Intelligence Thinking Cycle

In many companies, what’s called competitive intelligence is often just data collection presented as strategy. Real intelligence takes a disciplined approach to sort facts from assumptions, and assumptions from guesses, including what your competitor wants you to believe. It’s not something you can hand off to an intern using Claude AI.

Below is a framework, based on intelligence analyst methods, designed to help you use competitive intelligence that actually guides your decisions.

If you’re wondering what process to use to turn competitive data into reliable intelligence for strategy, this article outlines the full 20-step framework that professional intelligence analysts use, tailored for business.

The $3M Decision Built on an Assumption Nobody Checked in 2023
A SaaS platform was ready to invest $3M into building enterprise security features.

They reasoned that competitor X was discounting aggressively. They thought they were clearly losing market share. So if they built enterprise security while they were struggling, they would capture the enterprise segment before they recovered.

Someone asked, “How do you know they’re discounting because they’re losing market share?”

“Our sales team says prospects are getting lower quotes from them.”

“That’s what you’ve heard. Is that what’s actually happening?”

We used a structured intelligence analysis. In just three weeks, everything looked different.

What they knew: Competitor X was offering lower prices to certain prospects.

What they thought they knew: Competitor X was losing market share and discounting out of desperation.

What they didn’t know: Whether the discounting was across-the-board or targeted. Whether market share was actually declining. Whether the discounting was desperate or strategic.

What they needed to know: The actual reason behind the pricing changes before committing $3M.

The reality:
Competitor X was not losing market share. They were intentionally gaining ground in one segment: mid-market healthcare, where they had just launched a HIPAA-compliant product. The so-called discounting was actually introductory pricing for healthcare customers only. Their total pricing stayed the same.

They weren’t struggling. They were executing a planned vertical expansion funded by a quiet $20M credit facility.

The SaaS company was close to spending $3M on enterprise security features to take advantage of a weakness that wasn’t real. They were acting on an assumption from their sales team that wasn’t backed by enough information.

One unchecked assumption nearly cost them $3M. Meanwhile, the real competitive threat, vertical expansion into healthcare, was missed entirely.

Why Most Competitive Intelligence Is Just Organised Guessing
At Octopus Intelligence, we are a UK and US-based competitive intelligence agency founded by former British military intelligence analysts. For twenty years, we have worked with companies like Mercedes, Samsung, and European governments.

Here’s what we’ve learned: Most companies confuse data collection with intelligence analysis.

They gather competitive information. They organise it. They present it. They call it intelligence.

That isn’t intelligence. It’s organised data with no real analysis.

Agencies like MI6 and the CIA don’t operate this way. They use disciplined thinking to separate facts from assumptions, test ideas against evidence that might prove them wrong, consider possible deception, and rate their confidence in their conclusions.

The difference matters. Data collection tells you what’s happening. Intelligence analysis tells you what it means, how confident you should be, and what you should do about it.

Most competitive intelligence programs skip analysis altogether. They collect information, write reports, and miss the real insights.

Intelligence Thinking Framework
Here’s the structured analytical process that’s adapted from professional intelligence methodology for competitive strategy and our own experience:

Step 1: Define the Intelligence Question
Not “tell us about competitors.” Not “what’s happening in the market.”

A specific question tied to a specific decision.

“Should we invest $3M in enterprise security features based on Competitor X’s apparent pricing weakness?”

“Should we enter healthcare vertical given competitive dynamics?”

“Is Competitor Y’s growth sustainable or will they face financial pressure in 12 months?”

If your question is vague, your intelligence will be vague too. If your question is specific, you’ll get intelligence you can act on.

Most competitive intelligence fails during the initial stages of any project. Teams collect information broadly rather than drilling down on specific things. They write reports that answer questions nobody’s asking. All while ignoring questions that actually drive decisions.

Step 2: Establish What You Actually Know
Only write down the facts supported by the reliable evidence you have found.

Dont say “Competitor X is struggling.” That’s just interpretation.

