Water-to-coffee ratios: a Coffeed throwdown!

[nick]

I think a lot of folks have been itching for another lively topic on Coffeed. There's something that's been on my mind lately, and I think I'm just about at the point where I'm ready to step up to the dais and throw down the gauntlet. Ready? Here it goes:

60 grams of coffee per 1 liter of water.

I'm hoping to avoid pointing fingers at anyone in particular, and I have a feeling that people will probably step up to defend their practices, but it's frankly been bugging me a bit. 60g-per-liter. I'm seeing more and more folks with much higher dose (or 'throw') ratios--up to twice as much coffee-per-liter.

Most of the time, throws seem to go up because people are trying to brew coffee with shorter dwell/brew times than the 'old school' 3.5-5.0 minutes. They grind finer, and don't like how the cup tastes, so instead the throws go up and they like it much better.

These extra-high-throw brews seemed to really become popular with the Clover owners, but it seems pretty popular with siphon brewers, manual pourover, etc. What those brewing systems have in common is shorter brew times.

We're at right about 22g for 11.5 (340 mL) of water and a 3.5 minute total brew with our Abid/Clever dripper brew bar, and the coffees have been absolutely positively bad-ass. That ratio is admittedly ever-so-slightly higher than 60g/L (it's 64.7 g/L), but we'll be experimenting more this coming week.

I've got an ExtractMojo on order, so I'm hoping to learn a lot more using that in conjunction with our taste buds, but in the mean time here's my throwdown (admittedly) snarky challenging question:

Updosers... why must you waste so much coffee?

[Oliver]

I use 10 grams per 6 oz
so that would be be 53.33 grams for a liter.

I think it tastes awesome!

[Mike White]

From M. Petracco of Illy in "Coffee- Recent Developments" (page 143):

"Every time that a partially soluble solid is kept in contact with a given amount of water, at an appropriate temperature, not necessarily coincident with the boiling point, for a considerable amount of time, while allowing the concentration of solubles in the liquid to increase throughout the operation. Due to the law of mass action, the extraction rate decreases as the concentration increases, making an overly prolonged decoction ineffective and, as a side effect, possibly unfavorable to flavor and taste purposes due to volatile losses and hydrolytic changes."

Nick I think your question is dependent on extraction method. There's too many variables at play to assume that each brew method can abide by the same golden rule.

[Andy Schecter]

Nick, thanks so much! I have been meaning to post about this exact subject (on Portafilter.net, remember that place?), but I haven't gotten off my butt to do it. I will echo what you're saying, though, and add that I hope AVOIDING WASTE becomes an important issue in every serious coffee program.

My disappointment in the wasteful habits of coffee people goes back to the Seattle SCAA show. There at the Barista Guild booth I watched slack-jawed as "world-famous" baristas grossly overfilled their portafilters with artisan-roasted coffees, then swiped off the excess directly into the knockbox.

And this cavalier attitude certainly didn't end at that show. Nowadays third wave shops, in order to serve their customers Cup of Excellence coffees on an instant coffee timetable, massively updose their single-serve brewers in an vain attempt to make up for the ultra-quick brewing cycles. Sure, they can get the brew strength up that way, but the updosing inevitably produces an inefficient, underextracted brew. It wastes good coffee!

A lot is made nowadays of Relationship Coffee, Direct Trade, working closely with the farmers, etc. Do these cherished growers understand what happens to their coffee? Does the laborer who just struggled to carry a back-breaking load of hand-picked cherries down the mountain know that 25%, or 35%, or even 50% of their coffee is pissed away into the compost heap because of our absurd overdosed, underextracted brewing practices? How does it make them feel when we tell them? Do we have the guts to tell them?

There is a method that works admirably to quickly produce excellent quality single-serving coffee almost instantly. It's called THE ESPRESSO MACHINE! You probably already have one, and, used with sensible dosing, it works and works efficiently. Why aren't more shops offering an espresso-based, COE americano menu?

The Starbucks Automated Brewer looks cool and works quickly but doesn't extract properly. Because of its low-pressure design, open top architecture, and inadequate temperature control, it inevitably underextracts. It's an impressive gadget from a geeks point of view, but it does an automated, very precise job of wasting coffee, time after time. I feel badly for the independents who have invested tens of thousands of dollars in their Clovers, but honestly, it's the wrong tool for the job.

Some of you have worked very hard on your Automated Brewer parameters, and, using great coffees, you get good results. I bet almost all of you could get better results with traditional brewing methods, using lower doses. Of course it would take a few more minutes to brew each serving. But one has to ask, are business models that are based on inefficient use of the raw material "sustainable?"

Nick, you mentioned ExtractMoJo. For those who aren't familiar with it, it's a simple system (consisting of software and one or two special instruments) that helps the cafe owner to measure coffee strength and extraction yield, with the specific aim of optimizing coffee quality and minimizing waste. I'm AMAZED that more people haven't spent the few hundred bucks to get into this. Considering the huge sums of money you've sunk into your shops, it is a small investment that will pay off handsomely.

ExtractMojo also works for espresso. Once it gets wider use, super-updosed (super-wasteful) espresso techniques will hopefully go away, too.

[DISCLOSURE: I am definitely not impartial about ExtractMoJo. I played a limited role in helping develop the software. I received no money, had some fun, and got free instruments, software, and coffee.]

