Historical Information Review

Cost technique · See it on the map

Runnable here

Pulling actual cost, schedule, and estimating data from comparable past projects to sanity-check or ground a current estimate, instead of estimating from a blank page.

When to use it

At the start of estimating, when comparable past-project data exists and is trustworthy enough to compare against — most useful for an organization that has done similar work before and kept its actuals.

When to avoid it

Don't lean on historical data from a project that isn't actually comparable in scope, market conditions, or team, just because it's the data on hand — a plausible-looking number from the wrong reference project is worse than an honest estimate built from scratch. And don't use it as the sole basis for an estimate when the current project has no genuine precedent.

Steps

1. Identify past projects genuinely comparable in scope, scale, and conditions — not just the same general category of work.
2. Pull their actual costs, durations, and any documented estimate-versus-actual variance, not just their original estimates.
3. Adjust for known differences: inflation, site conditions, team experience, scope delta between the past project and this one.
4. Use the adjusted figures to check the current estimate, not to replace the estimating process entirely.
5. Record which historical projects were used and what adjustments were made, so the reasoning is auditable later.

What it produces

• A sanity-checked or adjusted estimate, with its historical basis and adjustments documented.
• A short list of comparable past projects usable again for future estimates.

Common pitfalls

• Using a past project as a reference because its data is convenient to find, not because it's actually comparable.
• Carrying forward a past project's own estimating error because only its original estimate was reviewed, not its actual outcome.
• Skipping the adjustment step and applying historical figures unchanged to a project with materially different conditions.

Worked example

A contractor estimating a 40-unit apartment renovation pulls actuals from a comparable 36-unit renovation completed eighteen months earlier: $1.42 million actual against a $1.30 million original estimate, a 9% overrun traced mostly to unforeseen electrical work. Scaling for unit count and adjusting for a documented 4% material cost increase since then, the estimator lands on $1.68 million for the new project and adds explicit electrical contingency, rather than repeating the prior project's underestimate.

Source

• PMBOK-6 §7.3.2.4

Where it comes from: this technique is named by the PMBOK Guide, 6th edition.