Start with the decision
Identify the uncertain quantities, factors and service assumptions rather than adding an unexplained confidence score. A reported decimal count says little about data quality. Begin with the aspects that could change feasibility or a comparative result.
Build the record
Record the source and basis of any input ranges. Keep measured variation, supplier estimates and scenario assumptions distinguishable. State the analysis method and avoid suggesting statistical confidence when only illustrative scenarios were tested. Include unknown dimensions that have no defensible range.
Check before acting
Compare outcomes across the agreed ranges or scenarios and show when the decision changes. Where options overlap or depend on missing information, describe the unresolved choice and next evidence request. Display rounding appropriate to the information and keep the detailed method available for a specialist review.
Worked example — illustrative
A fictional comparison depends strongly on how many deployments a retained system achieves. The team tests three clearly labelled utilisation scenarios rather than presenting a single exact saving. One changes the apparent ranking. The report explains that dependence and keeps future bookings separate from confirmed use.
Keep a useful handover
Close the “Compare uncertainty rather than displaying false precision” decision with a named owner and a dated evidence version. The next working record is “Build a lifecycle record”. Record whether that step is planned or confirmed; retain the prior decision if new information changes the instruction.
- Identify influential uncertainty.
- Document range sources.
- Label scenario versus statistical methods.
- Show changed conclusions.
- Avoid misleading precision.
