Skip to main content

​
Structured Honesty About Weak Thinking

What this solves: Strategic analysis that feels thorough but has never been stress-tested — where confidence comes from coherence rather than evidence.

​
The Problem

Most strategic analysis fails the same way: you build a coherent narrative from the evidence that supports your direction, and the coherence itself feels like proof. The analysis is internally consistent, well-structured, reads well — and misses something fundamental because nobody asked what it assumed, whose perspective was absent, or what alternative explanation fits the same data. The problem isn’t lack of rigor. It’s that rigor applied in one direction just builds a more convincing version of a potentially wrong answer.

​
How It Works

The skill enforces a thinking pattern with two phases: decomposition first, then mandatory blind spot detection through five lenses. Decomposition means breaking the question into components before interpreting anything. The premature leap to “so therefore…” is where most analysis goes wrong. Every conclusion has to trace back to premises you can name. When I catch myself thinking “obviously” or “clearly,” that’s exactly where the unexamined assumption lives. The five lenses are the part that earns the skill’s keep:
  • Assumptions — What did we accept without validating? Not just “what are our assumptions” but specifically: what’s the impact if this assumption is wrong, and how would we check?
  • Perspectives — Whose viewpoint is missing? Customers, competitors, adjacent teams, regulators, the person in the room who stayed quiet. Missing perspectives aren’t neutral — they skew conclusions in a specific direction.
  • Evidence — Where are we relying on secondary sources when primary data exists? Is our evidence current enough to trust? What sources did we not consult?
  • Logic — What alternative explanations fit the same data? Where did we mistake correlation for causation? Where did confirmation bias steer which evidence we weighted?
  • Scope — What’s outside our frame that matters? Adjacent market shifts, second-order effects, regulatory changes. The most dangerous blind spots sit outside the boundaries we drew.
The analysis must surface at least 3 assumption blind spots, 2 perspective gaps, and 2 alternative interpretations — prioritized by which ones, if wrong, most change the answer.

​
In Practice

I used this on a market entry analysis where the initial read was straightforward: strong product-market fit signals, growing category, clear competitive gap. The kind of analysis where everything points one direction and you start writing the recommendation. The assumptions lens caught that the “growing category” claim relied on analyst projections, not measured demand. The perspectives lens surfaced that the analysis had no input from the sales team who’d actually have to sell it. The logic lens identified that the “competitive gap” might be a gap because competitors had tried and abandoned it, not because they hadn’t noticed it. Three findings that didn’t disprove the thesis but changed it from “clear go” to “conditional go with three things to validate first.” Total time: about 20 minutes of structured thinking versus what would have been a 45-minute free-form analysis that probably would have confirmed the initial direction.

​
What I Learned Building This

The hardest part was making the blind spot detection non-optional. My first version had the lenses as a suggested checklist. I’d skip them when the analysis already felt solid — which is exactly when they matter most. Making them mandatory and requiring minimum counts (3 assumptions, 2 perspectives, 2 alternatives) was the change that made the skill actually useful. The scope lens produces the most important finding more often than any other. Assumptions and perspectives feel like the important ones — they’re dramatic, they challenge your thinking directly. But scope catches the things that aren’t wrong so much as incomplete. “You analyzed this correctly within the frame you chose, but the frame itself is too small.” That’s a different kind of insight and it’s the one that most often changes a recommendation. I also learned that holding competing hypotheses simultaneously — really holding them, not just listing alternatives to dismiss — is something that takes explicit structure to do well. The natural tendency is to evaluate each alternative against your preferred explanation and find it wanting. The skill forces you to ask “what evidence would I need to see to prefer this alternative?” which is a fundamentally different question than “why is this alternative wrong?”