← Back to Knowledge Graph

Convergent vs. Divergent Problem Solving: Why You Must Separate Idea Generation From Idea Evaluation to Create Breakthrough Offers

The Framework

Convergent vs. Divergent Problem Solving from Alex Hormozi's $100M Offers identifies the two cognitive modes that offer creation requires — and explains why most entrepreneurs produce mediocre offers by mixing them. Divergent thinking generates options without evaluation: brainstorm every possible customer problem, every possible solution, every possible bonus. Convergent thinking evaluates and selects: apply the Value Equation, use Trim & Stack, choose what stays and what goes. The critical rule: these modes must be separated in time. Evaluating while generating kills the creative output that exceptional offers require.

The Two Modes

Divergent Mode (Generate). Goal: maximum quantity. Rules: no criticism, no evaluation, no 'that's a bad idea,' no 'we can't afford that.' Every idea gets written down regardless of feasibility, cost, or practicality. The Problem Generation Matrix IS a divergent exercise — listing every possible problem a customer might have before deciding which problems to solve. The Brick Exercise trains this mode: list 30+ uses for a brick without evaluation.

Divergent mode feels uncomfortable because most adults have been trained to evaluate immediately. The discomfort IS the signal that the mode is working — the brain is generating ideas that the critical mind normally vetoes before they're fully formed. Many of the best Grand Slam Offer components (unconventional bonuses, surprising guarantees, counterintuitive pricing) emerge from divergent mode precisely because they wouldn't survive immediate evaluation.

Convergent Mode (Evaluate). Goal: maximum quality. Rules: apply specific criteria (the Value Equation), use structured evaluation tools (Trim & Stack), make decisive selections. Each idea is assessed on value-to-cost ratio, customer perception, operational feasibility, and contribution to the Price-to-Value Discrepancy. Low-value/high-cost items are eliminated. High-value/low-cost items are prioritized.

Convergent mode works best when it has a rich pool of options from the divergent phase. An entrepreneur who generated 5 ideas has 5 options to evaluate. One who generated 30 has 30 — and the best of 30 is dramatically better than the best of 5.

Cross-Library Connections

Fisher's Circle Chart from Getting to Yes prescribes the same cognitive sequence for negotiation: analysis (what's the situation?) → approaches (what are all possible directions?) → action ideas (what specific options exist?) → proposals (what should we offer?). The first two stages are divergent; the last two are convergent. Fisher's framework IS the divergent-convergent cycle applied to negotiation rather than offer design.

Cialdini's commitment and consistency from Influence explains why mixing modes produces mediocrity: once a person evaluates an idea positively (convergent), the consistency drive commits them to it — which stops the divergent generation of alternatives. The separation prevents premature commitment.

Hughes's Three Autopilot Bypass Categories from The Ellipsis Manual connect through the Confusion Operation: divergent thinking deliberately creates productive confusion (considering an object or problem from unfamiliar angles), which disrupts the habitual thinking that produces conventional solutions. The confusion IS the mechanism that produces unconventional ideas.

Berger's Inner Remarkability from Contagious predicts that the most shareable offer components emerge late in the divergent phase: the first ideas generated are conventional (and unremarkable), while the ideas generated after the obvious ones are exhausted tend to be surprising (and remarkable). The extended divergent session IS the remarkability generator.

The divergent-convergent separation also applies to team brainstorming: groups default to convergent mode even faster than individuals because social pressure makes people evaluate ideas before sharing them ('they'll think this is stupid'). Establishing divergent mode as the explicit norm ('we're only generating right now, zero evaluation') produces 3-5x more ideas from the same group. Wickman's Level 10 Meeting format from The EOS Life can incorporate this: IDS sessions (Identify, Discuss, Solve) should spend the Identify phase in divergent mode before the Discuss phase applies convergent evaluation.

Implementation

  • Schedule divergent and convergent sessions separately. Never on the same day if possible. The cognitive mode switch requires mental energy that same-day sessions deplete.
  • Set a minimum quantity target for divergent sessions. 30 ideas minimum for any brainstorming exercise. The quality threshold only applies in convergent sessions — divergent sessions measure quantity only.
  • Post the rules visibly during divergent sessions: 'No evaluation. No criticism. Every idea gets written down.' These rules counteract the default evaluative mode that adults revert to automatically.
  • Apply specific evaluation criteria during convergent sessions. The Value Equation (Dream Outcome × Perceived Likelihood ÷ Time Delay × Effort) and Trim & Stack provide the structured evaluation that convergent mode requires.
  • When stuck in convergent mode, switch back to divergent. If the evaluation process reveals that none of the generated options are strong enough, the solution isn't harder evaluation — it's more generation. Return to divergent mode and produce 20 more options before resuming evaluation.

    • 📚 From $100M Offers by Alex Hormozi — Get the book