Sweat More

From WeWillTry:

Management doesn't want to hear the truth. They just want to get as much effort as possible from the developers. (Well, then they should be happy: the software will be done when it's done, though treating the developers badly might get less output than treating them well.)

But getting the software done is not the same as getting as much effort as possible out of the developers.

For example, at the end of the first iteration of the project described in AnotherXpExperiment I (one of the team leaders) had a debrief with the project manager. We had (partly, I'm sure, because of our adoption of some of the XP practices) delivered on time while working normal hours. And the client's QA people were delighted with the product. The PM's response to this was to ask if the estimates had been padded, since we hadn't been working overtime. And, further, he said that he would have been happier if he'd seen us "sweating" a little.

Of course he would - management loves when they can squeeze a little more out of the staff by threats and carrots, and if you are not pressed for time, you start asking for real carrots instead of fake ones, and start treating the stick like a resume call. Also, remember that visible heroics get rewarded, while invisible excellence does not - who gets the kudos at the end of a DeathMarch - the folks who got all their work done on time, or the heroes who worked long hours to fix their buggy stuff?

Not only was the software done, it was done well, and it was done when we said it would be and that wasn't enough. It wasn't enough to get on this guy's good-books.

Getting the most from developers would seem to mean, sometimes, treating developers badly, sweating them, making sure that a huge effort is seen to be made. The worst outcome in this scenario is a heroic failure, and if that happens, well that's almost a success, because how could they have done more? In this way, the project manager is unable to fail in the eyes of their superiors.

I've no particular desire to be seen as a cynic (although no fear of that either) but, unless I've misunderstood, the response in italics at the head of the page is far too naive. -- KeithBraithwaite

You might want to try working overtime for about four straight weeks while tracking your progress. I'll bet you get less done, with lower quality. Then take that data to your manager and ask him which way he wants you to work. If he'd rather have the overtime and lower performance, well, you both have decisions to make.

Of course I'd get less done, and of course the quality will be lower. Some managers will interpret this as inefficiency (since you are doing all these hours you should be getting more done, right?). Mere numbers will not sway them. These results are so well known it beggars belief that any manager doesn't know them, but still those guys exist.

And, in fact, almost every project that gets into trouble with timescales and that does any tracking at all does exactly gather this information: estimates are not met, leading to longer re-estimates, that aren't met, and so on and so on. I've seen this several times, I'm sure you have too. Or have all your projects always been so rationally run that this hasn't occurred, whoever you are?

I personally have refused to work in this kind of regime, and will refuse to do so again. -- KB