Alright, let’s talk about EK43s then

Dale Harris of HasBean wrote a post called “We Need to Talk About EK43s“, about the current wave of interest sweeping through our industry in Mahlkönig’s EK-43 grinder.  I understand the sentiment behind the post, but I also disagreed with a huge amount of it.

For an industry that proudly tweets about the importance of dialogue, we’re generally terrible at long form discussion of any kind.  I want to make it clear that this is a disagreement of ideas, and not an attack on Dale or HasBean (both of whom I think are rather wonderful.)

Further disclaimers:  My company sells these grinders.  We use these grinders for all QC, and have done for some months now.  We did host Ben Kaminsky’s talks in London, but we did not profit from them in any way.  Ben Kaminsky is a close friend and someone I spend a lot of time talking to about coffee, because he often makes me feel stupid.  Which is good.

I’d recommend reading Dale’s entire post, but I am going to quote heavily from it throughout.  This is going to be pretty long, so either hit “Read Later” or let’s begin…

At the WBC in Melbourne Matt Perger, a super smart guy along with his coach Ben Kaminski [sic] (also super smart) presented a routine filled with new ideas and demonstrated a vision of how the espresso machine could be used: by using a grinder not usually associated with espresso preparation and by allowing a different control of extraction parameters. Around the same time, a few people I have great respect for have embraced some of these new concepts, putting them on bar alongside or to replace their existing menu.

Ben has run some training events in Europe, discussing the what, why, and how of these techniques and the drinks they create—I have not been to these sessions; I wish I had.

My first concern is that Dale hasn’t attended these events, and is speculating about their content from 2nd degree sources.  Some of these sources actually contain a few inaccuracies about the events, but that is by the by.

It’s been proposed that the EK43, a grinder designed years ago for the spice milling industry which exhibits improved grind particle size distribution can be the solution for poor (or poorly controlled) extraction for both filter coffee and espresso brewing.

Without giving Ben’s class away, I would disagree here.  Using this grinder and techniques allows us to increase the extraction of well roasted coffees into areas of the brewing control chart previously considered negative (i.e. above 22%) while seeing an increase in quality, rather than a decrease.

I have two issues with this:

1. Is it a good solution to the problem?
2. Is the problem really a problem?

I am not sure the problem is ever really clearly defined here.  Are we saying that underextraction of coffees is the problem?  I would argue that with many conical burr grinders extraction is a problem if you prefer lighter, but developed, espresso roasts.  Unless you’re roasting to tickle second crack you’re unlikely to easily exceed 19% extraction on a conical burr grinder.  This seems something of a shame if there is more tasty stuff that we’re knocking into the trash at the end of each shot.

Question No. 1: Is it a good solution to the problem?

The innovators and early adopters of these techniques and set-ups are getting tasty coffee with the EK43. That’s great! But, if we’re talking about it as a “solution” for both filter coffee and espresso, I’m finding it very hard to buy into as an idea.

Fair enough.  However, I’d be more comfortable with the dismissal of ideas after testing, rather than reading about online.  Before my own experimenting I had placed a hard limit on extraction at around 20%, and was extremely skeptical when told that coffee could taste good above that.  Testing revealed I was wrong, and my parameters have changed accordingly.

It is undeniable that the grinder achieves a different grind distribution and allows greater control over extraction % and different flavour properties in the cup–we can and should debate whether or not these results are improvements (only however whilst sharing the same drink, not different roasts of different coffees, brewed with different water in different time zones) but that doesn’t mean this grinder is a good solution.

It isn’t just control.  What probably hasn’t been made clear is that higher extractions simply aren’t possible with some grinders.  You can keep grinding finer, and seeing an increase in extraction, but then as you get finer still you actually start to see a decrease again.  There is a hard limit for many espresso grinders.

I don’t, however, buy into the EK43 as a grinder for espresso preparation. Here’s why:

  • New grinders are expensive – how many more cups of coffee do you need to sell (or how much more do you need to charge per cup) to cover a £1500 investment in a tool you happily lived without yesterday

I’m not really following the logic of this statement.  I lived without things in the past but their addition to my life has been an improvement, and I now consider them indispensable.  As for the cost argument, you could approach this from an actual answer – assuming most cafes make around 10-13% pre tax net, then you’d need to sell between £14,000 and £18,000 worth of coffee (including VAT). Depending on cost per cup obviously, but let’s just say 8,000 drinks for now.  This is before we factor in the reduction of waste.  For most people 8,000 drinks would take about 160+kg of coffee.  Most people’s coffee waste sits between 10-15%, and if that were reduced.  This could be reduced to significantly less waste, increasing yield per kilo.  You’d probably require 25kg less coffee to do the same amount of drinks (so multiply the wholesale price of whatever coffee you use by 25 for a bonus saving.)  Now work out the savings over the course of 1, 3 and 5 years.  I suspect the grinder paid for itself relatively early on.

