Some links on our site may earn us a commission at no extra cost to you. Learn more.
This article discusses caffeine and food-safety research. It is general information, not medical advice — talk to a clinician about your own caffeine intake.
Updated August 2026: Bean cost per serving computed from the Bureau of Labor Statistics’ July 2026 average price and each method’s published brewing ratio, with the arithmetic shown. The acidity research behind cold brew’s smoothness claim comes from two papers sharing the same lead authors, who reach different pH findings and never measured a stomach. The UC Davis Coffee Center’s sensory panel found held coffee earns no preference against fresh.
The verdict. Japanese flash brew is the iced coffee to make. It costs about 40 cents of coffee per 12-ounce glass at the Specialty Coffee Association’s published ratio of 55 grams per litre of water, and it is in your hand four minutes after you decide you want it. Cold brew is the expensive one: at the National Coffee Association’s published steeping ratio of 1:4 to 1:5, a 12-ounce glass of undiluted concentrate carries $1.46 to $1.82 of coffee, which is 3.6 to 4.5 times the flash-brewed glass — and the smoothness claim it charges that for is not settled by its own supporting research.
Prices are approximate, current as of publication, and change frequently — confirm the latest price with the retailer before buying.
When You Price the Beans, Cold Brew Stops Being the Cheap One
Start with the only number that recurs every single morning: the coffee.
The Bureau of Labor Statistics publishes an average retail price for 100% ground roast coffee, all sizes, per pound, U.S. city average. Its July 2026 figure is $9.317 a pound, down from $9.457 in June and up from $8.414 in July 2025. A pound is 453.6 grams, so the coffee costs $0.02054 a gram.
Now the doses. The Specialty Coffee Association’s Standard 310-2021 for home coffee brewers sets the brew ratio at “55 g of coffee per 1.000 kg of fresh water,” brewed with the slurry held between 90 and 96 °C. A 12-fluid-ounce glass is 355 millilitres. At 55 grams per litre that is 19.5 grams of coffee, which is 40.1 cents.
Cold brew is not brewed at that ratio. The National Coffee Association publishes 1 gram of coffee to 4 or 5 grams of water, coarse grind, roughly 12 hours of contact. Steep 355 millilitres of water at 1:5 and you have used 71 grams of coffee. At 1:4, 88.75 grams.
| Method | Published ratio, and who publishes it | Coffee per 12 fl oz | Cost of that coffee | Against flash brew |
|---|---|---|---|---|
| Japanese flash brew | 55 g/L — Specialty Coffee Association, Standard 310-2021 | 19.5 g | $0.40 | 1.00x |
| Hot drip, then chilled | 55 g/L — Specialty Coffee Association, Standard 310-2021 | 19.5 g | $0.40 | 1.00x |
| Cold brew concentrate, 1:5 | 1:5 — National Coffee Association | 71.0 g | $1.46 | 3.64x |
| Cold brew concentrate, 1:4 | 1:4 — National Coffee Association | 88.8 g | $1.82 | 4.55x |
Here is the part the arithmetic cannot close, and I am not going to paper over it. You do not drink cold brew concentrate neat. You cut it with water or milk, and no standards body, government agency or peer-reviewed paper publishes a dilution ratio for serving. The Specialty Coffee Association has no cold brew standard at all. So the true cost of a finished 12-ounce cold brew sits somewhere between 40 cents, if you dilute it back to hot-brew strength, and $1.82, if you drink it as it comes out of the jar. I can price the steep. I cannot price the pour, and anyone who prints a single figure for it is guessing.
Per litre of liquid actually steeped, the spread is cleaner: $1.13 for hot brew at the SCA ratio, $4.11 at 1:5, $5.14 at 1:4.
When Two Papers by the Same Authors Disagree About the Acid
Cold brew’s whole pitch is that it is smoother, less acidic, easier on your stomach. Two papers get cited for it. They share their lead authors. They do not agree.
