Build an Investment Thesis from SEC Filings in 60 Minutes
Build an Investment Thesis from SEC Filings in 60 Minutes
You can produce a decision-ready, one-paragraph investment thesis from a 10-K in 30–60 minutes using an AI-assisted filing workflow. The thesis follows a simple structure: one-sentence business summary + two to three quantified drivers with filing citations + primary risk with a falsifiable trigger. To mark it "decision-ready," you need the thesis paragraph, three supporting evidence points, two quantified valuation scenarios, and at least one monitoring trigger tied to a specific filing event. With Filingsiq, that timeline compresses to roughly 30 minutes. A structured 60-minute framework maps each filing section to a discrete thesis element, making the process repeatable across your entire coverage list.
Table of Contents
- What to extract from a 10-K/10-Q in 60 minutes
- How to convert filing evidence into thesis elements, phase by phase
- How do you validate the thesis and avoid weak arguments?
- What accounting red flags should you look for in footnotes?
- How should you use AI for filing analysis without losing research quality?
- How Filingsiq cuts thesis build time to 30 minutes
- Key Takeaways
- A workflow worth embedding in your coverage process
- Filingsiq makes the 30-minute workflow real
- Primary sources and further reading
- FAQ
What to extract from a 10-K/10-Q in 60 minutes
A time-boxed checklist keeps you from reading linearly, which is infeasible for active managers. Diffing against the prior-year filing is the most efficient way to isolate meaningful changes fast.
0–15 min: Business model and key metrics
- One-line business description from Item 1
- Revenue drivers and segment mix; customer concentration above 10%
- Unit economics: gross margin trend, revenue per unit or per customer where disclosed
15–30 min: Competitive position and capital spend
- Barriers to entry signals: capex as % of revenue, R&D as % of revenue, switching-cost language in MD&A
- Material contracts, exclusivity clauses, and key supplier dependencies
30–45 min: Financial quality checks
- Three-year series: revenue, gross margin, operating margin, operating cash flow, capex, free cash flow, total debt
- Debt maturities and covenant language in the credit-agreement footnote
- ROIC calculation from segment notes
45–60 min: Risks, valuation context, and write-up
- Diff risk factors vs. prior year; read every newly added item in full
- Scan critical accounting estimates for changes in revenue recognition, reserves, or tax provisions
- Check 8-K filings for material events since the last annual filing; scan DEF 14A for executive compensation structure and governance flags; review Form 4 filings for insider activity
- Draft the one-paragraph thesis
Pro Tip: Diff the MD&A outlook language word-for-word against the prior year. Subtle tone shifts — from "we expect" to "we anticipate continued pressure" — often precede guidance cuts by one quarter.
How to convert filing evidence into thesis elements, phase by phase
Experienced practitioners recommend reading MD&A and the core financial statements first to build context, then using footnotes for verification. That reading order is the backbone of this workflow.
Phase 1: Business case
Source sections: Item 1 (Business Description), MD&A, segment notes. Extract a one-line business summary and identify the two to three revenue drivers that account for the majority of revenue. Note page numbers for every claim you carry into the thesis.
Phase 2: Competitive position
Source sections: MD&A competitive discussion, capex/R&D footnotes, customer-concentration disclosures. Quantify barriers: capex as % of revenue over three years, R&D as % of revenue, and any disclosed switching costs or contract lengths.
Phase 3: Financial health
Populate the three-year financial template below from the income statement, balance sheet, and cash flow statement. ROIC and FCF conversion are the two ratios that validate capital allocation claims most directly.
Phase 4: Accounting quality
Read critical accounting estimates and the revenue-recognition note. Flag any changes from the prior year. Changes in critical accounting estimates can move reported earnings 10–20% without any operational change.
Phase 5: Risks and catalysts
Diff risk factors. Read every newly added item. Identify the top three risks and quantify each where the filing gives you the data. Note upcoming catalysts: contract renewals, product launches, regulatory decisions disclosed in the filing.
Phase 6: Valuation scenarios and monitoring plan
Write two falsifiable thesis statements and name the single next filing or event that would materially change your view. Assign a monitoring trigger to each.
Pro Tip: If time runs short, prioritize Phases 1, 3, and 5. Business case, financial proof, and quantified risk are the minimum viable thesis. Competitive position and accounting quality are the upgrade pass.
How do you validate the thesis and avoid weak arguments?
A common low-quality thesis error is relying entirely on multiple expansion without earnings growth. Multiple expansion is volatile and harder to predict than earnings-driven returns. Your validation checklist:
- Are the earnings drivers evidence-backed from specific filing disclosures?
- Is margin durability supported by three-year trend data and management commentary?
- Does cash conversion (FCF / net income) confirm reported earnings quality?
- Do capital allocation signals (ROIC trend, buyback vs. debt paydown timing) align with management claims?
- Is there a near-term catalyst with a specific filing or event date?
