40 synthesis patterns

Where 3+ operators converge on the same idea from different angles. Each pattern names the operators, the cards, and the implication.

• The AEO triangle, presence, relevance, manual-action propagation (10 ops)
• Agent-first GTM (rebuild, don't bolt-on) (8 ops)
• Context, not capability, is the bottleneck (8 ops)
• Substrate runs the loop; humans run alignment and taste (7 ops)
• Verification, not execution, is the irreplaceable human job (7 ops)
• Build for the next model, not the current one (6 ops)
• Copywriting craft is built on five reinforcing fundamentals (6 ops)
• Differentiation requires three checks, different, better, matters (6 ops)
• Differentiation, not competition, is what compounds (6 ops)
• Status quo / no-decision is the real competitor (6 ops)
• Agents mapped 1:1 to JTBD with named human checkpoints (5 ops)
• Distribution and earned channels are the new moat (5 ops)
• Frontline customer contact is the PMM substrate (5 ops)
• Generalists with taste, shipping end-to-end (5 ops)
• Make the implicit explicit (5 ops)
• Pricing is a behavioral-architecture problem (5 ops)
• Sell to the buyer's mindset, not to product features (5 ops)
• Sustainability beats optimisation (5 ops)
• AEO experience and structure gate (4 ops)
• Company story IS the strategy, name the shift, run the audit (4 ops)
• Decision quality at scale comes from process, not willpower (4 ops)
• Diagnose before executing, refuse the playbook ask (4 ops)
• Execution is becoming free; judgement is the part that doesn't compress (4 ops)
• LLM-as-OS, post-training as moat (4 ops)
• Pricing is the most leveraged and most under-invested function (4 ops)
• Principal/staff IC as force-multiplier archetype (4 ops)
• Quality and friction-as-feature are growth levers, not constraints (4 ops)
• Sales is an engineered system, not individual art (4 ops)
• Absolute counts + correlated short signals, not stage rates and long loops (3 ops)
• Agents are first-class users, design for output, not navigation (3 ops)
• AI defensibility comes from non-AI moats (3 ops)
• Evals are data analysis, single judge, binary rubrics, error analysis first (3 ops)
• Market and offer beat funnel optimisation (3 ops)
• Rapport surfaces what research cannot (3 ops)
• Repair, generous interpretation, identity-vs-behavior, parenting frames as leadership ops (3 ops)
• Research preview, frontier programs, seasons not roadmaps (3 ops)
• Specific knowledge × circle of competence: the moat is the boundary (3 ops)
• Specificity becomes profitable, long tail, niche, segment-aware pricing (3 ops)
• Subtraction-first operating discipline (3 ops)
• The Economic Turing Test, outcomes pricing, agent labor, revenue per employee (3 ops)