Finally, Nick, 60g/liter is not etched in stone. It depends on your personal taste, how strong your want your cup to be, and where you prefer your extraction yield. It will probably be slightly different for every new coffee you brew. The 65g/liter that you're using now may be right in the ballpark if you like your brew a little more concentrated.

Please keep in mind that the coffee-to-water ratio is only half the story. The other, crucial half: what percentage of the original dry coffee dose is extracted into your cup?

Whew. Rant off. Time to sit back and get my comeuppance.

[phaelon56]

This discussion may also be helpful to those roasters trying crack into the restaurant and office coffee market. I recognize that the coffee provided by many office and institutional coffee services is not high end but if it's an "okay" all arabica blend it should be far better than Folger's, Maxwell House et al. I typically have coffee or espresso at home in the morning and do without in the office but on occasion need a caffeine bump mid afternoon.

Our office coffee truly sucks and the reason became apparent this week. The fractional packs aren't labeled for weight of contents but I saw the master carton - 1.75 oz of coffee for 64 ounces of water. That is NOT a typo.... 50 grams of coffee to 64 ounces of water!

Why does this matter to the independent roaster? Because you may need to educate prospective customers who are puzzled by the "price per pot" or "price per cup difference between coffee service coffee and the better quality coffee you hop to switch them over to.

I've been opposed to the Clover concept but was always shocked at the amount of coffee used for a single cup. And I've had perhaps three very good and one stellar cup from the dozen or so cups I've ever been served from Clovers - a ratio not too impressive relative to what I've achieved using Melitta single cup pour-over at home.

[Aaron Ultimo]

Been thinking about this a lot too as I play with my new ExtractMojo, Abid, Chemexs, and Beehouses. Look forward to seeing where this discussion goes. I'll try to post some more when I'm not too sleepy on our own experiences at Ultimo.

[James Hoffmann]

I currently like the range of 60g/l up to a max of around 75g/l for some coffees.

I've probably been as guilty of anyone at slightly updosed underextractions in the past. Never got on with very high dose syphons though - my best cups came from lower doses.

In my head I find underextractions a bit depressing because I know I've left some goodness behind in the coffee that could make a positive contribution.

(And Nick, you ask the waste question way better than I did a few years ago!)

[scottlucey]

oh man that post about the bga ad is classic!!! a feisty one!

[Peter G]

All this talk about "underextracted" and "overextracted" begs the question: what is optimal extraction, which is the frame of reference for both the under- and over- versions?

Most people look at the control chart, and just accept it, not understanding the source of the "proper extraction" window. I think we should all understand that completely before throwing around the "underextracted" word.

Somewhere north of 30% of roasted coffee (by weight) is extractable by water. Everyone agrees that if you maximize that extraction, extracting all or nearly all of that material, you wind up with unpalatable coffee- as it turns out, there are some pretty yukky tasting compounds in your coffee. As it happens, many of those ugly tasting compounds are also difficult to extract, so if you reduce the efficiency of the extraction, you reduce those compounds in the cup, and voila better coffee. How far do you reduce this efficiency, though? If you reduce your extraction efficiency too much, don't you waste flavor compounds and leave the flavor "undeveloped"?

These days, most coffee professionals would say that "proper" extraction would be between 18-22% of the total soluble material. According to the Extract Mojo site, it seems that George Howell has narrowed that range slightly for HIS standard, to something like 18.5 to 21.5%. This extraction window goes back to the work of the Coffee Brewing Institute which, in the 1950s, did a series of consumer preference studies, which concluded that the majority of consumers like coffee best when extracted in the 18-22% range. This was good, exciting, legitimate work!!

However, it must be pointed out that calling a 17% extraction "underextracted" is relative to that study. Please understand- I support this standard and defend it. I hesitate to be too hard on those who practice a less efficient extraction, however, because I recognize that there might be a time and a place for less efficient extractions when used by a competent professional who knows what she is doing.

I personally find that less efficient extractions can emphasize sweetness and aroma, while de-emphasizing astringency and acidity. This may be just the ticket for some coffees or some consumers. By updosing, you will normally decrease extraction efficiency even as you increase solubles, or keep them the same. Seems to me that some may be expressing a taste preference for a reduced extraction percentage and an increased solubles concentration when using this strategy.

Now, one could theoretically make the same argument with INCREASED extraction percentages (low throw weights) but it seems the only people who do that are trying to scrimp on coffee and bilk consumers, and are rarely doing it for flavor reasons. However, throwing the "waste" word around is imposing a judgment on updosers, folks who are presumably making decisions based on flavor feedback. Do we want to do that?

All that said, my personal favorite ratio is right at 60 grams per liter. How about that?

Peter G

[John P]

Our brewing is just by the cup, but I like 12 g per ~8 oz water or 8 g per 5 oz water -- which is a slightly higher ratio, but for the coffees I use it for, it seems to work.

I would think that the combination of dosage/grind depends on brewing method and type of filter medium being used. On drip-brew machines, Gold filters are more porous than paper, but the result on paper filters on Melitta pour-overs may be different because the barista controls the dwell by how they pour, and different mesh fineness on various press will probably yield different results using the same dose, not to mention the different cloth filters and glass rod on the vac pots.

