Water for Pour Over Coffee: Why Japan Brews Soft
You bought better beans. You upgraded the grinder. You weighed everything, dialed in the ratio, timed the pours. And the cup is still... fine. Just fine.
There's one ingredient left that most home brewers never touch, and it makes up roughly 98 to 99 percent of what's in the mug: the water.
This isn't a small lever. Two people can follow the exact same recipe with the exact same beans in two different cities and get noticeably different coffee, because what comes out of the tap is different. It's also part of the answer to a question we get a lot: why does coffee in Japan taste so clean? Some of that is culture and technique, which we cover in our complete guide to Japanese pour over. But some of it is simply that Japan brews on very soft water.
Here's what's actually in your water, how to find out what you've got, and what to do about it.
Your coffee is almost entirely water
Brewed filter coffee is about 98 to 99 percent water by weight. Everything you taste — the acidity, the sweetness, the body — comes from the small fraction of the bean that the water manages to dissolve and carry out of the grounds.
That word dissolve is the whole story. Water isn't a neutral carrier. It's a solvent, and how well it does its job depends in part on what's already in it. Water that arrives at your coffee bed completely empty behaves differently than water carrying some calcium and magnesium.
So when we say "water matters," we don't mean it in a vague, wellness-blog way. We mean the mineral content of your water measurably changes what ends up in your cup.
The three numbers that matter
You don't need a chemistry degree. You need three concepts.
Hardness: calcium and magnesium
Hardness measures dissolved calcium and magnesium, usually reported in milligrams per liter (mg/L) or parts per million (ppm) — the same thing for our purposes.
These two minerals do real work during extraction. Too little of them and the cup tends to come out thin and hollow, no matter how fine you grind. Too much and things go the other way — the coffee turns dull and chalky, origin character gets muted, and your kettle starts collecting limescale.
You'll often read that magnesium specifically favors bright, fruity notes while calcium leans toward body. That idea traces back to a well-known 2014 study on dissolved cations in coffee extraction, and it's a reasonable working model. It's worth knowing that more recent work has pushed back on parts of it, finding that acid extraction may depend less on water composition than that study suggested. The broad point — that mineral content shapes the cup — is not in dispute. The precise mechanism is still being argued about.
Alkalinity: the acidity eraser
Alkalinity (mostly bicarbonate) measures your water's ability to neutralize acid. This is the sneaky one, and the number most people have never heard of.
Coffee's brightness is acidity — the citrus snap of a Kenyan, the apple note in a washed Ethiopian. Bicarbonate buffers that acidity away. High-alkalinity water doesn't make coffee taste bad, exactly. It makes it taste boring: flat, muted, like someone turned the treble down. If your light roasts consistently taste dull and you can't work out why, alkalinity is a prime suspect.
TDS: the total
Total dissolved solids (TDS) is everything dissolved in the water added together — the minerals above plus sodium, chloride, and the rest. It's a useful one-number summary, but a blunt one: two waters at the same TDS can be built completely differently.
The Specialty Coffee Association publishes a water standard for brewing. Here are the targets:
| Measure | Target | Acceptable range |
|---|---|---|
| Total dissolved solids (TDS) | 150 mg/L | 75–250 mg/L |
| Calcium hardness | 68 mg/L | 17–85 mg/L |
| Total alkalinity | ~40 mg/L | at or near 40 mg/L |
| pH | 7.0 | 6.5–7.5 |
| Sodium | ~10 mg/L | at or near 10 mg/L |
| Chlorine | 0 mg/L | 0 mg/L |
| Odor / color | Clean, fresh, odor-free | Clear |
Treat that table the way you'd treat a starting ratio: a well-reasoned default, not a verdict. Plenty of excellent coffee gets brewed outside those numbers — including in Japan.
Japan brews on soft water — and it shows
Here's where it gets interesting.
A nationwide survey of Japanese tap water found a mean hardness of 48.9 mg/L, with a median of 46.0. Under the standard classification — soft below 60 mg/L, moderately hard from 60 to 120, hard from 120 to 180, and very hard above that — Japan is soft water, almost everywhere.
There is regional spread, but it's narrow by global standards:
| Area | Approx. hardness |
|---|---|
| Chiba | 83.4 mg/L |
| Saitama | 82.8 mg/L |
| Okinawa | 68.1 mg/L |
| Tokyo | 65.8 mg/L |
| Hiroshima | 23.5 mg/L |
Now compare that to the SCA's calcium hardness target of 68 mg/L. Japan's national average sits below the recommended target — comfortably inside the acceptable range, but at the soft end of it.
That's worth sitting with, because it cuts against the easy story. Soft water isn't automatically "better" water. It's a starting condition, and Japanese coffee culture built a style around it.
Very soft water tends to produce a cup that reads clean, delicate, and transparent — you taste the bean rather than the water. What it doesn't hand you for free is weight. Body and texture have to come from somewhere else, and that is exactly what generations of Japanese technique have quietly solved for: patient blooms, controlled pour rates, careful grind selection, and — in the kissaten, Japan's traditional coffee houses — the nel drip, a cloth filter that builds a heavy, syrupy body a paper filter on soft water would struggle to match.
Put another way: the deliberate, unhurried technique you see in a Japanese café isn't purely aesthetic. Part of it is an answer to the water. For the cultural backdrop, our piece on kissaten covers where that style came from.
