Tool

Hypothesis Tracker & Evidence Log

A lightweight, structured form to record hypotheses, assumptions, acceptance criteria, experiment plans, evidence, decisions, and retrospective lessons. Designed so teams can find what was tried, why, and what was learned.

Interactive Tool

Hypothesis Tracker & Evidence Log

This lightweight tracker helps teams keep experiments honest and build an accessible record of hypotheses, tests, evidence, and decisions. Use it to reduce knowledge loss, avoid repeating failed experiments, and speed from insight to action.

Tip: Make each hypothesis specific and testable, tie it to acceptance criteria, link the experiment plan, and update evidence as results arrive.

A short, memorable title for this hypothesis (e.g., 'Faster replies → better retention').
State what you expect to happen and under what conditions. Avoid vague goals—make it falsifiable. Example: 'If we reduce customer email response time to under 2 hours, weekly retention will increase by >= 5%.'
List assumptions you are making that must hold true for the hypothesis to be valid (e.g., customers read emails, product changes are visible, baseline retention is stable).
What specific measurements will show the hypothesis is supported? Include metric names, thresholds, and measurement windows.
Optional: an estimated numeric change (e.g., +5 for percent points). Use this to prioritize and set sample-size expectations.
Link to the experiment plan, runbook, ticket, document, or repo where the test is described. Use an internal URL or identifier when possible.
Current lifecycle state of the experiment. Update as work progresses.
Optional date text (YYYY-MM-DD) for planning or record-keeping.
Optional date text (YYYY-MM-DD) when the experiment finished or is expected to finish.
Summarize the results, observations, and data collected. Be specific about metrics, cohorts, time windows, and significant anomalies. Link to analysis artifacts when possible. Update this field as evidence accumulates.
Paste URLs or identifiers for dashboards, queries, notebooks, spreadsheets, logs, or export files that contain the raw or processed evidence.
Select 'yes' if results contradict the hypothesis, 'no' if results support it, or 'unknown' if evidence is inconclusive.
Rate your confidence in the evidence and analysis (0 = no confidence, 10 = very confident). Consider sample size, noise, and measurement quality.
1.0 10.0
Record the team's decision after reviewing evidence. Add context in retrospective notes.
Who will do what next? Include owners, deadlines, and required follow-up analysis or implementation tasks.
Capture what was learned about the hypothesis, measurements, experiment design, tooling, and any process improvements to avoid repeating mistakes.
Comma-separated tags to make it easier to search and group entries (e.g., onboarding, churn, pricing).
Optional names or identifiers of related workstreams, OKRs, or product areas.
Optional: call out specific places reviewers should focus (assumptions, metrics, data quality concerns).
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.