Extract templates from your 3 best past examples by finding common structure — don't design templates from theory

Why This Is a Rule

Templates designed from theory ("What should a status report contain?") reflect what you think should work. Templates extracted from your best actual examples ("What structure did my three best status reports share?") reflect what has actually worked. The difference is significant: theoretical templates include sections you've never needed and miss structures that emerge naturally from practice.

The three-example method works by triangulation: one example might have a unique structure due to specific circumstances. Two examples might share structure by coincidence. Three examples sharing the same structure indicates a pattern — the common elements are genuinely structural (they appear across different contexts) while the differences are genuinely variable (they change with each instance). This is the same inductive logic as Build checklists from documented error history, not hypothetical failures — promote errors to checklist items only after they actually occur's empirical checklist construction: derive from observed reality, not imagined ideals.

Choosing your "three best" rather than three random examples biases the template toward quality. The template captures the structure of your highest-quality outputs, essentially encoding "how I produce great work" into a reusable format. Each future instance starts from the structural DNA of your best performance rather than from a blank page or a generic template.

When This Fires

• When creating a template for any output type you've produced at least 3 times
• When existing templates feel generic and don't capture your actual best-practice structure
• When you want to systematize quality across recurring outputs
• Complements Design near-complete templates for recurring outputs — section headings, formatting, and boilerplate pre-filled, requiring only variable content insertion (near-complete templates) with the method for creating them

Common Failure Mode

Theory-first template design: sitting down with a blank page and designing "the ideal template" based on what you think should be included. The result is a template that looks logical but doesn't match how you actually work — it includes sections you never use and misses structures you always create ad-hoc.

The Protocol

(1) Identify the output type you want to template. (2) Find your three best past examples of that output type — the three instances where the output was highest quality and best received. (3) Lay out all three side by side (Lay out multiple notes in parallel visual access for synthesis — sequential reading prevents the simultaneous comparison that synthesis requires's parallel display). (4) Identify what's common across all three: shared section headings, shared structural patterns, shared formatting, shared opening/closing patterns. These are your template's fixed elements. (5) Identify what's different across the three: the variable content, the context-specific sections, the instance-unique data. These become the template's variable placeholders. (6) Build the template from the common structure with variable placeholders. The result is a template grounded in your demonstrated best practice, not in theoretical ideals.