Most grind guides are written by people who already know how their grinder behaves, for people who don't. They describe grind size with words like "table salt" or "fine sand," then show you a photo of someone's hand holding both. It's useless. Grind size isn't a number on a chart — it's a variable you adjust by tasting what came out of it.
For two years I changed my grind setting by feel, picking a number that seemed right, then re-picking the next morning. The coffee was inconsistent, and I had no idea why. I was missing a feedback loop. Here's the one I use now.
Taste is the signal
Two flavors tell you everything you need to know about your grind: bitter and sour. They sit at the opposite ends of the extraction spectrum, and most other "off" flavors are decoration on top of one of them.
Bitter means over-extracted. The water has spent too long pulling compounds out of the grounds, or there was too much surface area for it to work with. Either way, you went too far — the answer is to grind coarser, or shorten the brew.
Sour means under-extracted. Either too little contact time, or the water couldn't reach enough surface area to get past the bright, acidic notes. The answer is to grind finer, or extend the brew.
Everything else — flat, hollow, ashy, papery — is a variation on one of these two signals, usually combined with something else (bad water, stale beans, wrong temperature). Start with the bitter/sour axis. The rest gets easier once that one is in range.
The three-step loop
Brew the recipe. Taste. Move once.
Pick a recipe — any reasonable one for your brewer — and follow it exactly. Don't improvise the first cup. When you taste it, ask a single question: is this more bitter than sour, or more sour than bitter? Then move the grinder one or two clicks in the corresponding direction. Don't move farther than you can taste; the goal is to learn the size of one step, not to skip ahead.
Tasting notes are a compass, not poetry.
You don't need to identify "stone fruit" or "panela sugar" or whatever the bag promises. You need to know which way the cup is leaning. Keep a one-line note in your phone for each brew: grinder setting, brew time, and one word — bitter, sour, or good. After two weeks you'll have a map of how your grinder, your beans, and your brewer interact. That map is worth more than every grind chart on the internet.
Change one variable at a time.
The fastest way to stay lost is to change grind and water temperature and dose all at once. You'll find a better cup eventually, but you won't know what fixed it. Hold everything else still. Move the grinder. Taste. Write it down. Then, if you want, change another variable.
This sounds tedious because it is. It also stops working tediously after about a week, when you've internalized the pattern and stop needing the notebook.
Where grind matters most
Grind isn't equally important across brewers. Espresso is the most sensitive — a single grinder click changes the shot from sour and gushing to bitter and choked, sometimes in fifteen seconds. If you have an espresso machine, your grinder is the most important piece of gear you own. It's not even close.
Pourover — V60, Kalita, Origami — is the middle ground. Grind affects flow rate and extraction time, but the window of "good" is wider than it is for espresso. You'll taste the difference between two clicks, but you won't get a bad cup from being one off.
Immersion brewers — French press, Aeropress, Clever — are the most forgiving. The water is in contact with the grounds for a fixed time, so grind affects extraction but not flow. You can hit "good enough" with a much wider range. If you're starting out, this is the most generous place to be.
One last thing
The hardest part of this isn't the framework. It's resisting the urge to chase perfect. A 7/10 cup every day is much, much better than a 9/10 cup that takes forty minutes to dial in and leaves you grumpy by the time you drink it. The point of having a system is that you can stop thinking about it after a while. That's what I'd hoped for, two years ago, when I was guessing. It took longer to get there than I'd like, but I'm there now. Most mornings I don't move the grinder at all.