THE SHORT ANSWER
Ask what must be true for the conclusion or plan to hold, which belief has the weakest evidence, what would prove it wrong, and what alternative explanation fits the same facts. Then test the assumption whose failure would matter most.
Turn hidden conditions into visible statements
- What are we treating as known?
- What must be true for this plan to work?
- Which assumption has the least evidence?
- What are we treating as fact that may be an inference?
- What observation would make us change our view?
A launch plan may assume customers understand the offer, the team can support demand and one channel can reach the intended audience. Naming these conditions makes them testable.
Test consequence as well as confidence
| Question | Purpose |
|---|---|
| How uncertain is it? | Avoid spending effort proving what is already well supported |
| What if it is wrong? | Find the assumptions with material consequences |
| How cheaply can we test it? | Choose a proportionate next step |
| Who sees it differently? | Expose missing experience or incentives |
Evidence & context: HM Treasury and Government Finance Function
Challenge the claim without attacking the person
Replace ‘That will never work’ with ‘Which condition is this plan relying on, and how could we test it?’ The aim is not permanent doubt. It is to make the reasoning answerable to evidence.
For deeper evidence evaluation, continue to How to Evaluate Evidence and Facts, Opinions, Assumptions and Inferences.
Evidence & context: National Academies of Sciences, Engineering, and Medicine
Write a precondition map
Take one proposed decision and complete: ‘This works only if…’ five times. Mark each statement as observed, inferred or unknown. Select one high-consequence unknown and define evidence that would increase or reduce confidence.
Sources & further reading
- The Green Book 2026
HM Treasury and Government Finance Function. Current official guidance on objectives, options, evidence, risk, uncertainty, appraisal and evaluation. It is designed for public decisions, while the general reasoning principles are adapted here.
- Reproducibility and Replicability in Science
National Academies of Sciences, Engineering, and Medicine. A consensus report on evidence, uncertainty, transparency, reproducibility and replication. Scientific standards need proportionate adaptation outside research settings.
Examples and exercises are illustrative unless attributed to a source. No independent expert review is claimed.
A correction, a counterexample or an experience worth sharing?
Join the conversation ↗