I haven't been able to find a good way to approximate the amount of caffeine in the coffee I make. Googling yields information like this Mayo Clinic article that mixes serving sizes and brewing methods, and it reports one 8 oz coffee contains "95-200 mg" of caffeine, which is a huge range. If I have three cups of coffee in a day, is that 300 or 600 mg of caffeine?

There are so many variables in prepared coffee, different extraction methods, beans, roasts, serving sizes, mixed with different amounts of water or milk.

Is there a general method for calculating the caffeine in prepared coffee? Something like "most brewing methods extract X mg of caffeine from each Y grams of ground coffee"?

Or more specifically, what method would I use to accurately estimate how much caffeine is in an 8oz serving from my 10 oz french press, made with medium- (not course-) ground medium-roast arabica beans, brewed 4 minutes before decanting?

  • 1
    FWIW the Cooking SE site had a similar thread that discusses scientific ways to measure caffeine levels yourself: cooking.stackexchange.com/questions/1927/…
    – Justin C
    Commented Jan 28, 2015 at 20:13
  • @JustinC that is brilliant! I am looking for something easier than actually measuring it though. For example, [more than 300mg per day is dangerous for pregnant women ](mayoclinic.org/drugs-supplements/…). If 8oz brewed coffee varies from 95-200mg, how should a pregnant woman estimate how much homemade coffee would be under that?
    – Nathan
    Commented Jan 28, 2015 at 22:45
  • 1
    Firstly, Mayo Clinic is like the w3schools of programming, there's no point in trusting it. Secondly, this question is quite nice. Right to the point and difficult, the way it should be.
    – qedk
    Commented Jan 31, 2015 at 14:56
  • This is one of the reasons I used to drink Rockstar. Total Caffeine listed right on the back of the can. Commented Feb 2, 2015 at 17:11

8 Answers 8


I read everyone's answer and did some of my own research to find an estimation method that satisfies me and might be helpful to others too. I think it should be roughly ±20% accurate.

For any amount of "good strength" American-style coffee by any brew method, weigh the dry coffee in grams and multiply by 0.008, or 80mg of caffeine for each 10g of dry coffee.

Weakly extracted coffee might yield 0.6% of its weight in caffeine and very strongly extracted coffee might get close to 1% caffeine.

These numbers are for typical 100% Arabica or mostly Arabica blends. 100% Robusta beans (uncommon) could be about double the caffeine.

Explanation below:

Maximum caffeine content

Typical premium roasted Arabica coffee beans like mine contain close to 1% caffeine by mass when measured by spectrophotometer (p 311) though Robusta (less common and usually blended with Arabica in cheaper coffees) has roughly twice that much.

If you have a health concern like wanting to stay under the 300 mg guideline for pregnant women, you should be able to use this as the upper limit: each 10g of Arabica coffee contains about 100mg of caffeine.

I use a 2 Tbsp scoop to measure beans before I grind them. One scoop of beans weighs about 12 grams (yes I tared the scoop, but I did only weigh once and am not comparing light or dark roasts which will vary in weight and volume).

2Tbsp scoop of beans weighing 12g

So if I eat that scoop of beans or grind them finely and drink them with any amount of water I will be consuming the 120 mg (1% of 12g) of caffeine present in them. That should be an upper limit to this estimation exercise. The strongest or biggest/weakest cup of coffee I can brew with that scoop of beans will have no more than 120mg caffeine.

How strong is my cup?

Typical American-style brewing methods and water ratios result in 1% (weak) to 1.65% (too strong) disolved coffee solids, with 1.25% considered optimal whereas "a typical espresso will contain on average 1.8-2.2%" solids

So 1 fl oz of brewed coffee or espresso weighs almost the same as water (29.57 g) and contains 1-2% coffee solids (so about 300-600mg, or 380mg solids for optimal brewed coffee and 600mg solids for espresso). That would be the upper limit if the coffee solids were pure caffeine, but they can't be.

