When I first moved into a position of leading developers I experienced a fair amount of good natured teasing that I’d do my best programming via Outlook and Omnigraffle. While this has been true (thankfully only to a point!) it has been outweighed by the ability to have a far greater say into the resourcing, architecture and implementation of projects and how my company does its work.
One thing that I’ve become painfully aware of is that the operational aspect of this position has a larger ‘people management’ aspect to it than I had previously thought. This is especially true when you extend your view to include recruitment, appraisals, motivating staff and the like and not just setting up your teams to deliver quality code as fast as they can. The soft skills of leadership tend not to come easily to those with a natural technical focus (describing almost every developer ever) so it only makes sense to take advantage of psychology research in the areas of skills assessment and motivation. These is a wealth of knowledge out there and I thought I would discuss a few of the more interesting/useful ones that I have come across.
Judging Ability – Dunning-Kruger Effect, Impostor Syndrome and the Dreyfus Model of Skill Acquisition
In theory you should be able to assign any part of a project to any developer. The holy grail is that is shouldn’t matter which developer takes which task but we know that it does. When we can share the work around evenly it ensures that we share the project knowledge around the entire team and don’t create knowledge silos but the reality is that you need to make sure that the story is appropriate to the developers level of ability. There are some stories that can be done by anyone. There are also some stories that you know are a little more complicated and will have much greater knock on effects that you know should be done by more experienced developers and we know that that doesn’t always mean the senior developers.
Recruitment also requires you to judge ability very quickly and ideally very accurately. HR screening and blunt telephone tests may provide a simple barrier to the wildly inappropriate candidate but nothing will replace sitting down with the candidate and doing a face-to-face interview.
When you know the technical topic judging how good someone is is fairly simple and intuitive but judging peoples ability when you have no expertise in that area is a tough task. The people that are at extreme ends of ability are obviously set apart from the norm but those that make up the bulk of the bell curve can be difficult to separate. Some psychological concepts that can help are the Dunning Kruger effect and the Dreyfus Model of skill acquisition.
The Dunning Kruger effect deals with the fact that ignorant people rate themselves more able than they are because they don’t realise how truly terrible they really are. The obverse of this is the Impostor syndrome - highly knowledgeable people rate themselves lower than they are because they are painfully aware than they don’t know everything and they think that their success is due to luck rather than their own work. Both these effect highlight how dangerous it can be to ask for self-described levels of competence but if you do, always make sure to follow up with a few in-depth questions. If they’ve described themselves as top rank but can’t seem to answer anything in-depth then their lack of awareness is a big sign to their true abilities.
Another way to establish expertise is to look at how people behave as they work and see where this behaviour falls in the Dreyfus Model of Skill Acquisition. The Dreyfus Model was originally designed to establish whether Navy pilots were competent before they were put in a plane and essentially, it looks at the way someone solves a problem. While this is fairly common sense, the objective framework it provides makes it far easier to describe the level of someone’s abilities, how good is their situational awareness, how do they plan and prioritise and when do they become aware that something is a new situation or another instance of a previous problem? I’ve been using this for appraisals for the last few years, covering someone’s past performance and using it to set goals for the future and it’s worked exceptionally well.
Motivation – Maslow’s Hierarchy of Needs and Self-Determination Theory
So you’ve now got the right people in your teams – now how do you make sure that they can do their job with enjoyment and passion? Passion is a productivity multiplier and a team who believe in what they’re doing can achieve almost anything. This is where ideas like Maslow’s Hierarchy of Needs and Self-Determination Theory come in. Maslow’s work lists out our needs and which order they need to be fulfilled, normally by the classic pyramid image with physiological needs on the bottom and self-actualization at the top.It’s a fairly based model but it provides an objective framework to assess people’s wants against and to prioritise them.
Self-Determination theory is an even simpler model with motivation separated in extrinsic and intrinsic motivation and the intrinsic motivation further separated into competence, autonomy and relatedness/higher purpose. Daniel Pink covers this superbly in his book Drive! and has given an amazing talk that if you haven’t already seen, you absolutely need to find the 10:48 minutes to find out exactly why these factors are so important and why money as a motivational tool may not have the effect you think.
It’s interesting to note that at a highly regarded company like Etsy, all engineers have three common annual goals: speak at a conference, write a blog post and open source code. I love these since I can’t think of any more concrete expressions of competence, autonomy and a common higher goal.
The Hawthorn Effect
One more quick one. Having read these and thinking that you can make a change to the way you do things, you need to be aware that any resultant improvement may be down to the fact that the people involved are aware that that there is a change. This effect is because people can change their behaviour irrespective of whether the change is actually an improvement – it’s kind of a placebo effect for workers. Hopefully your change is real but you need to be aware that improvements may be driven by something other than your changes so stay sceptical and always look for this and other external factors.
Like anything based on phenomenology, your mileage will vary. This is an absolute. The point of this article is not to prescribe fool proof ways of dealing with people or how to manipulate your staff (manipulation is a short term strategy that will always burn you) but to make you aware that their has been a lot of investigation into the area and that you can take advantage of this to make both you and your teams perform better.