Retrospective: The Clue is in the Name

I facilitated our regular end-of-iteration retrospective last week, and although the feedback from the team was positive, I was left with a feeling that something wasn’t right.

With our second major live release looming large on the horizon, I focussed the session on the theme of ‘Success’. My aim was to give the team a blueprint for a successful iteration to keep in mind when things were tough, and to help ensure that we were all pulling in the same direction by agreeing as a team what constitutes success for us.

I started by asking the team to score how successful, in their own mind, the last iteration had been, drawing a histogram on the flipchart of the results. Interestingly, there was quite a variance in the scores, with most people either grading the iteration 2/5 or 4/5, perhaps suggesting that different members of the team have different criteria for success, or that they just experienced the iteration differently.

I then asked them to reflect on the criteria they’d used to make that judgement, again in their own mind, capturing each of the criteria on a separate index card. We then spread the cards out on the floor, and mapped them into headings: teamwork, stability, delivery, happiness & joy… and a couple of others I can’t remember right now.

We then decided that our output would be some kind of vision statement under each heading, of the form “An iteration is a success when…” and worked through the various headings, discussing what it was that made us successful, and distilling that discussion into a bullet point on the chart.

Reflecting on what we’d covered, it seems to me that while the session was fun, motivating and affirming for the team, and thus valuable in it’s own way, it didn’t really enable us to focus on the major issues that are affecting us right now, such as the disruption of the recent office move. It also didn’t leave us with any concrete actions to take away and do in order to make things better. In terms of inspect and adapt, it really didn’t deliver.

When planning your retrospectives always remember your key goals (in priority order):

  1. Try to smoke out the most significant areas of waste in the team’s production of quality software
  2. Invent concrete actions you can take in order to reduce this waste
  3. Build and motivate the team

Whilst building and motivating the team is an important goal for a session like this, it’s often a side-effect of successfully satisfying the first two. Concentrate on facilitating a session that helps reduce waste while having fun, and you’ve hit the nail on the head.