Cold brew guide
Make cold brew with one decision before everything else.
Decide whether you are brewing a concentrate or a drink-ready batch. Once that is clear, ratio, steep time, filtering, and dilution stop fighting each other.

Cold brew recipes look contradictory when one writer is making concentrate and another is making coffee meant to drink straight. The ratios are different because the finished jobs are different.
Choose concentrate or ready-to-drink before you grind
Toddy's cold-brew cupping protocol is unusually helpful here because it labels ratios by intended strength. It lists 1:5 as concentrate, 1:8 as concentrated ready-to-drink, 1:10 as strong ready-to-drink, and 1:14 as typical ready-to-drink. The point is not that those four numbers are mandatory. The point is that “cold brew ratio” is incomplete without saying what kind of finished drink you want.
Concentrate is useful when several people want different strengths, when milk drinks are common, or when you want to adjust each glass. Ready-to-drink is simpler: filter it, pour it over ice, and make only small serving adjustments.
Two starting recipes that are easy to compare
These are Measured Bloom starting recipes built around Toddy's published strength categories. They are meant to give you a clean baseline, not to replace the instructions for a specific cold-brew maker.
| Concentrate | Ready-to-drink | |
|---|---|---|
| Ratio | 1:5 | 1:14 |
| Example batch | 100 g coffee + 500 g water | 50 g coffee + 700 g water |
| Grind | Coarse starting range | Coarse starting range |
| Measured Bloom test time | 16 hours | 16 hours |
| After filtering | Dilute to taste | Usually drink as brewed |
Why 16 hours? It is not a universal optimum. Toddy's home instructions allow a broad 8–24 hour room-temperature steep, OXO commonly uses 12–24 hours at room temperature, and Toddy's experiment guide explicitly compares 8, 12, and 16 hours as test variables. Sixteen hours simply gives us a repeatable place to begin inside those published windows.
Put the coarse coffee and water in a clean brewing vessel, make sure all grounds are wet, cover the vessel, and leave it alone. If your brewer's manufacturer gives a different procedure, follow the brewer-specific instructions first — especially when the filter system depends on a particular filling order.
Start coarse, but do not turn “coarse” into a visual religion
Toddy's current home instructions and OXO's cold-brew instructions both begin with coarsely ground coffee. A coarse starting point suits long immersion and generally makes filtration easier. It also avoids pretending that one grinder number transfers to another machine.
The useful part is consistency. If the grounds contain a lot of fines, the filter may drain slowly or the finished coffee may carry more sediment. If your batch filters cleanly and tastes balanced, there is no reason to make the grind coarser simply because another recipe uses the phrase “extra coarse.”
Steep time is a variable, not a strength control
It is tempting to leave a weak cold brew sitting for another ten hours. That mixes up two different problems. Ratio sets much of the drink's concentration; steep time changes how long water and coffee interact. Toddy's experiment guide treats brew ratio and dwell time as separate variables for exactly that reason.
Pick one steep time and repeat it. If the filtered coffee tastes balanced but too light for your preference, use more coffee or less water next time — or choose a concentrate recipe. If the flavor itself seems underdeveloped, then a time or grind experiment makes more sense.
Filter patiently and measure what comes out
Cold brew grounds hold onto water. That means 500 g of brew water does not become 500 g of filtered concentrate. OXO's compact instructions, for example, note that 24 oz of starting water yields about 16 oz of concentrate in that system.
This is why Measured Bloom does not estimate a universal filtered yield. Let the coffee drain according to the brewer or filter you use, then measure the liquid you actually collected. That number is what matters for serving and dilution.
If your filtration is muddy or painfully slow, check grind consistency and the filter before adding more steep time. OXO offers an additional paper-filter stage for its full-size system, while Toddy's home system uses its own felt and optional paper filtration. The filtration method is part of the recipe.
Dilute the concentrate, not the original brew water
Once the concentrate is filtered, measure a serving and add water or milk. Toddy's current home guide recommends beginning around 1 part concentrate to 2 parts water or milk. OXO recommends roughly 1 part concentrate to 2–3 parts water or milk for its concentrate recipe.
That spread is a useful reminder: dilution is a taste decision. Start with a measured ratio, taste it before the ice melts much, and write down the version you prefer. The next glass becomes effortless.
If the cold brew is too weak, too strong, harsh, or hollow
Start by naming the problem correctly. “Weak” describes concentration. “Hollow” or “underdeveloped” describes flavor. They can happen together, but they are not the same fix.
| What you notice | Try next | Keep steady |
|---|---|---|
| Balanced but too weak | Use a stronger brew ratio or less dilution | Grind and steep time |
| Balanced but too strong | Add more water or milk, or brew a lighter ratio next time | Grind and steep time |
| Hollow / underdeveloped | Test a little more contact time or a modestly finer grind | Ratio for the comparison |
| Harsh / woody / drying | Test a shorter steep or a slightly coarser grind | Ratio for the comparison |
| Muddy / difficult to filter | Go a little coarser and reduce fines | Ratio and steep time |
Do not change ratio, grind, steep time, and dilution in the same batch. Toddy's own experiment guide recommends isolating one brewing parameter while keeping the others constant. That is slower than guessing once and faster than never knowing what fixed the coffee.
Refrigerate the filtered coffee promptly
After filtering, keep cold brew covered and refrigerated. Toddy and OXO both state that concentrate made in their systems can be refrigerated for up to two weeks. Treat that as manufacturer guidance for those systems rather than a universal shelf-life guarantee for every homemade jar, filter, or handling routine.
Plan the batch, then measure the glass
Keep brewing ratio and serving dilution separate.
The Cold Brew Calculator scales concentrate or ready-to-drink recipes by weight, then switches to a separate mode for diluting the filtered concentrate you actually collected.
Cold brew FAQ
What is a good cold brew coffee ratio?
First decide whether you want concentrate or ready-to-drink cold brew. Toddy’s cupping protocol uses 1:5 as a concentrate reference, 1:10 as strong ready-to-drink, and 1:14 as typical ready-to-drink. These are starting categories, not universal recipes.
How long should cold brew steep?
Published home-brewing instructions vary. Toddy gives an 8–24 hour room-temperature window, while OXO commonly recommends 12–24 hours at room temperature and longer in the refrigerator. Start with one repeatable time and adjust only after tasting.
What grind size should I use for cold brew?
Coarse is a useful starting point for long immersion cold brew. The exact setting depends on the grinder and filter. If filtration becomes muddy or difficult, a slightly coarser grind can help; if the brew tastes hollow after a reasonable steep, test one small change at a time.
How do I dilute cold brew concentrate?
Measure the concentrate after filtering, then dilute that liquid rather than the original brew water. Toddy currently suggests starting around 1 part concentrate to 2 parts water or milk, while OXO suggests 1 part concentrate to 2–3 parts water or milk.
Why did I get less cold brew out than the amount of water I added?
The coffee grounds retain water, so filtered yield is lower than starting brew water. The exact loss depends on coffee, grind, filter, vessel, and draining method, which is why Measured Bloom calculates dilution from the concentrate you actually collect.