← All free resources

Assumptions and Decision Log

Keep supporting and disconfirming evidence attached to a decision.

For: Sponsors, product owners, and working teams
Use when: When an important decision depends on uncertain information
Suggested time: 15 minutes · Version 1.0 · October 5, 2026

Use this guide to structure a discussion, then record your answers in the editable template or printable PDF. Use synthetic or authorized information. A named person owns each decision.

Separate what you know from what you believe

Record one consequential assumption per row. Make the test observable, and look for evidence that would overturn the assumption.

Decision and assumption
What are you deciding, and what must be true?
Supporting evidence
What observation supports that belief?
Disconfirming evidence
What would show the assumption is wrong?
Test and owner
What can you inspect or try next?
Decision and residual uncertainty
Continue, change, defer, or stop? What remains unknown?
Review date
When will you revisit it?
See a completed companion example

Illustrative responses for a fictional workflow. These are planning assumptions, not client results.

Decision and assumption
Pilot policy drafting; assume source lookup is a major contributor to handling time.
Supporting evidence
Analysts report repeated policy searches; timed baseline still needed.
Disconfirming evidence
Most handling time comes from missing context rather than lookup.
Test and owner
Service-desk lead times a consistent sample over two weeks.
Decision and residual uncertainty
Defer time-saving claim until baseline is measured; quality benefit remains a separate hypothesis.
Review date
At the baseline review, two weeks after kickoff.

Record the next decision.

What will you do next, who owns it, and when will you review the evidence? Keep unresolved assumptions visible.

Continue with these tools.

By Vic Akosile · Built Into Work. Use and adapt for your own work or your team’s work; credit the source when sharing adaptations. These tools are discussion aids, not validated assessments. Fictional examples and proposed targets do not establish readiness for real use.