As a project manager, the question you often hear from stakeholders is when will it be done? It’s an easy question to ask but much harder to answer with any level of certainty.

The traditional approach is to detail out all requirements and estimate each task in story points or hours/days, and then use that data to come up with the project timeline. The problem is that people, in general, are not very good at estimating how long some piece of work takes.

Fortunately, there is an alternative approach, which is based on looking at your historical data and extrapolating that to your future work.

Here’s a chart that shows how much work a team has completed per week during the last 3 months:

If team composition is the same for the remainder of the project, wouldn’t it be helpful to use this information to estimate how long it takes to complete the remaining work?

A data-driven approach looks at your historical throughput and makes a prediction based on that. Here’s an example of such a forecast using the Forecasting chart

This chart looks at the weekly history of the team’s throughput and creates three scenarios based on the distribution of the work completed per week:

• Optimistic: 80th percentile

• Most likely: 50th percentile

• Pessimistic: 20th percentile

The optimistic scenario expects that your team will complete more work than in 80% of the past weeks. In the most likely scenario, your team completes the equal amount as what is the median week. In the pessimistic scenario, the team completes only as much as 20% of the past weeks.

You can learn more about the Forecasting chart from this how-to guide.