I’m aware the above back-of-a-napkin maths is massively oversimplified, and can be torn to pieces quite easily – the point is that in a well run business should be able to renew its equipment and invest in things that allow it to increase its efficiency and profitability.

  • This grinder is not made for espresso service – we’re taking a tool designed for a different purpose and pushing to the limits parts of it that just aren’t made for high volume use – things will break (and will thus be more expensive) in the long run

It isn’t made for espresso service, on this I absolutely agree.  I think it is unfair to speculate on reliability without any data though.

  • The blades are sharper and more expensive—they will dull faster. Let’s say burrs for a standard on-demand grinder are £80, these burrs are closer to £350, and likely to need replacing more often to maintain the virtue for which they are being prized

I’d really like to see where the data on burr age is coming from, as in my experience the way a burr ages is a complex mixture of burr material, quality of cutting, size, volume of coffee used, grind setting used etc etc.  If there is data that is available on this then I would welcome it.

  • Using the grinder is labour intensive – it takes longer. Does the subjective quality of the shots produced add up to enough benefit to charge more per drink your staff make? Even if slower service is acceptable to customers, it still reduces the amount of drinks a person can make within a given timeframe

Currently it is indeed more labour intensive.  I haven’t done the maths on how the reduction in waste tallies against the increase in labour cost, and obviously every cafes relationship between price spent per kg and wage paid per hour is different.  There would likely be some offset, if not a net gain.  However, no one is really proposing this grinder in this incarnation as a final solution.  It is simply a step forward.

I can’t reiterate the above point enough, and it should be noted that those experimenting with it are pretty much alpha, perhaps beta testers.  If you are not willing to participate with some risk, and a high incidence of bugs and errors then don’t buy a grinder yet.  No one is selling this as a finished solution, and to make out like anyone is creates a logical fallacy best known as the straw man argument.

These are incredibly boring reasons to supposedly write off the future of coffee preparation; in fact, I understand deeply the argument that cost should never be the deciding factor in a business focussed on delivering quality. However, it is still part of the equation: I don’t give a shit if you sold the best coffee in the universe if you’ve closed due to bankruptcy when I follow trip adviser to your door.

Is £1500 spent going to damage an historic, famous landmark coffee bar of the speciality movement? Probably not, but it could be the margin between success and failure for a less established business, and slower service and insufficient capacity are not something customers are crying out for right now.

The endemic financial unsustainability of serving coffee drinks is a problem, that needs more work for a solution.  However, if you are taken out by a £1,500 hit then you are either grossly underfunded, terrible at cashflow management or (and this is frankly extremely rare compared to the other two) you aren’t sufficiently profitable.  Expenses like this are inevitable in business, planned for or not.

I hope I have made something of a case above about why thinking like this and experimenting like this can lead us towards increased profitability and an increased likelihood of customer engagement.

If we want a more uniform grind size for espresso brewing, or even a fully uniform one, there needs to be better understanding of what’s causing higher uniformity of grind particle size (burrs, motor, direction/angle of burrs) and then the integration of those factors into a grinder body that allows on demand dosing so we can stop pissing around making coffee and focus attention on serving[1] coffee.

Well, obviously.  No one is arguing to the contrary here.  No one is claiming this is a finished product for the applications we are using it for.

Coming back to filter coffee and moving away from the grinder, to a topic I’m less familiar with – coffees shots: do we need to be brewing filter through an espresso machine? Are the flaws in manual and automated filter brewing so high that the solution would be to brew coffee exclusively through metal baskets attached to pump driven pressurised machines because at least that way temperature stability and pouring technique aren’t issues?

There are obvious implications in using espresso machines to brew coffee to filter coffee style volume and TDS% goals:

  • Time: whilst brewing filter coffee through many methods is slow, at least you can do it away from your espresso bar

What is the advantage of that?  Increased movement in a cafe is generally a bad thing, and economy of movement should be prized as a part of workflow in labour intensive operations.

