Duration Variance Percentage
This guideline describes how to measure Duration Variance Percentage to evaluate delivery time performance of completed projects.
Main Description

Overview

The Duration Variance Percentage metric is used for measure project duration (or delivery time) performance with respect to project duration estimates. This is frequently used to evaluate schedule management performance for both individual projects as well as project portfolios.

Measurement Method

Cost Variance = (Planned Project Duration - Actual Project Duration) / Planned Project Duration - where

Planned Project Duration = Original estimated project duration for a specified completed project, defined as the planned elapsed time period from the standard project start milestone (e.g., project kickoff date) to the standard project completion milestone (e.g., application deployment date). Duration can be quantified in terms of a common unit of time (e.g., days, weeks, months etc.).

Actual Project Duration = Actual project duration for a specified completed project, defined as the actual elapsed time period from the standard project start milestone to the standard project completion milestone. Duration must be quantified in the same units as Planned Project Duration, above.

The same formula can be used to evaluate multiple projects within a project portfolio for a defined evaluation period through histograms or scatter distribution charts.

Measurement Analysis

Duration Variance Percentage indicates how accurately projects are able to estimate actual delivery time, or alternatively, how well projects are able to deliver their defined project objectives within their original planned duration. Rendered as a percentage, this metric facilitates trend charting and comparisons from period to period to determine schedule management performance over time.

Duration Variance Percentage should fall between optimal values, frequently ranging from +/- 5% to +/- 30%. However, the optimal range is dependent on the type of project(s) under evaluation. For example, highly predictable projects, such as maintenance or small enhancement projects, require a narrower range of evaluation (say, +/- 5%), whereas highly risky projects require a wider range of evaluation (say, +/- 30%). Additional ranges may be used to establish tiered evaluation to characterize low, moderate, high variance conditions.

Upon establishing the evaluation ranges, if a project falls within the optimal range, the project is evaluated with optimal delivery time performance.

Otherwise, the project is evaluated at a lower level of delivery time performance. Alternatively, a multi-period histogram as below can classify portions of project portfolios by Duration Variance Percentage. This allows evaluation of portfolio performance from period to period.

Multi-period histogram classifying project portfolios by Duration Variance Percentage