Use this sensitivity table to stress-test whether the thesis survives multiple compression:
| Scenario | Base multiple | Multiple +10% | Multiple –10% |
|---|---|---|---|
| Base earnings | Target price | Target price | Target price |
| Earnings +15% | Target price | Target price | Target price |
| Earnings –15% | Target price | Target price | Target price |
Populate the target-price cells with your own EV/EBITDA or P/E inputs. If the thesis only works in the "earnings up + multiple up" cell, it is not defensible. A defensible thesis is a structured, evidence-based claim testable against risk factors and relative valuation — not a target price alone.
Common mistakes to avoid:
- Relying solely on multiple expansion with no earnings-growth evidence
- Ignoring changes in critical accounting estimates between quarters
- Underweighting customer-concentration renewal risk disclosed in newly added risk factors
What accounting red flags should you look for in footnotes?
Analysts consistently underweight footnotes. The highest-impact checks are:
- Critical accounting estimates: Any change in revenue-recognition timing, inventory reserves, bad-debt provisions, or tax provisions. A 5-percentage-point shift in a revenue-recognition assumption can swing EPS from a beat to a miss.
- Related-party transactions: Unusual terms, undisclosed conflicts, or transactions that lack arm's-length pricing language.
- Off-balance-sheet commitments: Operating lease obligations pre-ASC 842 adoption, variable-interest entities, or contingent liabilities in the legal-proceedings note.
- Non-recurring adjustments: Large items excluded from non-GAAP metrics; reconcile each to the GAAP income statement.
- Debt and covenant language: Any tightening of financial covenants or new cross-default provisions.
To verify, reconcile every footnote number back to the financial statements. Calculate the EPS and FCF impact of any estimate change. Diff footnote language against the prior year for wording shifts that signal increased legal exposure.
Pro Tip: Follow the footnote reference web. One footnote often cites two or three others. Map the chain before concluding the accounting is clean. This is exactly where AI assistance saves the most time — and where you must still spot-check the source.
Escalate to an accounting specialist when you see material estimate changes, a restatement, or any going-concern language from the auditor.
How should you use AI for filing analysis without losing research quality?
AI should handle the mechanical work. Human judgment handles the conclusions.
- Full-text search and change-diff: Use AI to surface every paragraph added, removed, or materially changed between the current and prior-year filing. This is the highest-leverage starting point.
- Numeric extraction: Prompt the AI to extract the three-year financial series and populate the template above, with page citations for each figure.
- Section summaries with citations: Ask for a plain-English summary of the MD&A outlook section, the risk-factor additions, and the critical accounting estimates — each with verbatim excerpts and page numbers.
- Thesis drafting: Use a phase-based prompt sequence (business model → competitive position → financial quality → risks and valuation) to generate draft thesis fields. Phase-based prompt templates reliably guide AI assistants to produce thesis components with specific metrics like R&D % of revenue and ROIC calculation steps.
- Cross-filing references: Ask the AI to link disclosed acquisitions or executive departures in the 10-K to the corresponding 8-K filings.
Verification checklist for every AI output:
- Confirm all page citations by opening the source filing
- Reconcile every extracted number to the financial statements
- Sample-check at least two to three footnotes line-by-line against the AI summary
Pro Tip: AI-generated summaries are a starting point, not a substitute. Spot-check risk-factor additions and critical accounting-estimate language directly in the source filing before carrying any conclusion into the thesis.
For a deeper look at where AI fits into the equity research process, the AI equity research guide for RIAs covers verification patterns and workflow integration in detail.
How Filingsiq cuts thesis build time to 30 minutes
Filingsiq automates the mechanical phases of the workflow so you spend your time on judgment, not document navigation. The platform extracts change-diffs across risk factors, MD&A, and critical accounting estimates; surfaces newly added risk-factor language; pulls a three-year financial series with key ratios; and crosslinks 8-K events to the relevant 10-K disclosures. For a detailed walkthrough of the 30-minute thesis workflow, Filingsiq's own guide covers the exact steps.
A practical 30-minute use case: open your target 10-K in Filingsiq, run the risk-factor diff (5 minutes), review the extracted three-year financial series (5 minutes), read the flagged critical accounting estimate changes (5 minutes), scan the 8-K crosslinks for material events (5 minutes), and draft the thesis paragraph using the AI-generated section summaries (10 minutes). Phases 1 through 5 of the workflow above are largely automated; Phase 6 — the valuation scenarios and monitoring plan — requires your judgment.
The verification rule holds regardless of platform: always open the source filing to confirm footnote language, accounting changes, and legal-risk disclosures before finalizing the thesis. Filingsiq accelerates extraction; primary-source spot-checks remain the analyst's responsibility.
Pro Tip: Use Filingsiq's section-targeted diff to focus on risk factors and critical accounting estimates alone. Reading the full diff on a 200-page 10-K is still infeasible; targeted diffing on the two highest-signal sections takes under 10 minutes.
