Sense
Spot opportunities, threats and risks in your environment early, and tell signal from noise. Built by the Planning and Collection phases.
Methodology
Competitive intelligence (CI) isn't "Googling the competition." It's a craft with its own process, its own sources, and its own quality control, and for a company it's an asset, not a cost. Here's how I work, in plain language, and why it matters.
Why intelligence is a company asset
In a crowded, turbulent market the winner isn't whoever has the most data, it's whoever notices a change first and can act on it. That's exactly what competitive intelligence does: it turns the scattered signals around your company into decisions that hold up.
In peer-reviewed research (Černý et al., Review of Managerial Science, 2026) I describe CI as a dynamic capability: a repeatable organizational habit that directly strengthens resilience. It rests on three functions: sensing, seizing, and transforming. A company that runs this loop quickly and accurately shrinks the gap between "something is happening" and "we know what to do," and that gap decides whether a market shift becomes an opportunity or a wound.
Put plainly: intelligence isn't a one-off report. It's a capability you build and keep, one you can measure and improve over time.
And this is the crux: in that research we don't just describe it, we demonstrate that CI isn't only a tool for supporting strategic decisions, but above all a tool that strengthens the resilience of the organization itself.
The three functions intelligence rests on
The whole cycle serves three capabilities. When one is missing, the whole thing stalls, so I track each on its own.
Spot opportunities, threats and risks in your environment early, and tell signal from noise. Built by the Planning and Collection phases.
Turn a finding into a decision and commit resources before the window closes. Covers Processing, Analysis, Synthesis, Dissemination and Decision.
Learn from outcomes and reconfigure both the company and the intelligence process itself. Driven by the Feedback phase.
The intelligence cycle, step by step
Every engagement runs on the intelligence cycle: eight phases that turn fragmented data into a decision. It comes from the world of state intelligence services and was adopted in business by disciplines like competitive intelligence and OSINT. For each phase I note what it's for and why it matters.
The textbook "cycle" is a useful model, not gospel. Real practice looks more like a non-linear loop than a tidy wheel, which is why I revisit and rewrite the brief whenever collection or analysis teaches me something new.
Six stages of CI maturity
Companies run these same phases with very different effectiveness. I describe CI maturity in six stages: from no CI at all to a fully developed capability. It's a map of where to move next.
These stages aren't a ladder with a guaranteed climb. Companies move up and slip back. Without constant attention, maturity decays. Resilience isn't a state you reach at the top; it's an equilibrium kept by running the cycle.
Estimate your company's maturityFour disciplines, one method
A systematic, ethical program for gathering and analyzing information about the competitive environment, whose primary role is strategic early warning (SCIP's definition). It isn't spying, it's enlarging the dataset your decisions are built on.
TI is the systematic process of monitoring the external environment with a focus on technologies that could affect your position (Brenner, 1996). Scouting is the search-and-validate practice underneath it: actively finding and confirming new ideas and technologies, including their feasibility, not just sourcing partners (Brenner, 1996; Rohrbeck, 2010). Same data, different output, which is why I don't mix them.
Intelligence produced from publicly available information, collected and delivered against a specific intelligence question (ODNI's definition). Legal and ethical. The edge isn't secret sources, it's knowing which public source to query and how to verify it.
A patent landscape should answer a specific question for a defined audience, not plot everything a database can (WIPO, 2015). FTO is its own three-stage process (describe, search, analyze) with limits stated explicitly (WIPO's FTO guide).
Where I use AI, and where I don't
AI is an accelerator, not a source of truth. Language models predict plausible text; they don't verify facts, which is why they "hallucinate" confidently and wrongly. In high-stakes decisions that's the worst failure mode: subtly wrong. I handle it with structure, not better prompts.
Sources & further reading
This page summarizes publicly available standards and peer-reviewed research. Where possible I link directly to the primary source.
Yes. OSINT works exclusively with publicly available information using legal and ethical methods. The value isn't in classified data, it's in knowing which public source to query and how to verify what you find.
Data is raw. Information is data that's been sorted and verified. Intelligence is information that's been analyzed and pointed at a specific decision. It only becomes "actionable" once someone uses it.
No. AI has made reading and first-pass research cheap, but it can't decide which question to ask, nor vouch for what's true. The edge today isn't access to sources, it's judgment, and a documented chain of evidence under every claim.
Yes, when done properly. CI is systematic work with legally available information under the SCIP code of ethics: no impersonation, no data theft, no industrial espionage. A clearly defined methodology is exactly what separates professional intelligence from the grey zone, and what makes it an output you can comfortably show your legal department.
A first working cycle (from key questions to the first decision brief) can be up and running in weeks. Building a genuine capability, the higher maturity stages, takes months to years and depends on leadership backing. That's why I start with a maturity audit: it shows where you are and what the smallest sensible next step is.