho (senior analyst).”
Before Julian can publish, the restricted list is checked (Corvane is clear in the compliance control room at 8:40 AM), with entitlements and maker-checker: Leah made it, Ines checked it, Julian publishes. v14 goes to the research management system and his workbook, and the portfolio managers who follow Corvane are notified. Every step, from the 4:12 PM pickup to the publish, is on the audit trail.
“Publishing writes v14 to the research management system and the workbook. Portfolio managers subscribed to Corvane are notified.”
The same 10-Q, seen by the five people who carry it — what earnings season looked like, and what it looks like now.
Seven agents read, map, foot and check every filing into the analyst's own model; an associate accepts, a senior analyst checks, the analyst publishes.
Watches EDGAR, exchange feeds, IR sites, bondholder sites and data rooms for every covered issuer; links each document to the issuer, the analyst's model and the earlier release, and starts the clock to the model's due time.
Reads every page and table in the original language, scans included, and records each value with its page, table, row, column and printed units.
Maps each issuer's label to the analyst's own line name, learning per analyst from the workbook and every decision people make.
Handles thousands against millions, currency at the house rate, sign conventions, year-to-date to quarter and each issuer's fiscal calendar.
Adds every subtotal — segments to consolidated, income statement, balance sheet, cash flow, each non-GAAP reconciliation to GAAP — and when something does not foot, searches the filing for the reconciling line.
Compares every prior period a filing re-reports against the published model, detects recasts and restatements, and proposes the update with the old value kept.
Cross-checks against the analyst's estimates and a vendor dataset where one exists, writes the accepted column into the workbook with a link and a note on each cell, and prepares the version diff and the evidence.
Accept and check before the analyst publishes. The agents propose; a named person decides.
Ask in plain words, or tell the agents what to do. Every answer cites its page, and anything that changes the model still needs a person.
What did the agents find in Corvane?
31 values mapped in 40 s; 45 of 46 checks foot. Segments add to $812.4M against $814.1M — the $1.7M is “Corporate, other and eliminations” on p.22. Aftermarket's definition widened (deck p.6), and four quarters were recast (8-K).
Treat restructuring as recurring for Corvane
Understood. For Corvane, “Adj. EBITDA ex-SBC” would stop adding back restructuring: Q3 2026 becomes 173.8 (177.0 − 3.2) and the four prior quarters change too. This changes Julian's definition, so it needs his approval.
Why not use Kisaragi's English summary?
The English summary rounds to ¥0.1 billion (“¥48.6 billion”), which would move Q2 revenue by up to ¥50M. The Japanese original gives ¥48,612M on p.1.
Is Corvane on the restricted list?
No. The compliance control room shows Corvane Industrial clear as of 8:40 AM. It is checked again at the moment Julian publishes; if it were restricted, publish would stop and Research Compliance would be told.
13 screens from the working solution, on its sample data. Pick one to see it large.
Thresholds, unit rules and service levels are settings. Each change is proposed by one administrator, approved by a second and versioned.
| Setting | Default | Choose from |
|---|---|---|
| Mapping asks a person below | 0.92 | Slider from 0.85 to 0.99, showing how many cells a day would come to people before saving |
| Scan reads go to a person below | 0.85 | House default |
| Suggested for column accept at | ≥ 0.97 and footed | A person still accepts every column |
| Footing tolerance | $0.1M or 0.05%, whichever is smaller | Changed by the Head of Research Data, approved by a second administrator |
| Highlight the difference from the analyst's estimate above | 3% | Set per analyst |
| Year-to-date to quarter | On | Derive the quarter from YTD less the prior YTD and show the arithmetic · on or off |
| Currency | House rate | Market data feed, 4:00 PM New York close on the period end, labeled with the rate and date |
| Model refreshed by | 9:30 AM local for filings after 4:00 PM | Escalates to the senior analyst 30 minutes before; a daily digest at 7:30 AM |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample data · “target” = the design goal, measured in the live solution · “estimated” = our estimate (where a ticker's update hours go, by hand and with the solution) · “typical” = published figures (SEC Form 10-Q general instructions — 40 days after quarter end for large accelerated filers, 45 for non-accelerated filers; a source-linked data vendor's published coverage of about 3,000–6,000 listed companies and its claimed 15–45 minutes saved per model, vendor, directional). Quillfield Capital and the people, issuers, identifiers and figures named on this page are fictional — characters and sample data in the working solution.
Financial data extraction, also called financial statement data extraction, is pulling reported numbers (revenue, segment results, KPIs, cash flow, debt) out of filings such as the 10-Q, 10-K, earnings releases and decks and into a model; KPI extraction is the same work for operating metrics such as orders, backlog and book-to-bill. Financial Data Extraction does it into the analyst's own Excel model: seven AI agents read every page, map each figure to the analyst's own line names, foot every subtotal and link each cell to its page. A research associate accepts each new column, a different senior analyst checks it and the analyst publishes.
Yes. The Table Reader reads every page and table in the original language — German, Japanese, Spanish, Korean, French and more — and records each value with its page, row and printed units. Units & Periods converts Tsd. €, Mio. € or ¥ millions and derives quarters from year-to-date figures, showing the arithmetic. Scanned values read below 0.85 confidence go to a person with the crop; the agents never guess a digit.
Your own. This is automated financial modeling for the model each analyst already keeps: the Mapper learns each analyst's line names from the workbook and every decision, and maps each issuer's as-reported financials to them; when a label changes meaning, it asks instead of breaking the series. A vendor dataset is used only to cross-check, never as the source, and the analyst's formula cells are never overwritten.
The Footing agent adds every subtotal and, when segments do not reach consolidated revenue, searches the filing for the reconciling line and proposes a row — it never plugs a difference. The Restatement agent compares re-reported periods with the published model and proposes the recast with the old values kept in each cell's history. Both always go to a senior analyst at check.
Yes. The agents watch exchange feeds, IR sites, bondholder sites and data rooms for every covered issuer, and an analyst can add an issuer and its sources. In the working solution, 214 of 1,186 covered issuers are outside vendor datasets, from a small-cap 10-Q filer to a private 144A issuer's scanned noteholder report.
They meet in the model. Financial Data Extraction does not build a new model from a template: its financial modeling automation keeps the analyst's existing one current — their line names, their adjusted definitions, their estimates. Each new quarter is mapped, footed, checked against restated history and linked to its page, and the analyst still owns the estimates and the view.
Yes. The agents propose and never accept, check or publish. A research associate accepts each column, a different senior analyst checks recasts and footing decisions, and the analyst publishes only after the restricted list and entitlements are checked. Every override needs a reason code, and every step is on an audit trail kept seven years.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your analysts' workbooks, line names, coverage list and filing sources — and it goes live once every quality gate has passed. First-time setup connects a workbook, reviews the mapping, picks the coverage, sets the checks and runs a trial on last quarter against what your team typed before anything is written.
Every call read the day it reports, every claim tied to its transcript line.
1-min demoEquity Research AutomationEvery result read, modeled and noted before the open, every figure tied to its page.
1-min demoFinancial Statement TranslationLocal filings in English the day they are published, every figure proven to the original.
1-min demoWe'll run Financial Data Extraction on last quarter's filings for your own covered names, and compare it with the models your team typed.