A decision with the evidence attached.
Gather what is actually known, weigh how good it is, and set out a recommendation you can defend.
Useful when a decision keeps being deferred because nobody can say what the evidence really supports.
Illustrative system example · fictional product and data · not client work
Evidence workspaceQUORUM · illustrative system, fictional data
You may need this when
Some of this probably sounds familiar.
- The same decision has been discussed three times without moving.
- Two credible sources say opposite things and nobody has reconciled them.
- A supplier's numbers are the only evidence you have.
- Someone needs to sign this off and wants to know what it rests on.
- You need to know whether an idea is worth building before building it.
What we could build
Evidence assembled so a decision can be made.
Including the honest answer when the evidence does not support a confident recommendation.
- Options appraisalTwo or three real routes, compared on what matters to you.
- Evidence reviewWhat is known, how strongly, and where it conflicts.
- Feasibility studyWhether the thing can be built, and what it would take.
- Market or competitor analysisWhat already exists and where the gap actually is.
- Cost and benefit modellingThe numbers behind a business case, with assumptions visible.
- Written recommendationA decision paper someone can take to a board.
See an example
How much of the conclusion rests on one source?
Change which evidence is admitted. The recommendation moves — because a conclusion is only as strong as what it stands on.
Every option is shown below. The selector needs JavaScript, so it is not offered here.
OutcomeWith everything admitted the case looks positive, but one source is the supplier's own.
OutcomeRemove the supplier's own projection and the confidence drops — the case largely rested on it.
OutcomeThe honest answer is not yet: two material conflicts should be settled before committing.
Illustrative interfaces and fictional data. Not client work.
Showing what the recommendation depends on is what lets somebody disagree with it properly, rather than just distrusting it.
What you receive
Something you can put in front of a board.
- A written recommendation with the reasoning set out
- The evidence base, with the quality of each source noted
- Assumptions and limitations stated rather than buried
- A clear statement where the evidence does not support a conclusion
- Agreed source code, data and infrastructure handed over according to the project scope and terms
- Handover documentation and any agreed support arrangements
Technical depth
Sourcing
Primary sources preferred, with provenance recorded so a claim can be re-checked later.
Appraisal
Each source rated on independence, recency and method, and weighted accordingly rather than counted equally.
Conflict handling
Disagreements between sources surfaced and reconciled explicitly, not averaged away.
Method
The analytical approach chosen for the question and stated up front, so the result can be challenged on its method.
Uncertainty
Confidence expressed in plain language, with the conditions that would change the conclusion named.
Reproducibility
Working, data and calculations kept so the analysis can be re-run when circumstances change.
Often combined with
What usually sits next to this.
- Simulation & modelling →When the question is what happens if you change something.
- Data & analytics →When the evidence needs building from your own records first.
- Forecasting & predictive modelling →When the decision depends on what happens next.
Tell us the decision you keep deferring.
Describe what you would need to know to make it with confidence.