In regards to the ExtractMojo, I am interested in how grind affects extraction on brewed coffee. Can you get the same extraction with (slightly) less coffee and a finer grind? Can the manipulation of coffee, grind, dwell time in different ways produce the same results?

[nick]

Good discussion! Still waiting to hear from more folks on the high-throw side of things.

However, it must be pointed out that calling a 17% extraction "underextracted" is relative to that study. Please understand- I support this standard and defend it. I hesitate to be too hard on those who practice a less efficient extraction, however, because I recognize that there might be a time and a place for less efficient extractions when used by a competent professional who knows what she is doing.

Thanks for making this point, Pete. This is part of what I was hoping to see discussed.

I can see that there could be a time and a place, as you say, and I'd love to hear from folks who believe they're doing that. But that's not really what I'm talking about. I'm talking about increasing the dose beyond 'typical' throw-ratios in order to get a palatable brew out of a non-traditional brewing device.

(Hmmm... let's look this up... okay, here it is!) A 2009 Lamborghini Murcielago gets 8 mpg city/13 mpg highway. A 2010 Toyota Prius: 51/48. Obviously a wildly different price for each, and they're both about getting from point A to B. However, the experience that comes from driving the Lambo is incomparable to driving the Prius, and while it might not justify the fuel consumption to some people, no one can dispute that the experience is wholly different.

However, I have to confess that I don't "get it" when it comes to high dosing in siphon and Clover brewing. The goal seems to be shorter brew times, but the best I've encountered of siphon or Clover brew falls well behind the best examples of a brew with a more "standard" brew time.

So what I'm sort of (and sort of unsuccessfully) trying to say is that what I feel like I'm seeing out there is a Prius that's getting 9 mpg, and it doesn't drive any better... it just sucks gas. Expensive and admittedly visually-interesting contraptions, with people proliferating techniques that seek shorter brew times at the expense of more coffee... I still haven't had a great cup like that. Not a great analogy, but I hope you get my point.

I've actually have had some great siphon brews... but at 3-3.5 minute total extraction times. I always wanted to try some Clover brews with 3-3.5 minute brew times. Is saving the one or two minutes really worth the extra trouble and the extra coffee?

In regards to the ExtractMojo, I am interested in how grind affects extraction on brewed coffee. Can you get the same extraction with (slightly) less coffee and a finer grind? Can the manipulation of coffee, grind, dwell time in different ways produce the same results?

Me too! I used to believe that you could dial up or down the grind to compensate for time going longer or shorter, but I'm starting to doubt that more and more.

Now I think that there's a correlation between turbulence and grind size (which after all, is about adjusting the surface area, so turbulence vs. grind size makes total sense), but I'm finding myself doubting whether fining-up the grind means you can shorten the brew time.

I mean, you CAN, but you won't get the same thing. Roasting a chicken vs. roasting the same chicken cut into pieces... sure the latter will cook faster, but it's not the same.

I've got my ExtractMojo on the way, and one of the first things I wanna play with is dialing in two different 'pathways' to the same solubles-yield by adjusting grind size and brew time. I'm interested to see if it's gonna be taste the same or noticeably different!

[John P]

Nick,

in regards to X brewing method and X dosage... For me, getting the correct dosage is such a multi-faceted thing that we may end up with a greater number of generalizations rather than some specific, all-encompassing, forumula. However, if the ExtractMojo does what it says, it will allow each user to easily pinpoint dosage for their coffee and their equipment.

If you have to change the grind between press and drip, is the dosage or the dwell time more likely to change in order to get the same results in the cup? OR because each brewing method IS different, shouldn't we expect different ratios, and flavor characteristics in the final cup? I would also be interested to know if certain brew methods are best at a particular ratio, and if you make the press ratio the same as the drip ratio, will it taste like @ss.

I encourage you to do about a month's worth of both necessary and unnecessary experiments and testing and present your findings in a live-streaming video.

[Andy Schecter]

it must be pointed out that calling a 17% extraction "underextracted" is relative to that study. Please understand- I support this standard and defend it. I hesitate to be too hard on those who practice a less efficient extraction, however, because I recognize that there might be a time and a place for less efficient extractions when used by a competent professional who knows what she is doing.

Sadly, I believe that many people who are "practicing a less efficient extraction" HAVE NO CLUE what they're doing. They haven't made a conscious decision, they don't understand brewing ratios and they have no understanding of extraction percentages. If they're using a Clover with 45 sec steeping times they are forced to overdose in order to get reasonable brew strength. All the while they're wondering why coffees --ones that are supposed to knock their socks off -- taste good, but not great.

A while back I measured some extraction yields with ExtractMoJo at a gorgeous third wave shop that served coffee from a world-famous specialty roaster. Their Clover brew was updosed at 80 grams per liter and the measured extraction yield averaged 14.6%.

The coffee, an exquisite, expensive Central American, tasted...OK.

Between Clovers, espresso machines and grinders alone this shop easily had $60,000 invested. And then people complain that a $600 Extractmojo setup is too expensive???

I personally find that less efficient extractions can emphasize sweetness and aroma, while de-emphasizing astringency and acidity. This may be just the ticket for some coffees or some consumers.

I agree that "less efficient" (ie, underextracted) coffees can emphasize sweetness, but they also often emphasize sourness, which I don't find personally satisfying. This is true with espresso, too, forcing cafes serving updosed underextracted espressos to restrict their single origin espresso choices.

throwing the "waste" word around is imposing a judgment on updosers, folks who are presumably making decisions based on flavor feedback. Do we want to do that?