On the other hand,the caffeine is probably more soluble than the other parts of the bean,so even though it's only 1% of the mass of the beans it's probably more than 1% of the dissolved solids. That's the hard part. Coffeechemistry.com says brewed coffee ends up with 8-15 mg caffeine per oz and espresso 30-50 mg caffeine per oz. That's a pretty huge range. It means caffeine is 2.5-4.0% of the solids (380 mg) in an ounce of "optimal" brewed coffee or 5.0-8.3% of the solids (600 mg) in an ounce of espresso.

So: I typically make a nice strong 10 oz cup of coffee with my 12g of beans containing a maximum of 120mg (or 1% of the coffee mass) caffeine. So let's say this maximum 12 mg caffeine per each oz of coffee corresponds to 1.65% or 500 mg solids (1.65% of 29.57 gr = 500 mg). In this case the 12 mg caffeine is 2.4% of the solids. If I brew it "too weak" and it ends up 1% solids (or 300 mg per oz) I would guess I'd proportionally get no more than 60% of the max caffeine, 7.2 mg (again 2.4% of the solids) per oz or 72 mg for a 10oz cup. So that narrows the range to 72-120 mg of caffeine in my cup depending on the brew strength.

I'm guessing a good cup ends up somewhere in the middle, so I'm going to estimate my typical cup contains around 100 mg of caffeine from its 12g dry coffee, or about 80% of the maximum caffeine present in the beans. From this I would suggest a decent rule of thumb for "optimal" strength American-style brewed coffee is around 0.8% (or 0.008) of the mass of the beans used, or 80mg of caffeine for each 10g of dry coffee.

Bonus corollary/sanity check

Using the same method, a shot of espresso made with 7g of ground coffee maxes out around 70mg caffeine (77mg if 10% Robusta sometimes typical for Espresso blends), but the serving size and strength are more standardized than brewed coffee, so the 64mg/shot reported by the USDA and others also ends up being 80-90% of the caffeine present in the beans.

  • 1
    One suggested adjustment to this calculation for future readers- the paper referenced found 1% caffeine by mass for green coffee beans, not roasted, so this calculation should take into account water weight lost during roasting. If we assume 20% weight loss but otherwise keep your assumptions the same, then we get 10mg caffeine per 10g dry coffee, which also happens to be an easier rule of thumb.
    – Joe
    Commented Jun 29, 2019 at 16:28
  • If you eat that scoop of 12g Arabica beans, it would contain about 92 beans which would contain about 552 mg of caffeine, not 120 mg. I think it is obvious that the chemical reaction of ground beans with water is not able to typically extract more than about 27% of the caffeine. Source: caffeineinformer.com/caffeine-content/coffee-beans
    – Matmarbon
    Commented Jun 10, 2020 at 11:18

Estimation is all well and good, but if you really want to know exactly how much caffeine is in your cup, you're going to have to measure it!

Basic Mode

You can get caffeine testing strips that you can dip into coffee, and they'll give you an idea of how much caffeine is in the brew.

D+Caff is one brand I found but there may be others. They look like this:

enter image description here

The strips have a couple of lines on them, marked D and C. Decaffeinated beverages will only rise to the D line, caffeinated beverages will rise above it.

You do have a problem that you're going to have to bust out some science to calibrate this thing yourself - out of the box it will not actually measure the amount of caffeine in a drink, but if you try it on some drinks with known amounts of caffeine you should be able to work out a system of measurement.

Advanced Mode

  1. Get yourself a Mass Spectrometer.
  2. ??????
  • 2
    +1 this is funny and the ???? link led me to some interesting stuff.
    – Nathan
    Commented Feb 3, 2015 at 20:11
  • 1
    Underpants gnomes?
    – apaul
    Commented Feb 4, 2015 at 0:05
  • Protip: be(friend/come) someone who works in a lab with a (GC)MS. It's expensive equipment and having someone who knows how to use it is a big plus. Obviously, the GC is an added bonus to do some aroma analysis.
    – JJJ
    Commented Jun 27, 2019 at 1:04

Well I am not sure how thay did it but here is the website about the amount of caffeine there is in certain cups of coffee (types will vary): Caffeine In Coffee
There will be caffeine intake of other drinks as well.

