Program Evaluation and Review Technique

Methods and techniques | trial

PERT (Program Evaluation and Review Technique) is utilized to assess the time required to complete a project. We use PERT during the project or feature planning phase to gain an understanding of the anticipated time and budget.

This technique involves three different time estimates for each project task.

Optimistic time The shortest time needed to complete a given task.

Pessimistic time The longest duration the task might take if complications arise.

Most likely The most probable amount of time required to complete the task.

In our experience, using PERT significantly improved the accuracy of our estimations and enabled us to identify potential risks. As the team engages with the pessimistic time, discussions regarding potential complications and uncertanties arise, which are then recognized as risks. Our approach involves frequent monitoring and thorough evaluation of these identified risks.

Here's what we appreciate about PERT.

Probabilistic thinking Instead of a single number, we arrive at a few estimates that come with respective confidence levels.

Faster estimation framework PERT helps us making estimation quicker, epecially after we simplified it even further by including Fibonacci sequence values—as hours or days—to limit the number of possible time estimates to choose from.

Dealing with uncertainty By considering three scenarios—pessimistic, optimistic, and most-likely—for each item, we're able to better accomodate for uncertainty. An additional benefit is that large differences between optimistic and pestimistic estimates may suggest unclear scope or potential technical challenges, so we're able to detect these issues early.

While the math behind PERT may seem complex, the process itself is quite straightforward. It requires breaking the project up into smaller items, and providing three estimates for each of the items. These estimates are then passed to a set of pre-defined equations to produce the final result. It's important to note that teams should feel free to experiment with different estimation techniques and processes to find what works best for them. The key is to remain open to learning and refining the process as the team gains experience and a better understanding of their unique needs.