The Critical Deal Framework is the five-step method that runs every Critical Deal engagement. Built for B2B research where the cost of being wrong is asymmetric.
Why the framework exists
Most B2B research processes look like this: a brief, a questionnaire, a fieldwork phase, an analysis phase, a deck. The research is structured around the data, not the decision.
The Critical Deal Framework is built the other way round. The research is structured around the decision, and every step is in service of the verdict the decision-maker needs.
The five steps
1. Name the decision
Every engagement opens with a one-page Decision Brief. Not a research brief, a decision brief.
- What is the specific decision being made?
- Who is making it?
- By when?
- What are the options on the table?
- What is the cost of being wrong on each option?
If we cannot get a sharp answer to all five, we keep working on the brief until we can. Vague decision = vague research.
2. Name the assumptions
Every decision rests on assumptions, claims that have to be true for the decision to be the right one. Once the decision is named, the second step is naming those assumptions explicitly.
- What has to be true about the market for this decision to be right?
- What has to be true about the buyers?
- What has to be true about the competition?
- What has to be true about the team's ability to execute?
- What has to be true about the economics?
Each assumption becomes a research target. The list is usually 5 to 12 items. If it grows beyond 20, the decision isn't framed sharply enough yet.
3. Go to source
For every named assumption, we identify the source of truth, and we go to it. Sources include:
- Buyer interviews. The people actually making the buying decisions in the category.
- Operator interviews. The people running the function at customer companies.
- Expert interviews. Senior practitioners with deep category knowledge.
- Quantitative surveys. Where the assumption needs sizing or distribution evidence.
- Existing data. CRM, product analytics, public filings, prior research.
- Competitive signals. Public moves, hiring, pricing, partnerships.
Most assumptions get more than one source. Triangulation is the discipline that turns multiple imperfect signals into a defensible read.
4. Triangulate the truth
No single source is enough on its own. Triangulation is the explicit step where we compare what each source says, surface where they agree, and surface where they disagree.
The disagreements are usually where the real insight lives. When the buyer interviews say one thing and the public signals say another, the gap is the finding.
We also explicitly surface dissenting voices. If 3 out of 25 buyers disagreed with the dominant pattern, we name them, characterise them, and read what their disagreement implies. Dissenters are usually the early indicator of where the dominant pattern is about to break.
5. Deliver the verdict
The output is a verdict, not a deck.
- The decision: Go / No-go / Go with conditions.
- The conditions: explicit, named, and operational.
- The assumption table: each assumption, the evidence, the confidence level.
- The red flags: anything the decision-maker needs to know that doesn't fit cleanly into the verdict.
- The evidence pack: transcripts, datasets, model, supporting analysis.
The verdict is short. It can stand on its own in front of a board, an investment committee, or an acquirer's diligence room. The evidence pack sits behind it for anyone who wants to dig.
The framework in one sentence
Name the decision, name the assumptions, go to source, triangulate, deliver the verdict.