Protocol

Build an outside-view reference class

Your plan is unique. Your kind of mistake often is not.

When it fits

A project estimate is being built mainly from the current team's internal plan and assumptions.

When to avoid it

Reference class forecasting is sensitive to class selection and data comparability; recent review work highlights unresolved methodological limitations.

Why it matters

Find a set of past projects that are comparable on the features that drive the outcome, then inspect the distribution of actual cost, duration or failure. Use that distribution as an outside-view anchor before adjusting for specific differences in the current case. Record the selection rule so the class cannot be quietly changed after seeing the answer.

Steps

The forecast target is defined consistently across cases.
Reference projects share the drivers relevant to that target.
Actual outcomes, not original plans, are used.
The inclusion rule is written before inspecting the desired percentile.
Case-specific adjustments are explicit and separately justified.

An example

Estimate a data-migration duration from comparable migrations with similar volume, interfaces and testing scope, not from the current task breakdown alone.

Check your result

The forecast can show both the historical outcome distribution and the explicit reasons for departing from it.

Keep this limit in mind

Reference class forecasting is sensitive to class selection and data comparability; recent review work highlights unresolved methodological limitations.

Connected ideas

Evidence and sources

Supports

Reference class forecasting uses outcomes from comparable past projects as an outside view, but recent review work highlights unresolved issues around reference-class selection and applicability.

A badly chosen class can give a confident but irrelevant base rate.

Reference class forecasting: promises, problems, and a research agenda moving forward · Systematic literature review and limitations

All sources (1)

Reference class forecasting: promises, problems, and a research agenda moving forward