BudgetOverrun.com is an independent reference site. Not affiliated with any PM software vendor. Statistics sourced from published research and cited throughout.

Three-Point Estimation (PERT): Formula and Worked Example

Three-point estimation forces estimators to express the uncertainty in their estimate, not just the best-case number. It is the simplest probabilistic estimating technique and the foundation of PERT scheduling.

The PERT formula

Expected value E = (O + 4M + P) / 6
Standard deviation SD = (P - O) / 6

where O is the optimistic estimate (best case), M is the most likely estimate, and P is the pessimistic estimate (worst case). The PERT formula assumes a beta distribution, which puts more weight on the most-likely value than a simple arithmetic mean would.

A simpler triangular-distribution variant uses E = (O + M + P) / 3, which gives a slightly higher expected value because it weights the tail more.


Worked example

You are estimating the cost of building a custom data migration module. The team gives:

  • Optimistic (O): 30,000 USD (everything fits, no edge cases)
  • Most likely (M): 50,000 USD
  • Pessimistic (P): 120,000 USD (legacy schema is a mess, two months of cleanup)
E = (30,000 + 4 x 50,000 + 120,000) / 6 = 350,000 / 6 = 58,333 USD
SD = (120,000 - 30,000) / 6 = 15,000 USD

So the PERT expected cost is ~58,333 USD with one standard deviation of ~15,000 USD. A budget set at E + 1 SD = 73,333 USD gives roughly 84% probability of staying within budget; E + 2 SD = 88,333 USD gives roughly 97.5% probability.

Crucially, the most-likely estimate alone was 50,000 USD. PERT recommends 58,333 USD just to break even, and a confidence-weighted budget closer to 73,000 USD. A single-point budget at the most-likely value would have under-funded the work.


Sources

  • Project Management Institute (2017). A Guide to the Project Management Body of Knowledge (PMBOK Guide), 6th edition. PMI, Newtown Square PA. See section 6.4.2.4 on estimation.
  • Malcolm D.G., Roseboom J.H., Clark C.E., Fazar W. (1959). Application of a technique for research and development program evaluation. Operations Research 7(5).

Related

Updated 2026-05-11