Niny Z. Rao and Megan Fuller published “Acidity and Antioxidant Activity of Cold Brew Coffee” in Scientific Reports on 30 October 2018. Six light-roast single origins, brewed hot and cold at 1:10. Their abstract: “The pH values of the cold and hot brew samples were found to be comparable, ranging from 4.85 to 5.13.”
Rao, Fuller and Meghan D. Grim then published a second study in Foods on 9 July 2020. One Colombian bean, three roast levels, same 1:10 ratio, cold steeped 7 hours against 6 minutes at 100 °C. Their finding: “Cold brew coffees from each of the three roasts were found to be less acidic than their hot brew counterparts, with mean values differing by 0.20, 0.26, and 0.34 pH units for R1, R2, and R3, respectively.”
| The claim | Who makes it | What the measurement says | Verdict |
|---|---|---|---|
| “Cold brew is less acidic” | Category marketing, near-universally | Rao & Fuller 2018, six origins: pH 4.85 to 5.13, described by the authors as comparable | Not supported by that paper |
| “Cold brew is less acidic” | Category marketing | Rao, Fuller & Grim 2020, one bean, three roasts: cold brew higher by 0.20, 0.26 and 0.34 pH units | Supported, by a fifth to a third of one pH unit |
| “Gentler on a sensitive stomach” | Category marketing | Neither paper measured a stomach, a symptom or a person | No evidence in either direction |
| Hot brewing pulls more acid | Rao & Fuller 2018 | Total titratable acid higher in hot brew in every sample tested | Supported — and it is a different quantity from pH |
| “Low acid” as a selling point | Category marketing | 21 CFR 114.3 defines low-acid as pH above 4.6. Every coffee measured in both papers is above 4.6 | The regulatory phrase means the opposite of the marketing one |
My reading of why the two papers differ, and it is my reading rather than anything either paper says: the 2018 study varied the bean and the 2020 study did not. Origin-to-origin spread in 2018 ran the full 4.85 to 5.13, which is wider than the 0.20 to 0.34 unit gap the 2020 design later resolved. Hold the coffee constant and a small effect shows up. Change the coffee and it disappears into the noise.
Both papers agree on the other half, which almost never makes the marketing: hot brewing extracts more titratable acid. pH and titratable acid are different measurements. A drink can sit higher on one and higher on the other, and cold brew does exactly that.
What neither paper supports is the digestion claim. Take the 2020 result at face value and cold brew is about a fifth of a pH unit gentler at light roast. That is the entire measured basis for a 3.6x price in beans.
When “Low Acid” Turns Into a Regulatory Category
The FDA’s definition sits in 21 CFR 114.3. Acid foods are “foods that have a natural pH of 4.6 or below.” Low-acid foods are “any foods, other than alcoholic beverages, with a finished equilibrium pH greater than 4.6 and a water activity (aw) greater than 0.85.”
Every coffee measured in both Rao papers, hot or cold, from 4.80 to 5.75, is above 4.6. Coffee is a low-acid food in the regulatory sense. That is why packaged cold brew in sealed, reduced-oxygen containers is a food-safety question rather than a flavour one — Death Wish Coffee recalled its 11-ounce Nitro Cold Brew cans in September 2017 over exactly that, and Cornell’s Food Venture Center wrote at the time that there was no scientific information showing cold brew is safe in reduced-oxygen packaging. The National Coffee Association has since published a challenge study reporting no pathogen growth or toxin formation in black cold brew held at retail temperatures.
Those two do not actually contradict each other. One is about shelf-stable sealed cans; the other is about black coffee held in a shop. Neither is about the jar in your fridge, and I could find no study that is. Keep it cold, keep it black, drink it inside a week, and you are outside the territory anyone has measured.
Before You Pour Out Yesterday’s Pot, Read the Sensory Data
Held coffee gets written off as flat and stale almost reflexively. That is an assumption, not a finding, and the research says otherwise.
