We have found that many organizations are struggling with predictability. One reason is that they don’t use common sense when estimating a project’s timeline. The Lite Schedule Estimating Matrix is a tool that helps to estimate the amount of time a project will take in any given phase. It leverages past experience combined with the critical few, key drivers that impact a project’s schedule. The matrix, which consists of five to eight inputs per phase, predicts the duration of each phase based on the range of typical times and the technical complexity of the design. An alternative to the time consuming, bottom-up Gantt chart, the tool helps inexperienced program managers become more successful.
The tool has a number of benefits, such as:
- Improving the predictability of schedules as organizations learn from their past experience
- Reducing conflict between the team and management
- Capturing institutional knowledge, regardless of turnover
- Establishing a common language and a consistent visual tool for estimating schedules, useful for the team and management alike
To construct this spreadsheet-based tool, place the phases of your project and the key drivers for each phase in the first column. Determine these key drivers by tapping the latent knowledge in your organization. Ask your most experienced project managers in particular. Examples of these drivers for a software project might include the number of screens/pages, the number of interfaces, the complexity of system integration, the percentage of testing that is automated, or the number of geographies involved.
The second and third columns represent the range (in numbers or in percentage) associated with each of the key drivers based on past experience. Place the lowest figure in the range in the second column and the higher figure in the third column. The fourth column is your estimate of the complexity of each portion of the project. Leave the fifth column open for comments. After entering all the estimates for the key drivers, the project manager estimates the duration of the phase as a whole.
The program manager can fill in this tool alone. However, the results will be more accurate if the cross-functional team works together to derive the estimates.
In the case of very complex systems, it is possible to break the estimate into sections for software, hardware, accessories, etc. After calculating the individual estimates, the program managers for the subsections should provide an integrated estimate, taking advantage of the parallelism available in most programs.
Products and services are developed more rapidly now than at any other time in history. For example, Web 2.0 development and smart phone applications can have releases every week!
With this fast release cycle, there is an opportunity to improve schedule estimation because the cycles of learning are so short. In general, organizations need to adapt and move more quickly, as the world has become much less forgiving when schedules fall behind.
Beyond lagging schedules, extended enterprise design and development require synchronization of disparate groups. With the rising standards in many areas, predictability is very important.
What Else You Should Know
We have found several factors that can limit the accuracy of this method. If the design paradigm changes, you may have incorrect or insufficient baselines. For example, if you have just started to implement agile software methodology, then your underlying process has changed dramatically and you will need to collect a new set of data from projects that used your new approach.
Another factor is that your estimate is subjective, so its quality is dependent on the team’s ability to make good inferences. However, in our experience, this tool is almost twice as accurate as an average program manager, though less accurate than your industry’s most capable program manager. An alternative to this tool is the Precise Schedule Estimating Matrix, which provides a higher level of schedule accuracy, but requires greater effort to implement.