This template structures thinking at three points: before coding, during review, and after release. It is not a bureaucratic checklist — it's a discipline for better decisions.
Rate each factor 1–5. Gates override scores.
| Factor | 1 (Low) | 3 (Moderate) | 5 (High) |
|---|---|---|---|
| Strategic relevance | No clear link to any strategic initiative | Supports a strategic goal indirectly | Directly advances a top-3 priority |
| Expected user/business value | Speculative benefit, no evidence | Reasonable hypothesis with some signal | Strong evidence of demand or validated problem |
| Cost of delay | Can wait months with no consequence | Moderate competitive or learning cost from delay | Market window, dependency, or strategic harm from waiting |
| Downside risk | Easily reversible, low blast radius | Moderate scope, some users affected if wrong | Trust-sensitive, hard to reverse, broad user exposure |
| Clarity/readiness | Requirements unclear, many open questions | Mostly defined, a few unknowns remain | Well-scoped, acceptance criteria clear |
| Verification readiness | No test strategy, unclear how to validate | Partial coverage plan, some gaps | Clear verification plan, automated where appropriate |
| Activation burden | Ship it and done — no cross-functional work | Some docs, training, or rollout coordination needed | Requires sales enablement, support training, migration, or marketing |
| AI suitability | Requires deep domain judgment, high ambiguity | Parts can be AI-assisted with oversight | Well-defined, pattern-based, strong AI fit |
Any item should be paused or rerouted if:
- Downside risk exceeds tolerance for the current team capacity
- Verification is too weak for the failure modes involved
- Clarity is too low to execute safely
- Required activation burden is unrealistic right now
Weights decide preference. Gates decide permissibility. A high-scoring item that fails a gate should not proceed until the gate condition is resolved.
- What better competing option exists right now?
- Why is this above that option?
- What would make this safer or clearer first?
- What non-dev work is required to realize value?
- If this slips two weeks, what strategic harm occurs?
| Factor | What to assess |
|---|---|
| Review depth needed | Does this PR warrant a quick scan, a careful read, or a deep walkthrough? |
| Evidence strength | Are there tests, screenshots, or data that demonstrate correctness? |
| Rollback sensitivity | If this breaks, how hard is it to undo? |
| Trust/compliance sensitivity | Does this touch user data, payments, security, or regulated areas? |
| Hidden complexity | Is the diff size misleading? Are there non-obvious side effects? |
| Pattern similarity | Have we seen this type of change cause problems before? |
- What type of failure is most likely here?
- What evidence would catch it?
- Have we seen this pattern fail before?
- What reusable guidance should come from this review?
| Factor | What to assess |
|---|---|
| Regressions | Did anything break? What was the blast radius? |
| Adoption | Did intended users find and use it? |
| Repeat usage | Did they come back, or was it a one-time interaction? |
| Strategic movement | Did it advance the initiative it was meant to advance? |
| Follow-on cleanup | How much hidden work did it create post-merge? |
| Trust/satisfaction | Did it build confidence or erode it? |
- Was a better option available that we didn't choose?
- What did we underestimate?
- Did value actually materialize, or did we just ship?
- Did the review level match the true risk?
- What should change for future similar items?
Different measures serve different moments. Mix them or you'll only see the past.
| Category | When used | Examples |
|---|---|---|
| Leading | Before work starts | Cost of delay, strategic alignment, risk profile, verification readiness |
| In-flight | During coding and review | Churn, review depth vs risk, emerging complexity, rollout burden changes |
| Lagging | After release | Adoption, repeat usage, regressions, strategic progress |
| Learning | Across patterns over time | Repeated issue classes getting safer, review guidance reuse, better AI routing |