Mackenzie E. Batali, Andrew R. Cotter, Lik Xian Lim, William D. Ristenpart and Jean-Xavier Guinard, working out of the UC Davis Coffee Center, published “The effect of post-brew holding time and carafe type on the sensory properties of drip brew coffee” in the Journal of the Science of Food and Agriculture, volume 103, issue 8, pages 4095 to 4106, on 23 January 2023. They held light, medium and dark roasts in three carafe types from 15 minutes to 3 hours, then had 93 coffee professionals compare fresh coffee against coffee held for 180 minutes.
The professionals showed no preference for the fresh cup. More sensory attributes fell than rose over the holding period, and the authors point at loss of volatile aroma as the main mechanism rather than anything happening to the acid.
Two honest limits on that. Batali held the coffee hot in carafes, not cold in a fridge overnight, and I could not find a single published study that measures hot-brewed coffee refrigerated for 12 to 24 hours. The nearest thing is Samuel Nicholas Lopane’s 2018 Clemson University master’s thesis on cold brew shelf life, which stored brews at 7 °C for 42 days and watched 3-CQA chlorogenic acid fall from roughly 1,157 to 345 parts per million in the cold-water extraction — about a 70 percent loss over six weeks, which tells you nothing useful about overnight.
So the honest position on last night’s pot is not that it is fine. It is that the only independent sensory work anyone has published on held coffee failed to find the penalty everybody assumes, and the overnight refrigerated case has never been measured. Taste it before you tip it.
When the Ledger Is Tallied, the Free One Wins
Five axes, weighted, each scored out of 100. Flavour is not one of them, deliberately: no credible body publishes a flavour ranking for these three, and a number I invented would be the weakest thing on the page. Disagree with a weight and the tally is yours to redo.
| Axis (weight, and why it carries that weight) | Flash brew | Chilled drip | Cold brew |
|---|---|---|---|
| Coffee cost per glass (30%) — the only figure you pay again every morning | 90 | 100 | 15 |
| A ratio a credible body publishes (20%) — the Specialty Coffee Association writes one down for hot brewing and none for cold | 70 | 70 | 45 |
| Claims that survive checking (20%) — a method that promises nothing cannot mislead you | 80 | 80 | 30 |
| Minutes from decision to first sip (20%) — the reason most people give up on a method | 95 | 100 | 55 |
| Independent evidence on holding it (10%) — small weight, because the evidence is thin for all three | 50 | 60 | 70 |
| Weighted total | 81.0 | 86.0 | 37.5 |
Chilled drip finishes at 86.0. It costs nothing extra, arrives in seconds, and has the only independent sensory work of the three behind it. Flash brew at 81.0 is five points back, which is not a tie, so no coin gets flipped here.
Cold brew finishes at 37.5 and is the named loser on this page. It is the most expensive coffee in the comparison by a factor of three and a half, it asks for twelve hours of planning, and the smoothness it charges for comes to about a fifth of a pH unit in the one study designed to detect it.
What the Ledger does not settle: whether you like it. Cold brew tastes different, and plenty of people will pay 3.6x for different. That is a preference and it is allowed. Paying it because a label said “less acidic” is a different transaction.
When Each One Is the Wrong Call
- Skip Japanese flash brew if you want iced coffee waiting for you rather than made for you. It is brewed to order, four minutes at a time, and it does not keep — you are standing at the counter every single glass.
- Skip chilled leftover drip if you do not already brew hot coffee in the morning. Its whole advantage is that the pot exists anyway. Brewing a pot specifically to chill it removes the only axis it wins outright, and puts you back at 40 cents with a wait.
- Skip cold brew if the reason you are buying it is the stomach claim. That is the one thing the research does not support. Buy it because you prefer the taste, because you want a jar in the fridge on Wednesday that you made on Sunday, or because you drink it with milk where the concentration works for you — those are real reasons and the price is the price.
One more thing that changes all three: grind. Cold brew wants coarse, flash brew wants medium-fine, and no method on this page survives a blade grinder. If yours is inconsistent, fix that before you change methods — the burr, blade and manual grinder comparison covers what that costs. For the machine that makes the hot half of this, see the drip, pod and pour-over comparison, and for the per-cup arithmetic on pods against whole beans there is the per-cup ledger. Everything else lives in Coffee & Espresso.
