Template

Experiment Log & Postmortem — Reusable Template

An interactive experiment log you can save to organizational memory. Capture hypothesis, design, analysis plan, results, decision, and lessons so teams learn faster and avoid repeating mistakes.

Interactive Tool

Experiment Log & Postmortem — Reusable Template

Use this form to record the full lifecycle of an experiment: the hypothesis, planned design and analysis, observed results, the decision you made, and the lessons your team learned. Saving structured experiment records builds organizational memory and helps future teams avoid repeating errors. Fields marked with * are required.

A short unique identifier you control (e.g., EXP-2026-001).
Concise name that makes the experiment findable.
Who is accountable for running and reporting this experiment. Include name and role.
Write a measurable hypothesis (if <em>we do X</em>, then <em>Y will change by Z</em>). Include the expected direction and rationale.
List the specific metric(s) you will use to judge success and how they are measured (units, collection frequency).
Current value of the primary metric(s), if known. Include units.
Minimum practical change you want to detect (same units as metric). Useful for power planning.
Describe who or what is included (segments, geographic scope, inclusion/exclusion criteria).
Use ISO date format YYYY-MM-DD. If unknown, estimate.
Use ISO date format YYYY-MM-DD. If unknown, estimate.
Recorded when experiment began (YYYY-MM-DD).
Recorded when experiment ended (YYYY-MM-DD).
Record planned and, later, actual sample sizes. Helpful for interpreting power and reliability.
How treatment and control were assigned (e.g., true randomization, alternating, cluster, convenience).
Describe the statistical tests or comparison method, how you handle outliers, and any subgroup analyses you plan. Pre-specifying reduces bias.
Give the key findings for the primary metrics (direction, magnitude, confidence intervals or p-values if available). Use concrete numbers and units.
Optional: paste a small table or plain-text summary showing control vs treatment values, sample counts, and key statistics.
The explicit decision reached based on results.
Rate how confident the team is that the decision reflects reality (consider statistical power, bias risks, and operational factors).
1.0 10.0
What went well, what went poorly, and what you would change next time. Be concrete so others can apply these lessons.
If adopting or iterating, record the immediate actions, owners, and timelines.
Paste links to dashboards, analysis notebooks, shared drives, or ticket IDs where full data and code live.
Advice for people who build on this experiment (pitfalls to avoid, context that matters).
You can explore this tool now. Sign in or create an account to save your responses and return to them later.
Sign In to Save Create an Account
Make this tool part of your work

Save a personal copy, bring it to your team, or tailor the questions and workflow to fit what you are hungry to improve.

Member customization and team collaboration are coming soon.
Discuss Ask the Librarian

Discussion

Comments and conversation will live here.