I’ve been on a bit of a coffee mission in the last few months, ever since discovering the wonders of Has Bean Coffee’s subscription coffee plan. One single-estate coffee through your door every week, what’s not to like?

And as a result, I’ve been working out how to make that coffee taste as awesome as humanly possible. Now, I’m a bit old-fashioned, and grew up with a heavy francophile influence, so my preferred way of making coffee is the cafetière (“French press” for the USAians amongst us, although I’m sure there are states where it’s now officially referred to as the “Freedom Press”).

Cafetières are awesome, actually. Baristas seem to reckon they’re one of the best ways of making coffee, period, and certainly one of the top ways that doesn’t cost hundreds or thousands of pounds. They produce coffee similar to that from a filter, but less burned-tasting than filter machines often produce, and with a real complexity of flavour depending on how you brew.

And there’s the rub. Making coffee with a cafetiere might look simple, but there’s a LOT of optimisation you can do…

h2. Optimise your grind

Hardcore types will already know that blade grinders, the cheapest type of coffee grinder, are fantastic for chopping chilli but sodding useless for coffee (although, IMO, if you have to use one a cafetiere brew is one of the less-horrible ways to do it). Pre-ground coffee is only of much use if you drink it within a day or so of grinding, and even then it’s nowhere near as aromatic as fresh-ground. So, one of the first stops on the coffee pilgrimage tends to be the burr grinder, with variable grind size. But here’s the rub - even burr grinders aren’t created anywhere near equal, and a bad burr grinder is particularly bad for cafetiere grinds, which need to be large and even. It turns out, you see, that cheap burr grinders can set a maximum grind size, but will randomly produce lots of finer-ground particles of coffee smaller than that grind size too, and that equals sludged-up cafetière and overbrewed coffee thanks to excessive surface area.

After screaming at the price of top-end grinders, I’ve settled for now on a Hario Skerton hand grinder. It’s a pain to figure out how to use it initially (I’ll write a guide on here at some point), and it’s certainly some fairly hard work to grind 50 grams or so of beans for four people, but it produces a very, very nice, smooth grind, and the smell when you first open it up after grinding is worth the price of admission on its own. Plus, I’ll probably study Brazilian Ju-Jitsu at some point, so the increased grip strength from using it every day ain’t going to waste…

h2. A/B testing cafetières’ brews

OK, let’s start simple. Get a set of scales with a 1 gram sensitivity. Better than that would be great, but they need to be able to handle a maximum weight of 2 kg or so, and most microgram scales can’t cut that, at least not on a non-lab budget. (Please do tell me if I’m missing a brand here - I’d love to get a set of microgram scales with a wide weight range).

Why do you need scales? Because the single biggest optimisation I’ve been able to make to my cafetière coffee has been precise weight measurement, both of the beans and the water. Measuring by volume works very, very badly for coffee - the ground beans froth varying amounts, the volume can be widely different dependant on the grind size, there are all the usual problems of volume compared to reasonably large eliptical objects. Get a set of scales in play and you can get granular on the problem. So to speak.

Now, there are four variables in the process. People who design websites, ads, software or Toyota cars for a living will be starting to make a matrix in their head already.

1. Temperature. Coffee is brewed somewhere in the 88-93 degree centigrade ( 190F to 200F ) range, but there’s not a lot of agreement beyond that as to what the ideal temperature is. (Yeah, I know I said 93 in the electric kettle article. But, as usual, it’s more complicated than that, much like finding a bloody yellow kettle turned out to be - true story.)

2. Brew time. Artisan Coffee in Edinburgh say 3 minutes, Square Mile Coffee say 4 minutes. My experience is that 3 minutes produces a more fruity, acidic brew, and 4 minutes produces a richer, heavier, and more caffinated brew.

3. Amount of coffee. Again, not a lot of agreement here. Somewhere between 5 grams and 8 grams per 100ml works well, with 8 grams working for heavier, smoother beans, and 5 grams working well for the lighter and fruitier, although it can sometimes produce a smooth brew too.

4. The bean. Different beans respond different ways to different approaches. A Kenyan Gethumbwhini will work well with a 4 min/7g/l combination, wheras the El Salvador Finca Argentina (a fantastic bean from this week’s subscription) seems to respond better to 5g/l / 4min.

So what do you do? Well, there’s a variety of approaches you could take. At the hardcore end, I’d recommend half a dozen small cafetieres, a big pile of ground beans, a matrix table, and some serious tasting. (I keep meaning to organise this in Edinburgh). You might want to Google “A/B testing”. At the lighter end, just vary your brew between the extremes every time you make a morning cuppa, and keep notes (that’s what I’m doing normally).

(Anyone got a systematised way for optimal testing of this sort of thing? My maths ain’t up to it.)

Either way, the important thing to remember is that each bean responds totally differently. So don’t assume a brew method for one bean will result in a good brew with another - I’ve had one bean which actually tasted plastic if given a long, high-density brew, but was fantastic at a much lighter brew.

h2. All cafetières are not created equal

For starters, there’s the temperature drop-off through the side - whilst the classic cafetiere is made of glass, that means that it’ll lose heat really quickly, meaning that the optimal extraction temperature drops off. (Hmm, we should really do some taste testing on this - watch this space.). There are a couple of ways to mitigate that - remember to pre-heat the cafetiere, for starters, as it really does make a difference. Wrap the cafetiere in something insulating, like a towel or some foam. And if you really want to get hard-core, you could immerse the cafetiere in hot water, or even water heated to exactly the right temperature in a water bath. (Again, must try that.)

Your best option for convenience and quality is probably to buy a double-walled cafetiere, though. Sadly Square Mile seem to have stopped selling their excellent-looking silver ones, but Amazon have a fair selection. They’ll hold heat like a thermos flask, meaning you get a much, much better brew.

The other variable, of course, is shape. As Mathias pointed out in the kettle article, the shape and material of a brewing vessel has a hell of an effect on the brew. In general, bigger, wider vessels should ensure greater extraction in less time, and they’ll also lose less heat to boot (volume and hence heat energy goes up as a cube whilst surface area only goes up as a square.). Having said that, the press will also work better the higher up the cafetière the coffee starts, so balance is important.

And that’s about it. Any other cafetiere tips out there?