Key Takeaways
A decision-ready investment thesis requires a one-paragraph claim, three filing-sourced evidence points, two quantified valuation scenarios, and at least one monitoring trigger — producible in 30–60 minutes with an AI-assisted workflow.
| Point | Details |
|---|---|
| Time-to-thesis benchmark | A structured AI-assisted workflow produces a decision-ready thesis in 30–60 minutes; Filingsiq targets 30 minutes. |
| Diff first, read second | Diffing risk factors and critical accounting estimates against the prior year surfaces material changes faster than linear reading. |
| Earnings growth, not multiple expansion | A thesis that only works on multiple expansion is not defensible; validate earnings drivers from filing disclosures. |
| Accounting estimates move earnings | Changes in critical accounting estimates can shift reported earnings 10–20% with no operational change; check every quarter. |
| Filingsiq accelerates extraction | Filingsiq automates change-diffs, financial series extraction, and 8-K crosslinks, cutting mechanical work to roughly 30 minutes. |
A workflow worth embedding in your coverage process
The analysts who get the most from this workflow are the ones who treat it as a repeatable standard, not a one-time exercise. Run it on your next earnings 10-Q or the latest 10-K for a top-hold name. Timebox each phase strictly. Then compare what you found against your current coverage note — specifically, whether the diff surfaced any risk-factor additions or accounting-estimate changes your note did not address.
Adoption tips that actually stick: set automated filing alerts on SEC EDGAR for every ticker in your coverage universe, standardize your AI prompts so the output format is consistent across names, and require at least one footnote spot-check per thesis before it goes to the PM. Tie your monitoring triggers directly to filing events rather than price moves. A watchlist built around filing dates and specific disclosure triggers beats broad, unfocused reading every time.
Measure improvement on three dimensions: time-to-thesis, number of new signals captured from filings that were not in the prior note, and rate of thesis changes triggered by filing disclosures rather than price action. If filings are not driving thesis updates, the workflow is not yet embedded.
Filingsiq makes the 30-minute workflow real
Analysts who run this workflow manually still spend 2–3 hours on a single 10-K. Filingsiq closes that gap by automating the extraction steps that consume most of that time: automated change-diffs across risk factors and MD&A, a risk-factor newness detector that flags additions lawyers added under disclosure pressure, an extracted three-year financial series with ROIC and FCF conversion pre-calculated, footnote crosslinks that map the reference web automatically, and AI prompt templates calibrated to the phase-based workflow above.
The 30-minute trial use case: upload a target 10-K, run the risk-factor diff, review the flagged accounting-estimate changes, pull the financial series, and draft the thesis paragraph using Filingsiq's section summaries. Primary-source spot-checks on footnotes and legal-risk language remain your responsibility — Filingsiq surfaces the signals; you confirm them. See Filingsiq's pricing and plans to start your first 30-minute thesis build today.
This article is general information for educational purposes. Confirm current SEC disclosure requirements and accounting standards with SEC EDGAR, FASB, or a qualified financial professional for your specific situation.
Primary sources and further reading
The authoritative source for all U.S. public company filings is SEC EDGAR, where 10-K, 10-Q, 8-K, DEF 14A, and Form 4 filings appear within hours of submission. For accounting standard verification, FASB's Accounting Standards Codification is the primary reference for revenue recognition (ASC 606), lease accounting (ASC 842), and critical estimate disclosures.
For practitioner workflow posts, the alphactor.ai piece on reading primary filings directly is the clearest published account of why diffing beats linear reading. The Filingsiq guide on analyzing a 10-K filing provides a stepwise methodology with AI-assisted templates. For business-report reading fundamentals that complement filing analysis, ExpressPlanner's practical guide covers extracting key risk and return factors from primary documents.
Suggested practice tickers for first runs: a mid-cap industrial with segment-level disclosure, a software company with significant deferred-revenue footnotes, and a retailer with material lease obligations. These three filing types surface the widest range of footnote complexity in the shortest time.
"The reader who goes to the primary source catches the information the summaries drop. The primary source is the one place all of it still lives." — alphactor.ai, SEC Filings: Reading the Primary Source Directly
FAQ
How long does it take to build an investment thesis from a 10-K?
A structured AI-assisted workflow produces a decision-ready thesis in 30–60 minutes. Filingsiq targets 30 minutes by automating change-diffs, financial extraction, and section summaries.
What sections of a 10-K matter most for thesis development?
Start with MD&A and the core financial statements for context, then diff the risk factors and critical accounting estimates against the prior year. Newly added risk factors are disproportionately informative because legal counsel adds them under disclosure pressure.
Why is multiple-expansion-only reasoning a weak thesis?
Multiple expansion is volatile and harder to predict than earnings-driven returns. A defensible thesis requires evidence-backed earnings drivers from filing disclosures, not just a valuation re-rating argument.
Can AI replace primary-source filing review?
No. AI summaries are a starting point; they lose footnote nuance and accounting-estimate detail. Always spot-check risk-factor additions and critical accounting-estimate language directly in the source filing before finalizing any thesis claim.
What is the minimum output for a decision-ready thesis?
One thesis paragraph, three filing-sourced evidence points, two quantified valuation scenarios, and at least one monitoring trigger tied to a specific filing event or date.
Recommended
Related insights
Ready to analyze filings faster?
Try FilingsIQ free and turn SEC filings into actionable research in minutes.