Well here is another website where it explains the amount of coffee per type (not brands). It was made by ICO, the International Coffee Organization: http://www.ico.org/caffeine.asp

  • That's an interesting comparison but it won't help me figure out my French Press. It does show that different restaurants serve beverages quite varied in caffeine per fluid ounce. Tim Hortons brews at 10mg/oz and Starbucks at double that.
    – Nathan
    Commented Jan 28, 2015 at 1:55
  • How about now ? Commented Jan 31, 2015 at 13:52
  • Why the downvote? I thought my second website made my answer more clearer. Commented Feb 6, 2015 at 0:33
  • 2
    Both of those sources give a very wide range of caffeine numbers and are hard to compare to my brewed coffee at home... but thanks for answering quickly!
    – Nathan
    Commented Feb 6, 2015 at 1:35
  • 1
    @PhytonMaster, the link is broken. I tried a quick Internet search but cannot find an alternative location. Maybe you know where it is.
    – MTSan
    Commented Nov 18, 2016 at 15:30
  1. Determine what kind of coffee you are working with. Different kinds of coffee have different amounts of caffeine. USDA-accredited values are:

    • Normal - 40 mg
    • Decaffeinated - 1 mg
    • Espresso - 212 mg
    • Decaffeinated espresso - 1 mg
    • Instant - 3,142 mg
    • Instant chicory - 21 mg
    • Instant french - 246 mg
    • Instant mocha - 360 mg

Sourced from Google which pulls its data from this USDA link - Basic Report: 14221, Beverages, OCEAN SPRAY, Diet Cranberry Juice. Tinker around and you'll find the other types.

The values are for 100 grams. This is best because g can be directly derived from an SI unit.

  1. Measure the amount of pulp. If your grounds are already wet, you have coffee pulp. Figure out how much you have by weighing it on a scale; it will help if you keep the pulp in the filter so that it does not spread and create a mess.

  2. Convert the pulp into dry measurements. Do this by subtracting the amount of water you added. You can find this amount by calculating exactly how much coffee you initially put into the coffee maker, minus the amount of coffee that was produced. This amount will be very small, but it will affect the mass of the final pulp. Measure out an equivalent amount of water (which is usually measured by volume rather than weight) and weight it on your scale, being sure to subtract the weight of the container from your final measurement. Subtract the weight of the water from the weight of the pulp.

  3. Multiply the dry measurement by the amount of caffeine in that type of coffee. If you made instant coffee and have 2 grams of dry instant, for instance, you will multiply this into 3,142 and divide by 100 for a final answer of 62.84 milligrams of caffeine in the coffee pulp.

Voila! Provided you are pretty skilled at handling and measuring things, you have an accurate answer. :)

  • Do you have a source for that USDA data? Why weigh "pulp" instead of dry grounds? Your list doesn't include normal coffee beans, only "espresso". But your step 4 is pretty much what I'm looking for...
    – Nathan
    Commented Feb 2, 2015 at 16:22
  • I've added the sources and also the amount of caffeine for normal coffee beans. :)
    – qedk
    Commented Feb 2, 2015 at 17:53
  • Looking at this more, the numbers make no sense and don't square with other sources. For example, there is no real difference between ground "coffee" and ground "espresso" so why does one have 40mg and one 212 mg? If it's measuring prepared coffee, why are the instant ones so high? This answer doesn't make sense.
    – Nathan
    Commented Feb 3, 2015 at 20:20
  • Well, I have no idea. USDA data. The links: goo.gl/lk2bPY goo.gl/gqrncP goo.gl/5yhLmD
    – qedk
    Commented Feb 4, 2015 at 12:33
  • goo.gl/SY5ydO goo.gl/UVUh2G goo.gl/jk63Dd
    – qedk
    Commented Feb 4, 2015 at 12:33

It really varies by size, bean origin, roast method and other factors.

