Why “best practice” must die
4 February 09
Anyone who’s worked in the web is aware of the “best practice” cult. To me, it’s a lazy creed that exhorts us to switch off and plunder others’ work, and the time has come to rebel.
Firstly, there’s the pure language involved. “Best” implies something that cannot be improved upon. A world of best practice gives us creationism, chariots, and gramophones. It negates progress.
There’s also a more sinister side, which is when it’s wheeled out as an argument in design projects that are heading off the rails:
“Ah, but that’s not how eBay do it”.
The unspoken implication is that eBay know better than I, and therefore I should defer to their wisdom. It’s an argument that I find misguided more than insulting. The web runs on the basis of meritocracy in a way that many other industries cannot. “That’s not how eBay do it” is industrial, corporate thinking, entirely irrelevant to the 21st century. For the truth is that large companies often don’t have a clue about design. One’s skill and knowledge are entirely independent of the size of your employer: I’m confident I know as much about my profession as the employees at any large company.
The best practice trump card also fails because it doesn’t understand the nature of practical design. It’s not a transferable commodity: you can’t just screw a design solution into place. Good design must be appropriate and relevant to the particular problem. The factors involved—technological, strategic, sociological—are far too complex and variable for a plug and play approach. To say “Well, a dropdown worked here…” is to ignore factors that can actually work in your favour. A company that rejects the easy route and takes the time to understand technology, strategy and users can offer designs that makes it stand out from the rest.
I’m not advocating isolating oneself from the surrounding environment. For instance, at Clearleft, we regularly perform competitor analysis at the start of a project. It’s useful to see where others’ strengths and weaknesses lie, and helps us understand the landscape. However, not once has it given me the answer to a design problem. That always comes later, with thought, with detail, and after many failed attempts.
So let’s not allow the enforced limitation and unvoiced threats of “best practice” to pollute our thinking. It’s harder work, sure, but standing out and being better always is.
Subscribe via RSS
Follow me on Twitter
8 comments on Why “best practice” must die
Interesting post Cennydd, I have a similar belief regarding “UX Patterns”, I’ll ping you once I’ve fleshed it out.
Thought provoking post! I agree we should certainly try to break the mould rather than just blindly following suit but there are some valuable conventions that people expect and are comfortable with that we shouldn’t just throw away.
This would make a great lunch time discussion topic for the IA Summit (if not a fully fledged presentation).
Des – oh, totally. Pattern libraries prioritise the artefact over the thought. Therein lies madness. That said, yes Amanda, there are definitely some conventions worth sticking to. Some existing de facto practice (eg. shopping cart functionality) is sensible and I would need a damn good reason to vary from it.
I suppose it’s more the label and the blind application of so-called solutions I’m rejecting.
I’d swear that no-one vists any other site than the BBC when i ask at the start of a project, ‘so what kind of sites do you like’.
Maybe it’s too open ended a question.
I’ve come across a variant of this phrase in the procurement of ‘IT Solutions’ which is best of breed. Strangely, more prescient than the originators of the phrase realise since they almost always buy ‘dogs’.
To my mind, “best practice” means delivering the best possible result, to the highest accepted standards feasible within the scope of the project.
I’ve not personally come across it being used as some pseudo-political argument for cutting corners, as your post suggests to me. I’d even go as far as to suggest that people using the term in this way don’t actually give a crap about “best” anything.
Given the presented context, “best practice” here sounds like hateful business speak for “acceptable minimum effort”. And to me this is entirely counterpoint to what “best practice” means.
For me, best practice is about doing your research thoroughly; it’s about getting your project spec to be as comprehensive and considered as possible; it’s about ensuring the end result is the very best it could be, given the scope of the project and the anticipated outcome.
I agree, that’s what it should mean. But yes, for whatever reason, it seems to regularly get misused as shorthand for “Facebook use dropdowns for this; let’s do that too.” It’s even more invidious when applied to practice that is clearly anything but best (e.g. the password antipattern).
I’m extremely lucky to have great clients who don’t think this way, but I still see this definition cropping up regularly. Oddly, there seems to be a cultural component to this too; in my experience (and confirmed by others’) some nations are much more slavish in their plagiarism than others.
I finally got around to explaining why I think patterns are a bad crutch that prevent innovation/creativity.
http://www.contrast.ie/blog/thinking-in-patterns/ if you’re interested.
Great post – and I couldn’t agree more. At present, I hear the term bandied around far too much and I’m starting to develop an allergy to it in similar fashion to when I hear the word “expert” use. I think both words / phrases as you rightly said suggest that something has reached it’s absolute potential.
The term “best practice” is only appropriate for the most refined and effective practice of any particular discipline the moment right before the world ends – anything else, as you say, is simply good practice.
Add a comment