Rather, “Competitor X offered three prospects in our pipeline lower prices than their published pricing in Q2.” That’s a fact, assuming your sales team reported accurately.

Not “The market is shifting to cloud.” That’s too broad a claim.

Look to be evidence-based, like “Four of our last six lost deals went to cloud-native competitors. Customer interviews in three of those deals cited cloud architecture as deciding factor.”

Listing only what you really know, with evidence to back it up, can be eye-opening. Much of what’s called knowledge is actually assumption, hearsay, or repeated interpretation that starts to feel true.

Step 3: Examine How You Know It
Every piece of information has a source. Sources have different reliability levels.

First-hand evidence: You tested the competitor’s product yourself. You saw their pricing page. You attended their conference presentation.

Documentary evidence: Published financial and competitive intelligence reports. Press releases. Patent filings. Job postings.

Corroborated information: Multiple independent sources saying the same thing.

Inferred information: You’ve concluded from indirect evidence.

Claimed information: Someone told you, but you haven’t verified it.

Most competitive ‘knowledge’ fits into these last two groups. Sales teams share what prospects tell them (claimed), and analysts draw conclusions from incomplete data (inferred).

Neither type of information is useless, but neither should be treated as fact.

A sales rep says: “Prospect told me Competitor X offered 40% discount.” That’s claimed information from a single source with potential motivation to exaggerate (prospect may have been using rival pricing as leverage).

Before you build strategy on that claim, ask: How do we know this? What’s the source? How reliable is that source? Is this corroborated?

Step 4: Identify What You Think You Know
This is the most dangerous category.

Information being treated as fact but actually containing assumptions, interpretation, or weak sourcing.

Example: “Competitor X is losing market share.”

Do you actually know this? Or did someone say it in a meeting six months ago and it’s been repeated until it became accepted truth?

Check: What evidence supports this claim? Is the evidence current? Is it from reliable sources? Could it be interpreted differently?

Most strategic mistakes come from this group: things the organisation thinks it knows but can’t prove, and assumptions that become accepted as facts just by being repeated.

We force every project team at Octopus Intelligence to complete five sentences before starting substantial research:

We know… (verified facts with identified sources);
We think we know… (information treated as fact but not fully verified);
We don’t know… (explicit intelligence gaps);
We need to know… (gaps that would change the decision);
We would change our assessment if… (disconfirming evidence we should look for)
That last sentence is the most important analytical habit in competitive intelligence. It makes you decide what would prove you wrong before you start searching for evidence that supports your view.

Step 5: Map What You Don’t Know
Explicitly list your intelligence gaps.

Most companies skip this step. They focus on what they’ve gathered and present it as complete. But often, the gaps are more valuable than another page of data.

Example gaps:

“We don’t know Competitor X’s actual churn rate” “We don’t know whether their discounting is company-wide or segment-specific” “We don’t know their current burn rate or runway” “We don’t know what their product roadmap prioritises for next 12 months”

Listing gaps does two things: It prevents false confidence in your analysis. It identifies where further collection would be most valuable.

Step 6: Prioritise What You Need to Know
Not every unknown weighs equally.

Some gaps could change your decision completely. Others are interesting but irrelevant.

Priority Intelligence Requirements (PIRs) are the gaps that would actually change what you do.

“We need to know whether Competitor X’s discounting is segment-specific or company-wide” — this directly affects whether we interpret it as a strategic move or desperation.

“We need to know Competitor X’s employee count” — nice to know but probably won’t change our decision.

Focus your efforts on PIRs. Don’t spend time or resources on questions that won’t affect your strategy.

Step 7: Surface Your Assumptions
Write down every assumption you can think of.

Like that, you assume Competitor X is discounting because they’re losing market share.

You assume the enterprise segment values security features above all else.

You assume our product is technically superior.

Or market growth will continue at current rates.

Now ask: What happens to our strategy if each assumption is wrong?

