Experimentation on Trial

Twice in two weeks, an engineering manager asked me for evidence that mobbing and pairing work. Both were worried about their team’s productivity. I heard the real request: show me the data before I allow my team’s choice.

Making a team convince a gatekeeper treats software development like a courtroom drama. Must the team argue its case for the right to try something new? Why, counselor, does the burden of proof sit in front of a team’s experiment?

What’s the downside of “that didn’t work for us”? What’s the upside of a team that keeps looking for better ways to work?

Demanding proof first is a sign of a pathological or bureaucratic culture, in Westrum’s terms. Commanders be commandin’. A manager in a generative culture encourages experiments and lets the team choose. That’s the culture we build in the dojo, and it’s why mob programming gets a fair trial there.

Remember the first line of the Manifesto for Agile Software Development:

We are uncovering better ways of developing software by doing it and helping others do it.

Finding the techniques that help a team perform means giving it room to experiment, learn, and sometimes fail. Stop using research as an obstacle to a team’s growth. Your people are intelligent adults. They don’t need an expert to pre-validate their experiments. That’s not how experiments work.

The defense rests.