Yes, we do. Dosing at 80g/l to get a 15% yield is...wasteful! It's pissing 20g/l straight into the compost bin. Peter, let's stop bending over backwards to be polite and simply call a spade a spade. This thing of ours will progress much more rapidly that way. Otherwise, we could just throw in the towel and let Starbuck's "15th Avenue Coffeeshops" overwhelm us with well-financed mediocrity.

[Andy Schecter]

I've actually have had some great siphon brews... but at 3-3.5 minute total extraction times. I always wanted to try some Clover brews with 3-3.5 minute brew times. Is saving the one or two minutes really worth the extra trouble and the extra coffee?

Is a 3-3.5 min extraction times slow for a siphon? That seems like the minimum time to me.

I'm told that 3-4 min Clover brews yield better results but you have to run the temperature at maximum and cover the brew chamber while steeping to retain heat.

[James Hoffmann]

When we had a Clover on loan at the roastery I ended up doing very long brews, but quickly stopped as it was easier just to use a french press.

All of this does beg an awkward question - if we need to rapidly produce cups of brewed coffee, at peak times, and have them taste great - were we too quick to throw out filter brewers and airpots?

[tim]

On the Clover we are doing 78g per litre (18g on the 18 cl dose which really is about 23 cl in reality) now with great results. We used to dose much higher, but after experimenting for 2 days last week I changed the ratios and grind size drastically.

On Frenchie we do between 60 & 65g pr. l.
Filter brewers I tend to do 70 g with coarser grinds now.

Still I have not got my hands on a mojo, but my taste buds are happier than ever...

[Marshall]

A lot is made nowadays of Relationship Coffee, Direct Trade, working closely with the farmers, etc. Do these cherished growers understand what happens to their coffee? Does the laborer who just struggled to carry a back-breaking load of hand-picked cherries down the mountain know that 25%, or 35%, or even 50% of their coffee is pissed away into the compost heap because of our absurd overdosed, underextracted brewing practices? How does it make them feel when we tell them? Do we have the guts to tell them?

I've heard this many times over the years and have always wondered how many coffee people have actually had that conversation with a farmer. Most of them are running pretty marginal, sometimes subsistence, operations. I may be stepping out on a limb here, but I would expect most farmers (well, maybe not Price Peterson) to think "Hmmm. 35% waste means they have to buy 35% more coffee. It's all good!"

[Andy Schecter]

I may be stepping out on a limb here, but I would expect most farmers (well, maybe not Price Peterson) to think "Hmmm. 35% waste means they have to buy 35% more coffee. It's all good!"

1. The best coffees -- the kind that single serve brewers are supposed to feature -- are always in short supply. There isn't enough great coffee around to make into compost.

2. Long term wasteful practices are not sustainable. Continued careless use of electricity, petroleum, soil resources, and water has often come back to bite us. Do we really want to start with coffee?

3. Institutionalized waste is a terrible business practice. Employees go more by what the management does than what the management says. When they realize that coffee is habitually wasted, it gives them a green light to waste everything else: milk, ice, electricity, time, supplies...the cafe's money.

P.S.: Peter, of course, had a very valid point when he said that not all updosers are "wasting" coffee. Some have made conscious, informed decisions based on taste. I didn't mean to dismiss your point, Peter, and I apologize if it came off that way.

But it's been my experience that quite a few updosing baristas don't have the tools and/or experience to understand the repercussions of what they're doing. That's where the "waste" word may be applicable. IMHO.

[Marshall]

I don't disagree with any of that Andy, just the frequent assumption that farmers would automatically be insulted by having some of their coffee wasted. I work with groups that are heavily involved with helping coffee farmers, and the stories I hear make me skeptical that many would be offended at any practice that increases coffee consumption.

[Ric Rhinehart]

There are a number of variables to consider here and we have generally assigned a wide range of "standards" to each of them. For example, extraction of 18-22% is essentially a 20% range or at least +/- 10% (whatever your statistical bent). The same can be said for dose - over here we sort of like 50 - 65 gms per liter, once again a 25% swing. Soluble concentration - 1.15 to 1.35 % is another one of these almost 20% ranges. I think any ranging of this magnitude ought to account for variations in coffee, roast style, etc. The rest are personal or potentially group preferences. In the US a solubles concentration of 1.35% is a strong cup, whereas our Norwegian brethren might prefer a range more like 1.25 - 1.5%. I have long suspected that coffee professionals who focus on espresso almost universally have a preference for higher concentrations in their drip cups.

Regarding under extraction vs over extraction, the case is fairly clear. Under extraction involves favoring the most soluble solids and volatile by products. Thus we get fast extraction cups that smell great but leave us wanting more - more mouth feel, more sweetness, more depth, more complexity. For coffees that are aromatically engaging, like the Esmeralda or a fine washed Yrg, this process can produce an exceptional sensory experience. Similarly, over extraction, resulting from too fine a grind or too long a steeping time results in distinctive bitter notes that may be painted on a palette of complexity and body, but are nonetheless off putting in their own right.