To get some general idea, here is full table according to Wikipedia:

Caffeine Content in Select Food and Drugs

And here is a full list according to Center for Science in the Public Interest:

caffeine content by Center for Science in the Public Interest

  • Thanks, but this is the sort of easily googlable info that I can't really use for my homemade coffee, which has a bunch of variables compared to everything here.
    – Nathan
    Commented Feb 20, 2015 at 20:07
  • Isn't that similar to my answer? Commented Feb 23, 2015 at 23:18
  • @PythonMaster Nothing wrong with similar answers (and certainly nothing wrong with an answer like this that puts the information on the page; this is generally encouraged as it prevents link rot).
    – user80
    Commented Feb 24, 2015 at 1:06
  • It looks like a duplicate of answer though Commented Feb 24, 2015 at 1:09
  • I think the question title is ambiguous and misleading and it's not clear unless you'll read the description what's about. Somebody will buy a Starbucks Coffee and will expect that this question to answer 'how much caffeine is in his cup', so this answer explains that. If you want me to remove an answer, please change the question like 'how to calculate/measure caffeine in the coffee I make', then it'll be more appropriate for the first answer.
    – kenorb
    Commented Feb 24, 2015 at 10:44

To be honest, with all of the variability in coffee, if you want completely accurate numbers, you'd have to prepare multiple samples of your coffee brewed the same exact way (preferable 3 or 4 samples to be sure of consistency), and send these off to an analytical chemistry lab, where they will use a Mass Spec instrument (maybe GC-MS) to determine content. The reason you don't see more cafes displaying numbers for their caffeine content is because this testing can get expensive.

Your specific example of brewing with a French press will also have trouble being determined because we don't know your grind size (exposed surface area), and we don't know your dose. Caffeine yield is directly related to dose.

Also that Mayo Clinic study is very, very, poorly cited because it just gives names of journals and years, not specific studies, issues, or authors. I based on there being no direct trace to its source, like just listing multiple sources that are properly cited, I would say that that article is not a reliable source of information.

  • Complete accuracy isn't necessary. Just wondering with all that variability, how someone would compare their caffeine from a few cups of coffee to a clinical guideline like "300 mg/day"
    – Nathan
    Commented Feb 2, 2015 at 16:24
  • If you're not going for complete accuracy, then err on the upper end of the estimate and say that there are 200 mg per 8 oz. Which would mean that 24 oz of coffee would meet that guideline.
    – John Snow
    Commented Feb 2, 2015 at 21:36

This should help you arrive at a more accurate number.

According to caffeine informer:

standard drip brew yeilds an average of 145mg of caffeine per 8 fl. oz.

french press yeilds an average of 107.5mg of caffeine per 8 fl. oz.

The difference is a 58 to 43 ratio.

Now, assuming most of the coffee brands listed on their caffeine database are brewed using the standard drip method, you could calculate the average french press ammount of caffeine per brand of coffee by taking the 58/43 ratio of depreciation into consideration.

let A = the average ammount of caffeine yield from french press

let c = the ammount of caffeine reported using standard drip method



According to here, there is 180mg of caffeine in an 8 oz. cup of Starbucks brewed coffee.

Based on these numbers, there should be approximately:


of caffeine in an 8 oz. cup of french press regular Starbucks coffee brewed at home.