American water is a different animal
The United States has no equivalent of Japan's national consistency. Hardness varies enormously by region, and it's one of the most common reasons a recipe that worked in one apartment falls apart in another.
The USGS uses the same classification — soft under 60 mg/L, moderately hard 61 to 120, hard 121 to 180, very hard above 180 — and much of the country lands in the moderately-hard-to-very-hard bands. A broad belt of hard water runs from the Dakotas down through Texas, and cities drawing on mineral-dense aquifers can test well past the SCA's upper limit of 250 mg/L.
If you live in that belt and your coffee tastes muted and heavy no matter what you change, you may not have a technique problem at all. Before you tear apart your method, it's worth ruling out the usual brewing mistakes — and then looking hard at the tap.
How to find out what's in your water
Three ways, cheapest first.
1. Read your utility's water quality report. Every community water system in the U.S. publishes an annual Consumer Confidence Report. It's free, it's public, and it usually lists hardness and alkalinity. Search your utility's name plus "water quality report." This costs nothing and takes about five minutes.
2. Use test strips. Aquarium and pool test strips measure hardness and alkalinity well enough for coffee purposes. Dip, wait, compare to the chart. Cheap and immediate — and unlike the annual report, it tells you what's coming out of your tap today.
3. Taste it. Drink a glass of your tap water cold, on its own. Chlorine, a metallic edge, or a chalky finish will all show up in your coffee, amplified. If you can taste it in the glass, you'll taste it in the cup.
Four ways to fix your water
You don't have to go far down this road to get most of the benefit. Options, roughly in order of effort:
| Approach | Effort | Best for |
|---|---|---|
| Activated carbon filter (pitcher or faucet) | Low | Almost everyone; chlorine and off-flavors |
| Bottled spring water | Low | Renters, quick experiments |
| Distilled water cut with tap or mineral water | Medium | Very hard water areas |
| Distilled water plus a mineral additive | Medium | Full control and repeatability |
Start with a carbon filter. A basic filter pitcher — Brita and similar — strips chlorine and many off-flavors. It won't dramatically change hardness, but chlorine removal alone is the highest-return, lowest-effort fix available, and for a lot of people it's enough.
Bottled spring water works if you read the label. Look for something in the neighborhood of the SCA targets, and skip anything marketed as high-mineral. Note that "purified" water is often distilled or reverse-osmosis water, which is a different thing entirely — see below.
Dilution is the move for hard-water cities: cut your tap or mineral water with distilled water to bring hardness down into range. Start at 50/50 and adjust by taste.
Mineral additives — Third Wave Water is the best-known, and DIY recipes using Epsom salt and baking soda are widely documented — let you build brewing water from a distilled base to a known spec every time. It's the most controlled option and the most fuss.
What not to use
Distilled or reverse-osmosis water on its own. This is the most common well-intentioned mistake. Pure water sounds like it ought to be ideal, but with no minerals present, extraction suffers and the cup comes out hollow and flat. Distilled water is an excellent base. It's a poor finished product.
Water from a conventional water softener. Ion-exchange softeners don't so much remove hardness as trade it — calcium and magnesium out, sodium in. You lose minerals that help extraction and gain sodium, which the SCA standard wants at or near 10 mg/L. Softened water usually makes coffee worse, not better. If your house has a softener, look for an unsoftened tap.
Untreated chlorinated tap water. Chlorine is volatile and assertive, and it will sit right on top of everything else in the cup. This is the easiest problem on the list to solve.
Temperature is a separate dial
Worth saying plainly, because these two get tangled: water temperature and water chemistry are unrelated problems.
The standard brewing range is 195–205°F (91–96°C). Getting your temperature right won't fix hard water, and fixing your water won't rescue a brew made with under-temperature water. Handle them separately. If you're chasing temperature control, choosing the right gooseneck kettle is the place to start, and our V60 recipe has the full set of numbers.
The two-cup test
Don't take any of this on faith. Run it.
Brew the same beans, same grind, same ratio, same temperature — twice. Once with your tap water, once with bottled spring or filtered water. Put the two cups side by side and taste them against each other while both are still warm.
Most people are genuinely surprised. If the two taste the same, congratulations: your water is fine and you can stop thinking about this. If they don't, you've just found your biggest available upgrade, and it costs less than a bag of beans.
FAQ
Does water really make that much difference?
For most people it's a bigger jump than upgrading a dripper and a smaller one than replacing a bad grinder. If your water is very hard, very soft, or heavily chlorinated, it can be the single biggest change available to you.
Can I use distilled water for coffee?
Not on its own. Use it as a base and add minerals, or cut harder water with it. Straight distilled water makes noticeably flat coffee.
Is Japanese bottled water worth importing?
No. You'd be paying to ship water. If you want to brew closer to Japanese conditions, use soft local water or dilute what you have — the hardness number is what matters, not the country on the label.
Will hard water damage my kettle?
It leaves limescale, which builds up over time and affects heating in electric kettles. Descale periodically with a commercial descaler or diluted white vinegar, then rinse thoroughly.
Start here
If you do one thing: find out your water's hardness, either from your utility's annual report or a test strip. Everything else follows from that number.
Get a carbon filter if you don't have one. Run the two-cup test. And if you're in a hard-water city, try cutting your water with distilled before you touch anything else in your routine — it's the fastest way to hear what your beans have been trying to say all along.