Most interesting to me is that regardless of perceived beverage strength, brewing method or brewing time, we seem to most often prefer coffee at or near the 20% extraction. An espresso might be 6 times as concentrated as a drip cup, but at 20% extraction presents a parallel kind of balance of flavor, aroma, complexity and sweetness. Go over 25% extraction in any method and see how quickly bitterness rises to the top of the flavor descriptor pile.

From where I sit its OK to voice a standard like 18-22% extraction because it is accurate (although not particularly precise) for essentially every form of brewing. Maybe there is a reason for making a 15% extraction, but it feels like "lite foie gras" to me.

Finally, John if I read your question correctly the answer is no. You can achieve just about any extraction from 0 - 30+% and you can hit a wide range of soluble concentrations, from 0 - nearly 90%, but you can't get a balance of extraction and concentration unless you keep the ratio of coffee to water inside the lines.

[Andy Schecter]

Vince Fedele from George Howell Coffee Company, developer of the ExtractMoJo System, asked if I could post the following info to this thread. I think it's very informative and clarifies a lot of the questions that have been raised:



Greetings to the group, kudos to Nick for starting an excellent thread, and best wishes for great success on your new coffee house!

Peter makes many excellent points regarding extraction and preference. The only clarification I might add is in reference to the statement that "17% is under-extracted only relative to the CBI study cited" (see corrected source, below). Actually, 17% would be considered under-extracted by many other references as well, and we have, unfortunately, seen coffee served by well-known coffee houses well below that level routinely....

The consumer study Peter mentions was not actually done by the Coffee Brewing Institute, it was performed originally by the Brewing Committee of the National Coffee Association, and was not 18-22%, but 17.5-21.2 (with a strength range from 1.04-1.39) see Chart below, note legend).


Original and revised Consumer Profiles, published by CBI and CBC

This study was used by CBI, until it was superseded by what the *Coffee Brewing Center* (CBI’s name transitioned to CBC in 1964) considered a more accurate and relevant, updated study performed by the *Midwest Research Institute* (MRI), and persists as current to this day. It is also relevant to note that many other international coffee institutions have separately derived their own consumer profile studies. Although modeled after the MRI/CBC charting format, they were developed independently, and all concur with the 18-22% range. These include the Specialty Coffee Association of Europe, the Nordic Coffee Brewing Center, and others.

It is interesting, however, that even though the studies generally agreed on the preferred extraction yield range, the preferred strength range differs in all of those studies for their profiled consumers.

Getting back to extraction yield, clearly, 18-22% is not strictly absolute to all coffees in all circumstances, I agree with Peter on that point, but exceptions are generally rare. Is the strength/extraction range really rectangular? Probably not. A more modern consumer taste profile would look something like the attached, check it out….it is only for illustration purposes, but is likely more realistic if performed scientifically for coffee using modern methods.


3D Consumer Profile (Courtesy Paul Songer Assoc)




Same 3-D graph viewed from the top in 2 dimensions. (Courtesy Paul Songer Assoc)


The coffee brewing control chart remains a fundamental tool to easily perform an objective analysis of any brewed coffee beverage, and it helps one to understand and navigate the chart while evaluating and adjusting the factors that influence brewed coffee quality. I agree with Peter that placing a box on the chart to collar the parameters for range of strength
and extraction might seem subjective, and could be argued one way or another because “some” coffees might be excepted from the general population. But the important point is that the brewing control chart is a crucial tool and offers a simple approach for analyzing and improving all brewing methods. The more people understand how to use it, the better all of our brewed coffees will taste. Furthermore, every coffee will brew differently in order to achieve the same brew strength and extraction yield. So, the more you change coffees, and roast profiles, the more you need to measure and perfect each brewing recipe, because the coffee will extract differently when anything changes. Plot the optimum range for the coffees you brew, and then repeat so that you can always enjoy them within that range.


Universal Brewing Control Chart: Peter G’s 1-Liter 60-gram preference

The brewing control chart is a simple graphical representation of the key elements needed to determine extraction yield: Brew Water-to-Coffee ratio (the brew formula line) and Concentration. Strength is in my view purely a personal preference, representing the amount of flavoring material (total dissolved solids) to the total amount of beverage [%TDS=solute/(solute + solvent) x100]. Extraction Yield indicates how much material was dissolved into solution from the original ground coffee, targeting 18-22%. Clearly, a very narrow range of extraction yield governs the best tasting soluble solids…..we think 18-20%, and we center our target solidly on 19%. Why do we center it there? We think it maximizes the natural sugars and sweetness of most of our coffees, which we tend to roast on the lighter side. We center it to allow for natural variations in dose of water and coffee from brewing
and grinding equipment, including single-serve portions measured by hand. Small portion errors change the brew formula ratio, and will therefore tend to move extraction yield slightly, so if one starts centered, say at 19%, then small errors will keep the batch within the 18-20 range, that we think tastes best for our coffees.

Concentration, or strength, is purely personal. I enjoy a stronger cup in the morning (1.40-1.45% TDS), less so in the afternoon or evening. I think it’s safe to say that everyone prefers a sweet cup (19% yield), and no one prefers either an over-extracted/bitter or under-extracted/under-developed cup. I have never agreed that there is “one” perfect balance between strength and extraction, because I view it this way: the final coffee solution (i.e., mixture) of soluble flavoring material (extraction yield) presented at the best level of concentration (strength) is completely based on personal preference and varies tremendously with the coffee used. It is very likely to be in the range of 18-20% Extraction Yield, at the
concentration YOU LIKE for a particular coffee. For example, I prefer Guat Bourbon, Ethiopia Yirg and Indonesian Peaberry at high concentrations, 1.45-1.55, because they all have a very light body. Whereas, Colombias, Kenya and Costa Rica have heavier bodies I prefer at 1.30-1.35% TDS.