If the strategy collapses when one assumption fails, that assumption is a critical vulnerability. It needs to be tested, not assumed.

Most strategies rely on five to ten untested assumptions. When strategies fail, it’s often because one of these assumptions was wrong. Companies that test their assumptions before investing come out ahead. Those that find out after spending $3M pay a high price.

Step 8: Develop Competing Hypotheses
Don’t fall in love with your first explanation.

Develop multiple plausible explanations for the evidence you’re seeing.

Competitor X is offering lower prices. But why?

H1: They’re losing market share and discounting to retain customers. A sign of desperation.
H2: They’re deliberately buying market share in a specific segment. A strategic investment.
H3: They’ve reduced costs through new technology and are passing savings to customers. They have improved their finances.
H4: The discounting is limited to specific customer segments or deal sizes. Is it targeted pricing?
H5: They’re preparing for acquisition and need growth metrics. Creating a story.
Each hypothesis leads to a completely different strategic response.

If H1 is true: Attack aggressively while they’re weak. If H2 is true: Defend your segment, don’t chase theirs. If H3 is true: Investigate their cost reduction and determine if you’re at a structural disadvantage. If H4 is true: Understand which segments they’re targeting and why. If H5 is true: Prepare for a well-funded acquirer entering your market.

The same evidence can lead to five different explanations and five different strategies. Most companies pick H1 because it feels safest. They assume the competitor is weak and decide to attack, without checking if that’s actually true.

Step 9: Test Each Hypothesis Against Evidence
For each hypothesis, ask: What evidence would we expect to see if this hypothesis were true?

If H1 (losing market share) is true, we’d expect:

Declining customer count (check their case studies, logos, references)
Employee departures (check LinkedIn)
Reduced marketing spend (check SimilarWeb, ad tracking)
Defensive positioning in marketing (check messaging changes)
If H2 (buying market share) is true, we’d expect:

Discounting in specific segments only (check across deal types)
New capabilities targeting those segments (check product releases)
Hiring in those segments (check job postings)
Maintained investment elsewhere (no broad cost cutting)
Now look for evidence that contradicts each hypothesis, not just evidence that supports it.

If you find evidence that contradicts H1 (they’re actually hiring aggressively and investing heavily), H1 becomes less likely regardless of how much confirming evidence exists.

The hypothesis with the least disconfirming evidence is most likely correct.

This is called Analysis of Competing Hypotheses (ACH), the same method CIA analysts use to evaluate threats. It works for competitive intelligence because it helps you avoid confirmation bias.

Step 10: Search for Collection Bias
Are the sources available to you distorting the picture?

Your sales team sees competitor pricing (but only in deals they’re in). Your customers see competitor products (but only the features they evaluate). Public financial reports show revenue (but not segment breakdown). Former employees know internal dynamics (but their information is dated). Industry analysts report trends (but from their analytical framework).

Each source only sees part of the picture. No single source sees everything.

If your intelligence comes primarily from sales team reports, you’re seeing the competitive landscape through a sales lens. You might miss product development shifts, business alliances, or market approach changes that don’t show up in sales conversations.

Using a variety of sources helps reduce collection bias. When several independent sources with different viewpoints agree, your confidence in the conclusion grows.

Step 11: Consider Deception
Could someone be deliberately misleading you?

Competitors don’t just compete on products and pricing. They compete on information.

A competitor announcing “AI is central to our strategy” might be:

Genuinely investing in AI.
Trying to scare you into expensive AI investments
Positioning for fundraising (AI narrative attracts capital)
Distracting from their actual strategic moves
Public statements are meant for strategy, not for sharing real intelligence. Keep that in mind when you evaluate them.

We add a sixth sentence to our analytical framework specifically for competitive intelligence:

If we were the competitor, what would we want our rivals to believe?
This question introduces deception awareness into the analysis. It asks: Is the competitor’s public behaviour designed to make us react in a specific way?

If Competitor X wants you to believe they’re struggling (so you get complacent), discounting might be a deliberate signal rather than desperation.

