Competitive Landscape Analysis for Investors: An AI Workflow

An investor's competitive landscape should explain who competes for the same customer decision, how bargaining power and profit are distributed, which firms are gaining economic share, and what could change the structure. AI can collect peer disclosures and maintain the comparison. The analyst must define the market, decide which measures are comparable, and connect competitive change to forecasts and valuation.

This guide is based on public regulator and company-source methods. It does not rank AI products or report a hands-on comparison. We build AllMind, and the cross-company system described near the end is ours.

Define the market before naming competitors

A list of companies that use the same category label is not a market. Write four boundaries first:

The US Department of Justice and Federal Trade Commission's 2023 Merger Guidelines are written for antitrust enforcement, not equity research, but they are useful reading because they treat market definition, substitution, concentration, entry, and competitive effects as separate analytical questions. Investors should not use a legal screen as an investment conclusion. The discipline of defining the arena before calculating share is the transferable part.

Test at least two plausible market definitions. A narrow definition may reveal pricing power; a broader one may expose a substitute that category reports ignore. Keep both until the evidence favors one.

Build an evidence-status competitor map

Start with facts that can be compared and label the rest.

“Reported” means the company or regulator discloses the measure. “Calculated” means the inputs and formula are visible. “Estimated” requires a source, method, and range. A generated market-share percentage without those fields should not enter the model.

The SEC's company filing search is the starting point for public-company disclosures. Search every peer for the same concept, but preserve differences. One company may report end customers, another paid accounts, and another transactions. Matching the labels through embeddings does not make the denominators comparable.

Measure concentration carefully

The Herfindahl-Hirschman Index squares each participant's market share and sums the results. It is a compact description of concentration, not a forecast of margins or a legal conclusion.

For a hypothetical market with shares of 40%, 25%, 15%, 10%, and 10%, square each share and add the five results. The HHI is 2,650.

This calculation is reproducible, but its usefulness still depends on the market boundary and share data. If the 40% participant's share is global while the others are US-only, the arithmetic is precise and the analysis is wrong.

Always show a sensitivity table:

AI can recalculate these scenarios and trace the share inputs. The analyst decides whether the market definition makes economic sense.

Map the profit pool, not only revenue share

Revenue leadership can coexist with weak economics. Split the value chain into stages and estimate where gross profit, operating profit, working capital, and reinvestment sit.

Use filed segment data when available. If a company does not disclose the stage separately, mark the value as estimated or unavailable. Do not force a sum to a market-size report that uses another definition.

The important investment question is often the direction of the pool. A distributor may gain revenue share while suppliers retain pricing power. A platform may compress customer prices while expanding its take rate. The peer table should therefore track the mechanism, not just the outcome.

Compare strategic responses as commitments

Management language is cheap. Look for observable commitments:

• capacity under construction, with expected commissioning date;
• research and development or sales investment tied to the strategy;
• acquisitions and divestitures;
• disclosed pricing or packaging change;
• distribution expansion;
• contractual minimums, backlog, or customer concentration;
• cannibalization explicitly accepted by management.

Build a strategy ledger:

This is a better use of AI than generic sentiment. The system can compare what management said across periods and flag missing evidence. The analyst determines whether the commitment changes competitive advantage.

Convert the landscape into forecast variables

A competition section that never touches the model is background. Translate each mechanism into an observable variable:

Record the expected direction and time window before the next disclosure. This turns a narrative into a test.

Where AI helps and where it misleads

AI is useful for maintaining a peer grid across filings and transcripts, extracting comparable passages, tracking strategic commitments, and generating sensitivity tables from visible formulas. It is especially valuable when the same question must be asked across ten or twenty companies every quarter.

It is unreliable when asked to invent a definitive market size, infer private-company share from web mentions, or merge incompatible metrics into a single ranking. General web assistants are effective for vocabulary and source discovery. Licensed research platforms add broker and expert perspectives. Private-market databases add funded entrants and transaction history. A document-analysis product can reason over a bounded diligence set.

We designed AllMind for cross-company work that joins filings, market data, licensed research, and a firm's own evidence through an entity map. Our live data-source catalog documents private-company profiles and modeled financials, funding rounds and investors, M&A transactions, ownership and holdings, company registries, software adoption, hiring, web and app activity, patents, trade flows, and other alternative signals. That makes AllMind a built-in private-to-public landscape system, not merely an analysis layer over user-supplied documents. The buyer should still test the hardest comparability problem, preserve each provider and as-of date, and avoid treating a missing or estimated field as fact.

Review with an adversarial checklist

Before the landscape reaches an investment memo, ask:

1. Would the competitor list change under a reasonable alternative market definition?
2. Does every share figure have the same geography, period, and denominator?
3. Which numbers are company-reported, calculated, or estimated?
4. Does revenue share differ from gross-profit or operating-profit share?
5. Which management claims are backed by committed capital or observable action?
6. What substitute is missing because no public company reports it?
7. Which result changes the forecast, and which is merely interesting?
8. What evidence in the next two quarters would reverse the conclusion?

We did not calculate a real market's concentration or compare vendor performance. The HHI example is hypothetical and included to make the method reconstructable. Legal market analysis and investment market analysis have different purposes; consult qualified counsel for antitrust questions.

The most useful first artifact is the evidence-status competitor map. If the same denominator cannot be established, publish the incompatibility instead of a share chart.

Sources and methodology

• DOJ and FTC 2023 Merger Guidelines, used for primary-source concepts around market definition, substitution, concentration, and entry. No legal conclusion is drawn here.
• SEC Search Filings, used as the primary public-company disclosure source.
• Calculations and templates in this article are original explanatory artifacts; the concentration example is hypothetical.