However, that said, I don’t like ANY coffee extracted to a yield of less than 18.5% nor more than 20.5% because it never really “blooms in the cup” for me, there is always something missing.

Many people who habitually add sugar to their coffee are actually offsetting the over-extracted bitters they must have grown up thinking was “normal” extraction….when they come to our open houses, some have actually thought we added sugar the coffee, when they tasted ones extracted to 19%. We have measured well-known coffee houses serving under-extracted coffee at 0.91% strength and 15% yield (French-Press, Drip, Clover), or 2.1% and 14% Yield (VacPot updosed and under-extracted = sour) as well as others over-extracted
to 2.87% TDS and 26% yield, in many of these cases, not even on the chart….literally off the page!

There is vast room for improvement. Many coffee drinkers who love their coffee but are not “specialty coffee” fanatics habitually add sugar to their coffee - they are actually offsetting the over-extracted bitters they must have grown up thinking was “normal” extraction….when they come to our open houses, some have actually thought we added sugar to coffees, when they tasted ones properly extracted to the sweet spot, 19-20%. We've had hundreds of "converts" who now drink their coffees properly brewed and black, after years, decades in many cases, with cream and or sugar.

So, you see, as Peter G asked, “All this talk about "underextracted" and "overextracted" begs the question: what is optimal extraction, which is the frame of reference for both the under- and over- versions?”

Hopefully this helps understand the stated question, and provides a few answers. The feedback we’re getting from everyone using MoJo and the new Coffee Refractometer has been very positive to both understanding how to improve the final result as well as customer responses when they taste a truly sweet coffee, black w/o cream and sugar.

If you know where you are on the chart (and you can know now that there’s an instrument that works), you are able to navigate to where you want to be, it’s as simple as that.

Vince F.
George Howell Coffee Co.
GotMoJo?
http://www.ExtractMoJo.com

[Peter G]

I'm afraid this throwdown isn't turning out as contentious as Nick may have hoped.

The discussion above is priceless by any standard, particular thanks to Mr. Fedele for his obviously well researched opinion.

A couple thoughts after reading through the above:

-Vince, thank you for your correction to my incorrect attribution of the original taste study done by the Brewing Committee of the National Coffee Organization. Were the other studies you reference also consumer taste panel surveys? I always assumed these various sources were all referencing the same original taste panel work.

-I've always been interested in the raw data from those studies. It is clear that most preferred extractions between 18 and 22 percent, but how did the preference actually shake out? Is this a statistically normal distribution? This question is in part based on my observation that extractions over 22% taste offensive and "bad", yet the worst anybody seems to say about a less-efficient extraction (under 18%) is "lacking" or "underwhelming". Even professionals who describe taste experiences for a living wind up saying things like "I don't know, I just felt there was something missing."

-As for myself, I see overextraction as a much bigger problem than underextraction. Overextraction produces awful coffee. And it's a hugely common practice, in coffeebars, restaurants and homes all over the country. These hordes of overextractors are ruining great coffee all over this land, producing bitter awful coffee for paying consumers, in the interest of saving a few pennies.

On the other hand, underextraction produces, at worst, "uninspiring" coffee, still containing sweetness and aromatics. It is practiced regularly by a tiny number of experimenting cloveristas, pourover brewers, and coffee fanatics. Ok, which is the problem more deserving of a coffeed thread? I'm just saying. I understand and agree with the comments about wasting valuable taste material...but still.

-One thing I vigorously agree with: coffee professionals should pay more attention to the numbers behind extraction percentages and beverage concentration, there is a tremendous amount to be learned here. The extract MoJo is a huge step forward in terms of coffee professionals having the tools needed to accurately assess these variables, and diagnose brews accurately, correlating taste data with objective measurement.

-I would really love to see some sort of indication of the progressive nature of extraction: i.e. a study showing what flavor components extract at what stages of extraction. We all assume that sugars and aromatics dissolve early, and alkaloids and other bitter compounds extract late. Anyone doing work or have data to support this idea?

-Last but not least: I know there are lots of clover users out there using high doses who aren't weighing in. Kudos to Tim for sharing his formula. I know that I, myself, have had some delicious coffees from clovers using high doses and therefore presumably low extraction percentages. While I respect the opinion of those who have been left wanting by clover extractions, I myself have experienced delicious ones, and I know many of my colleagues- good tasters all!- have as well. I understand that there is some hope that there will be a scrap over whether less-efficient extractions are de facto "wrong", and I am heartened to see that we're all in agreement that, in the hands of a skilled barista, we can push the limits and still get great coffee.

Peter G

[Luke Shaffer]

I respect the opinions of the folks that have commented in this thread, and I hope to get a Mojo sometime soon to see where the brews I enjoy most fall in the spectrum... and where the tools take me from there. I am a numbers guy, and I'd like to have the capability to put a value on what I'm doing and cross reference that with what I taste. Did it taste good by accident, because of something I did... what was it? These are answers anyone pursuing better coffee would want to know. I own/operate a couple of synesso's as well as a few clovers, and other stuff. We invested a lot of money in equipment yes, but I'm still looking forward to the day we buy a dishwasher. I get annoyed when the focus is on the cost of everything, when we as a business are trying to invest our money in areas that give us better control over our product- one cup of coffee. We recently purchased an electronic Mazzer, and it's been fun/nerdy charting our doses that we arrived at by taste versus age of the coffee. That information may come in handy if I ever get time to really look at it.

