when will it be done? is a question every team that’s building software has to answer at some point.
Customers or business stakeholders want to know when a specific set of functionality will be ready or they want to know what scope of functionality they can get by a given date.
In my experience the question ‘when will it be done?’ results in an estimation meeting that mostly involves doing story point estimates. No one likes these meetings and no one really believes in the estimated release date. That’s why I got really interested when I learned about the books Daniel Vacanti wrote on the subject of predictability with the hope to learn about better ways. The books didn’t disappoint.
After sharing a summary of what I learned from Daniel’s book with the Team Leads at In The Pocket about a year ago and indicating that I was going to start trying things out with my team a community of practice formed. We started exposing flow metrics, sharing them with our teams and started experimenting by putting into practice what we learned from the books which resulted in positive outcomes: finishing work items sooner and stabilizing the system so we became more predictable. Once we became more predictable we also started experimenting with forecasting using Monte Carlo simulation.
At one point Frederik De Bosschere raised the idea to interview Daniel Vacanti for Studio Shift and that’s what I ended up doing. I interviewed Daniel Vacanti but wasn’t available during the live show on the 20th of October 2021 and so two great colleagues both involved in the predictability community of practice at In The Pocket, Thijs Morlion and Quentin Braet, gave some additional context on parts of the recorded interview during the live show. You can listen to the recording of the show where we talk about work item age, continuous forecasting, Monte Carlo simulations, story points and probabilistic forecasting (Daniel Vacanti and myself in English, context by Thijs and Quentin in Dutch):
Of course the interview only covers a very small amount and some key points on the topic of predictability. If this sparked your interest I recommend you to check out the books. Daniel recommends reading his 2nd book first: When will it be done? Lean-Agile Forecasting to answer your customers’ most important question.
Next to the books I can also recommend the Drunk Agile Youtube channel where Daniel Vacanti and Pradeep Singh talk about agile software development including predictability and forecasting. Drunk Agile can also be found on SoundCloud.