Much Ado About Performance

Posted by in Freelancing

I think I may have finally read one too many posts on “how-to-make-your-WordPress-site-faster”.

It is true that the monotonous litanies, their vaguely suspect tone of link-bait, have had at least one positive side-effect on me: they’ve convinced me that nothing beats looking at your code with the umost attention, and reading whichever documentation is available. The other side-effect being, of course, to stop reading embarrassing fluff; how many more times do I need to read infinite variations of “minimize-css-on-top-javascript-at-bottom-install-cache-plugin”? None.

Yes, of course there is another kind, one which I will keep on reading. They are usually written by developers, for developers, based on experimentation, objective metrics, and sound analysis of the specific problems at hand. Case-studies, not recipes.

Is the whole problem how to tell them apart? I maintain that it is not. The whole problem, for us site owners, maintainers, developers, editors, administrators or whatever, is actually deciding on which side of the fence we mostly sit.

Using a car analogy, let’s say you buy a production car, one that wasn’t built specifically for you, one that can be bought in the exact same configuration by thousands of people. But hey, you want your car to be yours, unlike that of thousands of people: hence you personalise it, decorate the interiors, apply racing stickers, maybe even a million “performance” accessories (all of which are usually esthetic). If you’re into that sort of thing, you can maybe even, tune the injection a bit, or use a different oil, special tires, a new exhaust system, and so on. You car will run faster (perhaps), and some of you will be even be tempted by the ways of Fast & Furious and go street-racing. That’s all fine and dandy, but a few questions remain to be asked:

  • Why did you buy the car, in the first place? To go street-racing (to take one extreme), or to comfortably go from point A to point B (the other extreme)?
  • If you answer was “street-racing”, do you plan to tune your car enough, all by yourself, to say, enter the 24 Hours of Le Mans? Or maybe you thought that a 600hp engine is an absolute requirement for commuting to work every day (it’s a jungle out there)?

Because if you do, you see, at some point you will need (at the very least) a mechanic, someone who knows engines, and cylinders, and valves, and crankshafts, and combustion ratios, and air intake systems, and exhaust systems, and a million other things. You’ll probably need a whole customized car, come to think of it.

So, what is your business? Is it car tuning? Or is it something else, for which you need a car?

The same applies to your website: again, what is your business? Is it tuning websites? Or maybe some other activity that needs a website?

I am not against a few standard tweaks that can make your website run faster, far from it, but you have to realise that the bigger your site becomes, and the more you rely on said website to do business, the more complex its code becomes, and implementing yet another “minimize-css-on-top-javascript-at-bottom-install-cache-plugin” will lead to trouble down the road. Not because the code is bad (okay, sometime it is), but simply because things crash for stupid reasons, and you, as the website owner (et al.) have no skills to fix them. When that happens, where’s your business then? In forums, trying to decide which of the two thousand answers to the same issue is the one that applies to your case? Reading yet another “performance” post”? What about your original business? Weren’t you supposed to be doing that, instead?

In short, a little tuning is cool, go ahead and do that, step by step, and make sure to thouroughly test at every step. Just keep in mind that there’s only so much you can do.

If you want enter faster races, however, go hire a mechanic already. Having the fastest car around isn’t going to help, if it explodes before the finish line.