Opinions are like brewing methods- everyone's got one. I'd also like to put these opinions into perspective. Let's start with "by the cup" brewing. Lot's of people/shops have all manner of by the cup brewing devices. How many of these businesses offer 100% by the cup? By the cup means brew it & serve it, no holding time. I want to hear from those folks first, because they've successfully balanced the equation of quality/freshness/service time/price/choices/profitability/expectations. What do you do? In the retail world it is absolutely a balance. It's my opinion that if I am able to get quality equal to or better than my manual methods I use at home, not hold the coffee for any amount of time, serve it up quickly, sell it at an acceptable price, provide choices, make enough money to pay everyone that works for me, and do it with a smile repeatedly I'm doing well thank you very much (takes a breath). Finding this balance in our market, which admittedly is not nearly as sophisticated as other markets, has been an awesome challenge and a battle some days. I think we're winning people over. I look at every gram of coffee as a precious resource like every hour of payroll, I want it to taste great every time, and I want my greatly underpaid employees to deliver it to our customers flawlessly. If we do this long enough we can pay people better, take a harder stance on the philosophical stuff, and offer even more/cooler/better stuff to our customers. It bothers me to read the bashing/"you're wrong" remarks here, because in my opinion it's easy to see things in black and white when you aren't trying to balance all these the variables.

Sometimes I updose, sometimes I don't. For the record, I'm liking most of my brews in the 66g/L range (Chemex yes, Press little less, Syphon and v60 little more), but I think that may have more to do with what I like about each brewer and being "in the mood" for the profile that I get. That's where I start and I may go up or down from there.

I was just following the thread until I took the "waste" and "pissing" bait. You sucked me in dammit! With Clover, yes we're "updosers" (I'll wear a scarlet "U" on my chest) and I suppose we may "piss" away some coffee to get to a flavor we like in a time that works for our business. I can also see pissing off customers by asking them all to wait 6 minutes for a cup, in which case that great coffee goes unsold and is wasted as whole bean. I also see big airpots of coffee turning over quickly that while it may be fresh it just doesn't have that nuance and finer detail you get with a smaller brew- the coffee was wasted because it could have tasted "better" with a "less efficient" brew method. I also see customers "pissing" in these great coffees by dumping condiments into them without letting it cool and trying it first even though we made polite and friendly recommendations-waste. I see shops all over the internet and in my travels touting their by the cup programs but what's that I see? Bunch of airpots and a whole bunch of sparkling clean by the cup brewers. How long has the coffee been sitting there? 10, 20 minutes? It's wasted. You can hold it longer and sell it, but it's too late it doesn't taste as good- waste. Or you can toss it down the drain and waste it that way. I also see good businesses with the right mix of passion and work ethic go under because they can't build a following "sticking to their guns". What a waste that shop went out of business- they had the best coffee, everything was made to order/table service/no to-go cups, etc.

In a way this is similar to how I get on some ppl's cases around here for still having blenders and giant sizes- I think that is a waste, too, but I can't judge them because they are also trying to balance the "equation of their situation". So flame away!

P.S. -Here's the next bomb, sorta related...

*The Abid/Clever is essentially a slowed down Clover without the vacuum. It is currently fancied (flavor of the week) because it is repeatable and hard to mess up assuming you follow a tasty recipe. It isn't as sensitive to pouring styles like a v60, and you can brew it as long/short as you want. As a result I'd imagine there are many ways you could brew a tasty cup with the Clever, depending on the coffee you're using and your personal taste. With the Clever, the user has control over all the variables- ratio/time/temp/grind, it brews one cup at a time, and successfully removes that pesky sludge from the cup. Like the Clover, the skill lies in tasting and adjusting recipes (assuming you have developed a good/repeatable way to stir your Clover). I like the Abid, I like other brewers too- but I grow tired of the "your way is wrong" stuff.

Excuse me I have now go update our POS system as we're raising our prices tomorrow

[John P]

Luke,

nice post. Now go raise your underpriced coffee.

We have always been a "by the cup" place. For coffee, it's either been Americano, Press (by the cup) (2005-2007) and siphon/vac pot by the cup late '07 to present. The Press, and siphon, have exclusively been served as "for here, and black only". I have been called "The most hardcore operator in America" by Jay Caragay, which I am very proud of, and he can attest, when it comes to by the cup, I don't f*** around. In going on four years of "by the cup" service, I've had no more than five or six ask for cream or sugar, one customer's disrespectful mother have her youngest son sneak it once, and one "I'll punch you in the face" response. Any problems were early on and we quickly learned to explain upfront how these coffees are served when they ask about it. We have never compromised our principles on our coffee. But in all cases, there is a (hopefully) polite and educational explanation of WHY we do it this way, and most everyone "gets" it, or we will happily offer them a wonderful Americano or another beverage instead. But the education is out there... and at some time most who have asked about the "by the cup" service are intrigued enough that that either later in their stay, or their next time in, they can't resist. We've had an abundance of success converting naysayers. I think experience has allowed us to refine our delivery of the information to make it more desirable. I think it's the STORY behind the coffee that often hooks the customer.