  • Water usage: cafes in many european cities using RO are quite aware of the effort it takes to produce perfect water for brewing[2] more water volume through your espresso machine means greater storage requirement of filtered water – there goes another cover

This is incorrect.  If you require 250g of water to brew a filter coffee then an RO doesn’t become more or less efficient depending on whether that water will be heated in a counter top water boiler or an espresso brew boiler.  In fact, our tendency to rinse filter papers means that an RO would need to produce more water to brew a pour over.  (I put most people’s rinse at around 100g of water).  Pour over coffee uses more water than a coffee shot, and therefore produces more waste from an RO.

The easier point to make here would be that in most scenarios the energy usage in the water boiler would be better, as they are generally more thoroughly insulated than espresso brew boilers (and the fact they are not properly insulated on so many machines drives me insane!).

  • Using tools in a way they are not designed: I don’t believe the design of any modern temperature stable espresso machine has figured for regular output of 250ml shots – some of the older ones (pre 1940) were designed for it – but PID’s weren’t included

This will vary dramatically from machine to machine, but I would argue it is significantly more repeatable and controllable than the water coming from a pouring kettle.  Again – actually testing this (hard to do without messing around with your Scace device), combined with actually tasting the results, would be better grounds for dismissal at this point than speculation.  I admit to having retained a hangover of skepticism around coffee shots, despite the fact that I’ve had several that have been exceptional and unique in certain qualities to that brew method. (In a good way.)

As well as flaws, there are virtues to the way we currently brew filter coffee. I enjoy the clarity of flavour that reduced undissolved solids allow in my Chemex, and I appreciate the difference between my consistently well brewed 6l drip brewers output[3] and the cup profile I get from my Eva solo. In fact, I adore tasting a coffee a different way and finding a completely different reason to love it.

Dismissing the clarity of a properly made coffee shot, without tasting one, is a little unfair.  No one is proposing that these methods disappear, however I’m not really sure either is suited to a typical cafe environment.

Personally - and I fight internally over this all the time – I don’t think I want every cup of coffee I drink to come from a box behind a bar, with the promise that it was ground at that point the only remaining nod to ‘prepared for you’ the same way as every other drink being served from there was. That doesn’t suggest to me great engagement between staff and their product and it doesn’t help me appreciate that making coffee involves knowledge and craft. In fact, pre-portioned doses ready for any possible menu item reminds me more of the McFlurry machine.

I don’t care about how my coffee is made if it tastes delicious.  Our continued fetishisation of brewers we consider fashionable, regardless of customer experience, often in the name of “craft” is a problem for us.  Most of our customers really don’t care about the prep work, they care somewhat about the product and absolutely about the experience.  If we can spend a little less time pouring self-absorbed swirls of water over coffee, and a little more time being nice to people, then that would be great.

Question No. 2: Is the problem really a problem?

 I don’t want to be the guy who stands up for the status quo, for the good enough, and for the adequate – I don’t want to say that selling coffee is a numbers game that requires fast food style service and high volume to succeed in a market with high rent and low unit value…

 I am, however, going to question whether coffee made and served the way speciality cafes have been trying to over the last 10 years is that bad.

 Let’s start by clearly defining my bias:

  •   I’ve had delicious espresso, more than once
  •   I’ve had delicious filter coffee, more than once

These seem like really silly things to write down – I would seriously ask why anyone would passionately drink, let alone sell coffee if they couldn’t say this – but bear with me, you’ll see why it’s important.

 None of these delicious drinks needed modal grind size, or low grind size distribution – they needed good coffee, and people working hard with the goal of delivering consistently something they thought was nice tasting and worthwhile.

I have said this every year I’ve been in coffee, and every year I’ve worked to set myself a new benchmark.  I’m interested in better in every aspect of what we do.  This does not invalidate what I enjoyed in the past, but I have no intention of settling at any point.  So I do not believe the above to be a valid argument against experimentation and innovation.

I am arguing that we are able to serve delicious coffee if we so wish, and whilst different taste goals should and can be explored, there is value in what we’ve built so far.

Another straw man I’m afraid, because no one is arguing that we’re not capable of delicious coffee now. No one is making a point contrary to the above.

Question No. 3: Were We All Wrong?