If Competitor X wants you to believe they’re pivoting to enterprise (so you defend enterprise while they attack SMB), their announcements might be misdirection.

Most competitive intelligence treats competitor communications as fact. But that isn’t real intelligence; it’s just reading press releases.

Step 12: Define What Would Change Your Mind
This is the most powerful analytical discipline.

Before you conclude, define: What evidence would cause us to revise this assessment?

“We assess Competitor X is struggling. We would revise this assessment if: they raise new funding, hire aggressively, launch new products, or win deals in segments we expected them to lose.”

If you can’t explain what would make you change your mind, your conclusion is just a belief, not a real intelligence assessment. Beliefs don’t change with evidence. Intelligence assessments do.

Step 13: Red Team Your Conclusion
Give someone the job of attacking your prevailing conclusion.

Not to agree. Not to be constructive. To find every reason your assessment might be completely wrong.

“We think Competitor X is struggling because they’re discounting.”

The red team response was that three alternative explanations exist.

“Their discounting is sector specific and not company-wide. They have just hired 15 engineers and their LinkedIn posts show confidence. Not desperation. The sales team reporting discounts. But they be listening to exaggerated claims from prospects using competitive pricing as leverage.”

Red teaming can feel uncomfortable because it challenges the story everyone believes. That’s exactly why it’s so valuable.

Step 14: Adopt the Competitor’s Perspective
Ask: If I were this competitor, what would I be trying to achieve?

What constraints are they facing? What do they know that we don’t? What would be the rational move from their perspective?

Most companies analyse competitors from their own perspective. “We think they’re struggling because we’d be struggling if we did what they’re doing.”

But they’re not you. They have varied constraints, different information, different objectives.

A VC-backed competitor burning cash isn’t struggling. They’re executing a strategy where growth matters more than profitability. From their perspective, discounting to buy market share might be perfectly rational because growth rate determines their next funding round.

If you judge their strategy based on your own finances, you might misunderstand what they’re really doing.

Step 15: Develop Alternative Futures
Intelligence shouldn’t produce a single prediction. It should develop multiple plausible scenarios.

Most likely
Competitor continues current strategy for 12-18 months.

Optimistic
Competitor faces financial pressure and retreats from our segment.

Pessimistic
Competitor’s strategy succeeds, and they capture significant market share.

Disruptive
Competitor gets acquired by a larger player with more resources.

Each scenario needs a different strategy. If you plan for one outcome but something else happens, that’s how companies get caught off guard.

Step 16: Set Up Indicators and Warnings
For each scenario, determine what you’d expect to see beforehand.

If a competitor is succeeding
Increased hiring. Growing customer references—expanded marketing. Conference presence increasing.

If a competitor is struggling
Executive departures. Hiring freeze. Product stagnation. Customer complaints increasing. Marketing going quiet.

If a competitor is preparing for acquisition
Leadership meetings with potential acquirers. Strategic positioning changes. Growth acceleration regardless of profitability. PR narrative shifting to “market leader” positioning.

Review these indicators every week. When you notice them, you’ll know which scenario is unfolding before others do.

Step 17: Assess Source Reliability Separately from Information Credibility
A normally excellent source can deliver incorrect information. An unknown source can occasionally provide something accurate.

Ask both questions:
“How reliable is this source generally?” (track record, access, motivation) “How credible is this particular piece of information?” (consistency with other evidence, plausibility, corroboration)

A trusted former employee may be offering outdated information. A reliable source, potentially stale information. A prospect may tell you about accurate competitor pricing. That’s unknown reliability, but verifiable information.

Don’t trust information just because you know the source. Don’t dismiss information just because the source is unfamiliar.

Step 18: Distinguish Fact from Analytical Judgement
Your intelligence products should make clear the difference between:

We know
Facts with reliable evidence.

We assess
Analytical conclusions based on evidence and reasoning.

We believe
Judgements with limited evidence but supported by logic.

We don’t know
Acknowledged holes in understanding.

