Intelligence Collection Disciplines — Conduct Competitive Research Like an Intelligence Agency

Skills Overview

Intelligence Collection Disciplines is a set of competitive intelligence collection methods designed for product managers. It transforms the eight intelligence collection disciplines (from OSINT to MASINT), signal inference chains, and all-source fusion processes into a competitive research workflow that relies solely on public sources and can be carried out by anyone.

Use Cases

1. Validating competitors’ strategic moves: When you suspect a competitor is building a platform, entering a new market, or adjusting its pricing, use multiple independent channels—patents, hiring, financial reports, and website change records—to cross-validate the evidence and distinguish genuine commitments from public messaging before the competitor makes an official announcement.
2. Building a defensible TAM/SAM/SOM model: When market-size data is challenged in a review meeting or fundraising context, use government statistics, industry benchmarks, and transaction capture rates derived from financial reports to conduct a bottom-up estimate, with a traceable source for every step.
3. Establishing a sustainable competitive monitoring system: Scan websites and pricing changes weekly (SIGINT), organize public sentiment and personnel developments monthly, and conduct deeper reviews of financial reports and patents quarterly—turning one-off research into an continuously operating market intelligence system.

Core Features

1. Eight collection disciplines and source map: OSINT (open-source sentiment), FININT (financial reports and tenders), GEOINT/DEMINT (statistics and demographics), TECHINT (patents and technology), HUMINT (recruiting and personnel), SIGINT (website and pricing changes), and MASINT (supply chain and operations) are each accompanied by lists of free and paid sources, signal interpretations, and the product documents they support (battlecards, TAM models, roadmaps, etc.). Select just two or three disciplines based on the decision at hand; there is no need to run every discipline for every question.
2. Signal-to-inference chain: Every observable signal—such as a cluster of more than five similar patents filed within 12 months, 30+ specialist positions recruited in a single quarter, or an SSL certificate for a new subdomain—has a bounded interpretation and is labeled as a “fact,” “inference,” or “hypothesis.” This prevents jumping from a single signal directly to sweeping conclusions.
3. All-source fusion and confidence accumulation: An independent finding from one discipline is an observation; findings from two disciplines constitute a working hypothesis; findings from three or more mutually independent disciplines constitute actionable intelligence. An official announcement counts only as “intent,” while funding, hiring, permits, and contracts count as “commitments.” Cross-validation turns three weak signals into one strong conclusion.

Frequently Asked Questions

How is this different from ordinary competitive analysis?

Ordinary competitive analysis typically relies on a single channel—such as a company website, press releases, or review sites—and draws conclusions from one signal. In essence, it is anecdotal. This methodology emphasizes the independence of disciplines: a press release may exaggerate, but customs records, patent filings, and job postings come from entirely different sources. Confidence should be upgraded only when they point in the same direction. It also requires a clearly defined decision—such as a roadmap, positioning, or pricing decision—to be established before research begins. Research without a decision is merely a hobby.

Is competitive intelligence based solely on public sources legal?

Yes. All sources used in this methodology consist of information that has been published, officially filed, or is publicly observable, including government statistics, regulatory filings with the SEC, Companies House, and similar bodies, job boards, patent databases, and web-change records. It also clearly defines the boundaries: do not misrepresent your identity to obtain information, access content protected by an NDA, scrape websites in violation of their terms of service, or specifically target employees in order to obtain a former employer’s confidential information.

Do I have to purchase paid tools? Are free sources sufficient?

Yes, free sources are sufficient. Each discipline comes with a complete set of free sources, including SEC EDGAR and corporate registries in various countries, Eurostat and national statistical agencies, Google Patents, free BuiltWith lookups, the Wayback Machine, and ImportYeti. These are enough to complete single-discipline research and cross-disciplinary validation. Paid tools such as Klue, AlphaSense, and Panjiva are primarily useful for large-scale, continuous monitoring at the team level and can be added as needed after their value has been validated.