  • 1
    Thank you, that's the sort of thing I was looking for. But this method doesn't include the amount of dry coffee grounds, which can vary considerably. Also, I'm not sure Caffeine Informer is a reliable source—it differs from other sources and is overtly biased
    – Nathan
    Commented Dec 1, 2015 at 0:50
  • @Nathan Yeah, at first I thought they were caffeine enthusiasts like myself and then I found out otherwise. They have a sources page with links to federal databases and I also understand that consumer reports has some data as well. Also, the industry standard is supposed to be 1 tbsp per 6 us fl. oz. but who knows what some of the different places use as McDonalds has about half the caffeine of Dunkin Donuts and they both use drip brew arabica as far as I know.
    – mchid
    Commented Dec 1, 2015 at 2:17
  • @Nathan Remember, the larger the sample population size, the closer the statistics get to a true and accurate mean or something like that.
    – mchid
    Commented Dec 1, 2015 at 2:25
  • @Nathan Kind of a late reply but I have some new information. 1 tbsp per 6 fluid ounces of water is for [usually] cheaper robusta or robusta blends that have more caffeine and 2–2.5 tbsp per 6 fluid ounces is recommended for arabica and I believe this is supposed to compensate for the difference in caffeine between the two varieties.
    – mchid
    Commented Apr 2 at 18:53

Arne Preuss at Coffeeness in Berlin shelled out for a laboratory to analyze the caffeine content of 15 different brew methods on the same roast batch and published his results at https://www.coffeeness.de/en/how-much-caffeine-in-coffee/

I did some math on his results and got Pour over: 12.33 milligrams caffeine / 1 g coffee grounds French press: 15.56 milligrams caffeine / 1 g coffee grounds

Some arabica beans have more caffeine than others so it's rough, but this is the only laboratory-backed info I've found so far.

I posted a question to Arne's blog post yesterday which has not been approved yet, so doesn't yet appear in the comments. Here is my math from my comment, by which I meant to ask him if he agrees my math is correct, or if I missed something:

Your pour over brew ratio is 30 g grounds / 500 ml water. The caffeine result is 74 milligrams caffeine / 100 ml brew. That's the same as 370 milligrams caffeine / 30 g coffee grounds, which is 12.33 milligrams caffeine / 1 g coffee grounds If I brew with 30 g grounds, that's 370 milligrams caffeine

Your French press brew ratio is 18.3 g grounds / 320 ml water The caffeine result is 89 milligrams caffeine / 100 ml brew. That's the same as 284.8 milligrams caffeine / 18.3 coffee grounds, which is 15.56 milligrams caffeine / 1 g coffee grounds If I brew with 30 g grounds, that's 467 milligrams caffeine.

If I hear back from him, I may try to nudge him to post results in the form of milligrams caffeine per 1 gram coffee beans, by brew method, on his site, either in that article or a new one for people like us. You want to nudge him too?

For now I am ignoring how water density changes with temperature (I've asked at what temperature Arne measured the 50ml brew sample) and that the coffee grounds absorb about 1.5x their weight in water. Brew yield ml is less than the water added to make the brew.

  • Welcome to Coffee! Can you describe a bit more which results you used from that link and what mathematical operations you applied to arrive at those numbers? While the link is very informative, answers on this site are expected to be somewhat self-contained. Even with the numbers you provide it's not really clear what assumptions were made (by the study in the link and in your calculations). Adding that nuance would be really helpful for others stumbling upon your answer. :)
    – JJJ
    Commented May 5, 2022 at 23:11
  • 1
    It seems that you've created a new account to add further elaboration. Seeing that the other answer builds on this one, I've taken the liberty to edit that information into this one. You would normally be able to do this yourself, but you may need to register your account (or register a new account) and then use the contact form (as explained here) to link all this post to your new registered account.
    – JJJ
    Commented May 7, 2022 at 23:27
  • Hi Sue and @JJJ ! I'm glad my few hours of nerding out about this question almost 9 years ago are still interesting. Arne's experiments definitely seem more rigorous than my napkin math, yet the comments section is full of additional questions. Clearly some real research is needed :)
    – Nathan
    Commented Nov 29, 2023 at 0:12

Not the answer you're looking for? Browse other questions tagged or ask your own question.