Key Assumptions Check

Description

Systematically lists the assumptions underpinning a judgement, rates each as solid, caveated or unsupported, and identifies what would invalidate them.

Every analysis rests on assumptions that are rarely stated. The key assumptions check, done at the start and end of a project, asks the team to list working assumptions, then for each ask: why do we believe this? what could make it untrue? has it been true in the past? if wrong, would it change our conclusion? Assumptions are rated as solid, correct with caveats, or unsupported, and unsupported key assumptions become research questions or monitoring indicators. In product research, this exposes taken-for-granted beliefs about users, markets and technology before they shape study design and conclusions.

Attributes

Loss type
Unmigrated
Stage
Strategy
Effort
Hours

Steps

  1. Gather the team and state the current judgement or plan (e.g. product strategy, study conclusions).
  2. List all assumptions, stated and unstated, using prompts ('what must be true for this to work?').
  3. For each, ask why it is believed, under what conditions it fails, and whether it has held historically.
  4. Rate each as solid, caveated or unsupported, and as key or non-key for the conclusion.
  5. Convert key unsupported assumptions into research questions, tests or monitoring indicators.
  6. Repeat the check at the end of the analysis and when new evidence arrives.

Sources

  • Heuer & Pherson, Structured Analytic Techniques for Intelligence Analysis (2nd ed., 2014)
  • US Government, A Tradecraft Primer: Structured Analytic Techniques for Improving Intelligence Analysis (2009)

Also known as

  • KAC
  • Assumptions audit (intelligence)

Categories

  • Research Operations
  • Research Design & Rigor