It has long been a sacred cow in agile circles that teams that are not co-located are not as productive as co-located teams. For me, the recent COVID-19 pandemic is a rare opportunity to test this. This will be a 2 part blog post. Firstly, I’ll explain the methodology I’ll use. The second part will be the results, which I’ll post once sufficient data is available.
As an Agile Coach and “statistics enthusiast”, I set out to collect some data to test my hypothesis below.
To get started I took a metrics framework that we used across a number of our teams at Trayport. Our framework is based on the work of Troy Magennis, Larry Maccherone, and others. The framework we use covers 6 areas and each team defines their own metrics within these areas. Across teams, we focus on trends rather than comparisons.
- Quality – Are we delivering quality products to our customers?
- Productivity – Are we working efficiently?
- Responsiveness – Are we able to change direction at short notice?
- Predictability – Are we able to say reliably how long we will take?
- Right Things – Are we working towards the right outcomes?
- Resilience – How sustainable is our performance?
The reason we have 6 areas is that we want to avoid improving any area at the expense of others. So we keep an eye on all of them even when we are only trying to improve a specific metric area.
My intention is to compare the % change across each of these metrics at the team level. Something like the following graph (which is based on made up data).
What's The Next Step
- Keep observing changes in team performance. I expect this experiment will run for another 1-2 months before we have sufficient confidence to reach a conclusion. Ultimately, draw a conclusion on whether the hypothesis holds true or not.
- Based on that conclusion, think about how we could / should / would like to change our ways of working.
“”Damien O'ConnorHead of Agile Delivery