My first cold brew was a quart mason jar of coarse grounds and tap water that lived in the fridge door for most of a summer. I would top it up when it got low. It never occurred to me that I was running a two-month continuous culture of whatever had been on my hands the day I filled it, and nothing bad ever came of it, which is the least useful kind of evidence there is. I make it differently now, and the changes are boring: a ratio I write down, a lid, and a date on the jar.
Cold water buys you nothing, so you spend hours instead
Hot brewing works because heat gives the water enough energy to pull solubles out of the grounds in a few minutes. Take the heat away and the same compounds still come out, just slowly and in a different order. The bright acids and the aromatic top notes that hot water lifts immediately are underrepresented; the sugars and the heavier body-forming compounds arrive anyway, given long enough.
That is the whole flavor signature. Cold brew tastes round, low-acid, and slightly flat compared with the same coffee brewed hot, and people who love it love it for exactly the trait people who dislike it complain about. It is not a stronger version of drip. It is a different extraction of the same bag.
Two numbers, and one of them decides the other
Decide first whether you are brewing concentrate or a ready-to-drink jug, because everything downstream follows from that. Concentrate sits around 1:5 to 1:8 coffee to water by weight and gets cut before you drink it. Ready-to-drink lands near 1:15, close to the numbers in the coffee-to-water ratio notes, and goes straight into a glass.
I make concentrate, for one reason that has nothing to do with flavor: a fridge shelf holds a lot more coffee that way. A 1:5 jar at 200 grams of coffee gives me a week of iced coffee in a container that fits behind the milk, and I control the strength at the glass rather than at the jar.
Steep time then covers a wide, forgiving band. Twelve to eighteen hours on the counter, or eighteen to twenty-four in the fridge, since colder water works slower. Past about twenty-four hours the returns flatten and the cup starts picking up a woody, papery edge. Under about ten and it tastes like a rinse.
Coarse means coarser than you think
Grind for cold brew sits at the far coarse end, well past French press — closer to raw sugar than to sand. Fine particles are the enemy twice over here: they over-extract into bitterness over a long steep, and they turn the filtering step into a silt problem you cannot fix afterward.
This is where a blade grinder struggles most, because it produces a mix of dust and boulders at every setting and the dust does the damage. If a burr grinder is not in the budget, the grinder comparison covers the honest workarounds, and for cold brew specifically the best of them is buying the coffee ground coarse from a shop that will do it for you.
The counter or the fridge
Room-temperature steeping extracts faster and gives a fuller, sweeter cup. Refrigerated steeping is slower, cleaner, and brighter, and it keeps the whole process cold from the first minute. Commercial producers overwhelmingly steep cold, and their reason is not flavor.
At home I split the difference on a schedule that fits my kitchen: I mix it at night, leave it on the counter until morning, then filter and refrigerate. That puts the room-temperature phase at roughly ten hours rather than an open-ended one, and the finished concentrate goes cold immediately rather than sitting out while I get around to it.
What the food-safety research actually says
This is the part I did not know back in the mason-jar days. Researchers at the University of Georgia's Center for Food Safety looked at whether cold brew can make you sick, and the finding is more nuanced than either camp on the internet claims. Graduate researcher Angela Parra found that pathogenic bacteria do not grow in cold brew if they are not already there. What they do is survive: nine to twelve days for the ones she tested, and over thirty days for Bacillus cereus.
So the risk is not that a clean jug spontaneously turns dangerous on the shelf. The risk is that cold brew never gets a kill step — no boil, no pasteurization — and its low acidity means anything that goes in at the start is still there a week later. Whatever is on the jar, the spoon, the filter cloth, or your hands is a permanent guest.
Which makes the fix pleasantly simple, and identical to the fix for everything else in this journal: it is a sanitation problem, not a chemistry problem. Wash the vessel and the filter properly, use drinkable water, keep your hands off the grounds, and refrigerate the finished coffee.
Getting the grounds out without a filter cabinet
Two stages beat one. First, pour off the liquid through a mesh strainer, or lift out the brew bag if you used one, and let it drain on its own rather than squeezing it — squeezing pushes fines and bitterness straight through. Then let the strained coffee sit in the fridge for a couple of hours and decant it off the sediment that settles.
If it is still hazy after that, run it once through a paper filter in a dripper, which is slow but final. What I avoid is filtering hot-brew-style through paper as the only step, because a full jug of unstrained cold brew will clog the paper and you will spend twenty minutes hovering over a cone.
How long a jug actually keeps
Straight concentrate in a clean sealed jar in the fridge holds up for about ten to fourteen days, with the flavor fading before anything else does. Once it is cut with water it is a three-to-four-day proposition, and once it is cut with milk it is a same-day drink and belongs on the same clock as the milk itself.
The habit that made this workable was a strip of tape with the brew date, which sounds fussy and takes four seconds. It is the same instinct behind dating a bag of roasted beans in the storage notes: the point is not that the coffee expires on a schedule, it is that you stop having to guess.
The Sunday jar
Two hundred grams of coarse coffee, a liter of filtered water, stirred until every ground is wet, lid on, left on the counter overnight. Monday morning it gets strained, decanted, and dated, and it lives in a jar at the back of the shelf. A glass is roughly one part concentrate to three parts water or milk over ice, adjusted once and then left alone.
The whole thing takes about ten minutes of actual attention per week. If you have been buying cold brew by the bottle, run this once with a coffee you already like hot — the interesting part is not the savings, it is finding out how much of the bottled stuff tastes the way it does because of the coffee, and how much because of how long it sat.