peer-review-template

name	peer-review-template
description	Structure a peer review of analytical work before stakeholder delivery. Use when reviewing a colleague's analysis, dashboard, or query for quality, correctness, and clarity.

When to use

• Reviewing analytical work before it reaches stakeholders or goes into production
• A dashboard is being promoted to production and needs a formal review
• An analysis will inform a strategic decision and independent validation is warranted
• Any analysis that will be cited externally or used in a regulated context

Process

1. Clarify review scope — agree with the author on what the review covers: analytical logic, statistical methods, code quality, presentation, or all of the above
2. Assess analytical rigour — verify: the research question maps to the methodology, assumptions are documented, data supports conclusions, and uncertainty is communicated
3. Technical review — if code is included: check reproducibility, correctness of logic, code readability, and performance; run scripts/qa_runner.py if applicable
4. Provide structured feedback — categorise findings as: must-fix (blocks delivery), should-fix (quality issue), or nice-to-have (enhancement); be specific and actionable — "line 47: join produces fan-out, add DISTINCT or aggregate first"
5. Author response — author addresses each item and documents: resolved, won't fix (with rationale), or deferred
6. Sign off — reviewer confirms critical items are resolved; document in assets/review_signoff_template.md

Inputs the skill needs

• Required: the analytical output (notebook, report, dashboard spec, or SQL)
• Required: agreed review scope
• Required: reviewer name and relationship to the work

Output

• Completed review document with categorised feedback (must-fix / should-fix / nice-to-have)
• Author response log documenting resolutions
• assets/review_signoff_template.md (filled) — formal sign-off confirming critical items resolved