Most competitive intelligence presents everything with the same confidence level. “Competitor X is struggling and will face financial pressure by Q3” reads as fact when it’s actually analytical judgement based on limited evidence.

Being clear about these differences helps you avoid confusing uncertain opinions with facts.

Step 19: Assign Confidence Levels
Every conclusion needs to have a confidence indicator like this:

High confidence
Multiple independent sources
Strong corroborating evidence, and
Hypothesis that has survived disconfirmation testing
Moderate confidence
Some supporting evidence
Limited disconfirmation testing, or
Evidence from few sources
Low confidence
Limited evidence
Significant gaps, or
Evidence that could support multiple hypotheses
“We assess with moderate confidence that Competitor X’s discounting is segment-specific rather than company-wide, based on pricing data from four deals and two customer interviews.”

Confidence levels encourage honesty. They prevent analysts from presenting uncertain conclusions as facts and help decision-makers decide how much to trust each assessment.

Step 20: Answer “So What?” and “What Next?”
Intelligence becomes useful only when connected to a decision.

So what
What does this assessment mean for our strategy? What opportunities or threats does it create?

What next
What action should we consider? What additional intelligence should we collect? What indicators should we monitor?

If your intelligence doesn’t answer ‘so what’ and ‘what next,’ it’s just an academic exercise, not a useful business tool.

The Intelligence Thinking Cycle
At Octopus Intelligence, we compress this into a repeatable analytical sequence:

This is an image of a cycle and circle for an article called What Do We Know, and How Do We Know It by octopus competitive intelligence agency
This cycle never really ends. You don’t do it just once. As new evidence appears, repeat the process, update your assessments, adjust your confidence, and rethink your strategy.

Most companies do competitive intelligence as a project. Collect data. Write a report. File it.

Intelligence is a process, not a one-time product. If you don’t keep it going, your conclusions will soon be out of date.

Why This Matters More Than Most Companies Realise
The difference between data collection and intelligence analysis is the difference between:

“Competitor X is discounting” (data)

and

“We assess with moderate confidence that Competitor X’s discounting is limited to healthcare segment, driven by strategic vertical expansion funded by $20M credit facility, and represents deliberate market share investment rather than desperation. If correct, their pricing will stabilise within 6-9 months and they’ll have established healthcare positioning we’ll find difficult to challenge. We would revise this assessment if we see company-wide discounting, executive departures, or reduced product investment.” (intelligence)

The first tells you what’s happening. The second explains what it means, how confident you should be, what actions to take, and what could change the situation.

That’s the difference between guessing and knowing, between reacting and planning, and between wasting $3M on the wrong move or investing where it matters most.

What to Do This Week
Before your next decision, complete these sentences:

We know… (verified facts only)
We think we know… (assumptions treated as facts)
We don’t know… (explicit gaps)
We need to know… (gaps that would change the decision)
We would change our assessment if… (disconfirming evidence)
If we were the competitor, what would we want our rivals to believe? (deception awareness)
This exercise takes just 30 minutes. It will show you how much of your competitive “knowledge” is really just assumption, interpretation, or repetition.

At Octopus Intelligence, we apply the full intelligence thinking cycle to every competitive analysis we conduct.

We’re a UK, Dubai and US-based competitive intelligence agency built by former British military intelligence analysts serving companies like yours.

We don’t collect data and call it intelligence. We apply structured analytical methodology that separates facts from assumptions, tests hypotheses against disconfirming evidence, considers deception, and assigns confidence levels to every conclusion.

We help you make decisions based on assessed intelligence, not organised guessing.

Get in touch. Tell us about a competitive decision you’re facing.

We’ll apply the intelligence thinking cycle and show you the difference between what you know, what you think you know, and what you need to know before committing resources.

The gap between what you know, what you think you know, and what you need to know is where the big mistakes happen. Life-threatening mistakes for some intelligence analysts.

https://www.octopusintelligence.com/intelligence-thinking-cycle/