I’ve long worried about “Best Practices”. Sure I usually play along at the time but lurking in the back of my mind, waiting for a suitable opportunity are two questions:
- Who decided this was best practice?
- Who says this practice can’t be bettered?
I was once told by someone from the oil industry that it was common for contracts to specify “best practice” should be used. But seldom was the actual practice specified. Instead each party to the contract would interpret best practice as they wished, until something went wrong. At that point, after an accident, after money was lost they would go to court and a judge would decide what was best practice.
Sure practice X might be the best know way of doing things at the moment but how much better could it be? By declaring something “best practice” you can be self limiting and potentially preventing innovation.
Now a piece in MIT Sloan Management Review (Why Best Practices Often Fall Short, Jérôme Barthélemy, February 2018) adds to the debate and highlights a few more problems.
Just for openers, sometimes people mistakenly identify the practice creating the benefits. Apparently some people looked at Pixar animation and decided that having rest rooms (toilets to us English speakers) in the centre of an office floor enhances creativity. They might do, but there is so much else happening at Pixar that moving all the toilets in your organization will probably make no difference at all.
But it is worse than that.
Adopting best practice from elsewhere does not mean it will be best practice in your environment but adopting that “best practice” will be disruptive. Think of all the money you will need to spend relocating the toilets, all the people who will be upset by a desk move they don’t want, all the lost productivity while the work is going on.
The author suggests that in some cases that disruption costs are so high the “best practice” will never cover the costs of the change. Organizations are better shunning the best practice and carrying on as they are. (ERP anyone?)
It gets worse.
There is risk in those best practices. Risk that they will cost more, risk that they won’t be implemented correctly and risk that they will backfire. What was best practice at one organization might not be best practice in yours. (Which might imply you need even more change, even more disruption at even more cost.)
In fact, some best practices – like stock options for executives – can go horrendously wrong and induce behaviours you most definitely don’t want.
So what is a poor company to do?
Well, the author suggests something that does work: copying good practices. Not best but “just OK”. That works. Copy the mundane stuff, the proven stuff. The costs and risks of a big change are avoided. (This sounds a bit like In Search of Mediocracy.)
In my world that means you want to be getting better at doing Agile instead of trying to leapfrog Agile and move to DevOps in one bound.
The author also suggests that where your competitive advantage is concerned keep your cards close to your chest. Do thinks yourself. Work out what your best practice is, work out how you can improve yourself.
I’ve long argued that I want teams to learn and learn for themselves rather than have change done to them. But I also want teams to steal. When they see other teams – at home or elsewhere – doing good things they should steal practices. The important thing from my point of view is for the teams to decide for themselves.
Sign-up to receive these posts by e-mail and free eBook of Xanpan
I’m surprised you didn’t bring Cynefin up
I once worked on a project that was let out on contract from the U.S. government. We were obsessed with best practices for two reasons:
1. The government doesn’t actually know how to make software, but is also scared to death of being gypped by their contractors. So they adhere to best practices as a religion, because they think it saves them from peril.
2. Companies that do government contracting very often are in the “government contracting” business more than they are in the “software development” business and know as much about delivering software as they do about taming wild bears or making delicious pate de foie gras. They cling to best practices as if they were a how-to manual. Do these things and you’ll have success!
The amusing thing is that government contracts usually come with something called Independent Verification and Validation, or IV&V. This is a government-contracting company hired to make sure the government-contracting company building the software is adhering to best practices. Hilarity ensues when IV&V disagrees with the company doing the work on what the best practices actually are. There are so many to choose from, after all.