Now, because of the attention to the coffee and the individual, it is readily apparent to most everyone that this is how our siphon coffee is served, and for those who come just to have that, they never think otherwise. I think it's about creating an environment and a level of service that encourages a particular behavior. At some point, it doesn't need to be said, it just happens.

I roast everything on site, so, as much as I can, I tailor the roast to the brew. I'm often successful. And other times it becomes part of an experimental espresso blend.

Our ratio has generally stayed the same. We do 12-13 g per about 7.5-8 oz of water. We adjust dosage for smaller cups. There are no larger cups. After doing XX number of these, we use the eyeball test on the water. That's pretty much in the neighborhood of the 60 g per L ratio, but it was arrived at by testing dosages rather than knowing what the dosage "should" be. I only have time to read so much, and I'd rather let tasting be my guide (based on other exceptional coffees I've had) and THEN see where that falls in line with what has been worked out by other brew-heads. The only thing I really changed between the brewing methods is grind and steep time.

I do see a definite need at times for something like the ExtractMojo because there are coffees I have to adjust slightly because they're not quite right with the formula I use. I experiment with dose and steep time, but I always keep my grind consistent... too many variables otherwise. I've never had to go above 13 g or below 11g for 7.5-8 oz. of water. Generally it's steep time that I will change, but it's minimal. But I think this kind of tool would make it much easier to pinpoint the "sweet spot".

I love the marrying of the science and "art" of coffee, but there are several times I have been to a some places that, I believe, have more knowledge than I do with respect to the science, but end up being utterly unimpressed by their coffee. Of course all the science in the world won't help us if we over roast our coffees, or serve old coffee. Frankly, I can't believe anyone is sourcing these phenomenal Direct Trade/Relationship coffees and doing anything but "by the cup" service. It really baffles me.

Q for the many with a larger coffee brain than I:

Do different coffees extract differently due to how they absorb water?

Does the same coffee with different roast levels absorb water differently?

Is 60g per 1 L brewed the same or different than 60 g per 800ml and then 200 ml added after brewing is finished? If different, in what respect? I don't have a grasp on that.

[Andy Schecter]

-I would really love to see some sort of indication of the progressive nature of extraction: i.e. a study showing what flavor components extract at what stages of extraction. We all assume that sugars and aromatics dissolve early, and alkaloids and other bitter compounds extract late. Anyone doing work or have data to support this idea?

Well, you're probably aware of Jim Schulman's comments a while back:

This work was done by Ted Lingle, who grouped the flavors by molecular weight, with the light weight ones dissolving quickly, and the heavy weight ones dissolving slowly. There are two light weight, fast dissolving, families that are fully present even in under-extracted coffee.

* Fruit acids have fruity or floral aromas and flavors, crisp tastes in sweeter brews, and sour tastes in less sweet ones. These dissolve the fastest
* Maillard compounds have the aromas and flavors of toasted grain, wood, tannins, or nuts; and tastes which are sharply bitter in less sweet brews, and warm, round, and malty in sweeter ones. These dissolve more slowly than the fruit acids, but will still all get into even the most under-extracted cup or shot

There are also two heavy weight, slow dissolving families which require high solubles yields to reach their full strength.

* Caramels have caramel, vanilla or chocolate flavors and a sweet taste. Since almost all sugars in green coffee are caramelized during the roast, these are the primary source of sweetness in coffee. Dark caramels, which taste bitter-sweet, dissolve more slowly than light caramels, which taste more sugary. Some light caramels will get into even lower extractions, but require higher extraction rates to be completely dissolved.
* Dry distillates are reduced (burnt) caramels and maillard compounds that become dominant in dark roasts. They have the aromas and flavors of clove, tobacco, peat, or turpeny, a dully bitter, ashen taste in less sweet brews, and a bitter-sweet molasses taste in sweeter brews. These dissolve very slowly, but are tasteable at very low concentrations. Their presence is usually a good reason to keep extraction levels fairly low.

I understand that there is some hope that there will be a scrap over whether less-efficient extractions are de facto "wrong"

I wasn't looking for a scrap. I was hoping that there'd be at least an acknowledgment, not that some extractions are "wrong," but that some extractions are much less "efficient" than others.

And when you combine that with say, a statement like this (Intelligentsia)...

when you see the Intelligentsia Direct Trade logo on our bag, you know how much effort is invested in each bean

or this one (Counter Culture)...

Our vision is to pursue coffee perfection by creating partnerships that ensure prosperity for all people, improving the natural environment and operating efficiently to minimize our environmental impact.

...I personally can't help but think that "efficient" utilization of raw material should at least be in the conversation when developing a cafe/roaster program.

I understand that taste rules, and that brewing speed can be critical, and that made-to-order brewed coffee is revolutionary, and that a sexy machine like the Clover can sell a lot of expensive cups. But, I'm just sayin', please consider extraction efficiency, too.

And I'm ALSO sayin' that I believe 18-20% extractions usually taste better than 14-17%, so it's not only about efficiency. And I'm also sayin' NOT to extract above 20% yield. That would be more "efficient" still, but it wouldn't taste good.

Thank you,