We wanted to set up a small team who’d look after the areas that overlapped across projects, the common code, framework, that kind of thing. So we started experimenting with using index cards - we just had little summaries on them and we tried to keep them as small as possible, but they were still very big for what I’d call a task now.
That was working quite well and then a project came along which was to deliver an upgraded version of some software we were already selling to a big pharmaceutical supplier. So we decided we were going to do at iterations. We got somebody who was a customer-dude (and I got the Marketing Manager to do it) and that worked quite well and we spoke to him regularly. He was not always changing his mind, but things were moving and evolving very quickly and we were trying to keep up with it. It basically became too much, so we realised that we had to do the iteration thing properly - where we would listen to him all the time but we would actually only change direction every periodic bit and get him to accept that. I think we were doing three week iterations. For the next the next three weeks we’ll do this, and we continued to discuss with him what he was going to want next, but then we wouldn’t alter what we were doing until the next iteration. That worked quite well.
We did that for a bit and we had some mixed success, we managed to introduce loads and loads and loads of bugs and other quality issues. I mean it was very nearly, “Stop what you’re doing and let’s go back to the way it was”. Then I got [R] and [D] in to do some training for us.
It was very successful from our point of view, but I think the executive team shat themselves at the money involved - at the time I didn’t have the words and arguments to support all this stuff. The main problem was that they were seeing it as training and paying x pounds per day, it didn’t make sense to them. Eventually, about a year later, I managed to explain it in terms of “Well, we can send someone on a course for five days, which will cost £5000 or we can spend the same £5000 having someone sit with that person two days a week for six months” and then the penny dropped.
At this point basically we were trying to do stories but nobody was very good at them. The usual story is like this: we’re trying to do acceptance tests, but nobody was very good at them. We were doing iterations, but people were always slipping. There were problems with estimates, everything else. But by and large the projects were coming out and the reason for that is because the focus had changed from having a fixed amount to do, and the time it’s going to take to finish it, to changing all the time…
So from the business point of view, to use the technical jargon, we had risk control. So we were able to release whenever we wanted, it was just a question of maybe not having all ten features in, might only have seven.
delicious
digg