 In separate discussions with two people working in bars who are using the EK43 in part of their service – individuals who I think talk a lot of sense about coffee and have served me many tasty things –  words were said that caused an almost primal reaction within me.[5]

Within one conversation  it was suggested that, ‘the coffee tastes different, I’m not sure it always tastes better, but we’re selling it because we want to extract the best we can and this may enable us to do it.’

I can’t speak to other people’s quotes.  I would be happy to go on the record to say that I have tasted shots and brews that exceed the quality of what I have experienced before on traditional equipment.

Why would we stop selling something we enjoyed yesterday and replace it with something we’re not sure we enjoy – isn’t that what the back room/ R&D is for?

As above – we’re beta testing.  Those who wish to go live with it are welcome to make that decision, but they make it for their own business.  No one is saying you have to do anything different from what you are doing now.  This may solve something you consider a problem.  It may not.  Get as much information as you can and try to make an informed decision.

Within another, ‘as part of adopting this we may need to develop our understanding of what good espresso tastes like’, that perhaps we’ve been apologising for or skirting over the fact that espresso was acidic, when in fact it was our inability to brew correctly.

I don’t really understand this point clearly.  Higher yields have resulted in increased balance and mouthfeel, and increased sweetness for me.  (There are certain other attributes that wreck my head, and how I understand espresso, but I promised not to give Ben Kaminsky’s entire class away…)

When I judge espresso as tasty it is a reaction to  flavour and texture, not a filing of it against a context we’ve established as a movement. When espresso has acidity and we enjoy it, that’s because it’s nice. When it’s too acidic, too bitter, or even too sweet, we should adjust our recipe, adjust our technique, or not serve it.

What we enjoy from coffee is up to us, and arguably some of our positions allow us to influence what some people drink and how much they enjoy it – but the market (or our chosen part of it) will always tell us what they really think in the long term.

The invitation will always be open to come and share espressos with us here, and discuss what we collectively taste.  Writing off how an espresso tastes from an EK, without ever tasting it, is a problem for me.

’m not saying that those using these new systems are wrong, or that it won’t be part of or inform our shared coffee future. In fact, I’m proud to work in an industry that has many people, trying many new things, and I can’t wait to taste them all—Innovation is good, taking (calculated) risks is good, following your nose, pallet, or gut is good.

However, I’m shamed by our rush to embrace a silver bullet to fix our problems, by our desire to be the coolest kids on the block, and our fear of being left behind.

Here is, for me, the real meat of Dale’s post.  This is something that I absolutely agree with.  Please don’t jump on the EK43 bandwagon without a thorough understanding of what is being experimented with, and what the potential ROI is.  There is no rush.  This is a new idea, this is early days.  There may be competitive advantage in it (as outlined above), but that inevitably comes with caveats.

Lets get this straight – if you buy into this now you will not be the first person to do this in a shop, there will be no first place prize, the novelty is not valuable – in 6 months lessons will have been learnt and the cutting edge will have moved on, possibly leaving scars, possibly some people gloating at my idiot naysaying and almost definitely some tidied up lessons that may or may not include aspects extolled today as progress.

Agreed again. 100%.  I’ve tried, repeatedly, to differentiate innovation from novelty (something we seem to struggle with as an industry).

 Doing something expensive, risky, and different because someone else thinks it’s the future of your industry isn’t clever—it’s certainly not innovation, taking a calculated risk, or following your nose. It’s playing follow the leader where the leader doesn’t know you, how your business operates, how your staff work, and who your customers are.

I’m disappointed by our lack of faith in the drink we gave to someone yesterday, and I’m pissed off that we still think a new toy will fix everything when the proven reality, not just in coffee but in every industry, is that growth and improvement come with small, boring improvements and an aggregate of marginal gains, not an old deli grinder.

Again, more agreement.  However, I believe that it is also disappointing to see something so thoroughly dismissed without proper testing or hands on time.  I agree with so much of the sentiment, but I believe it is absolutely healthy (and essential) to be dissatisfied with what we are doing now – if our businesses are focused on quality.  When that goes away we start doing stupid things like talking about “the perfect cup”, and we discover all too late that we’ve been left behind (something we see in this industry often, as well as many other industries.)

Summary:

Some of us are playing with the grinder.  It is being used in a way it is not designed to be.  We believe there are commercial and qualitative upsides to this. We do not think we have a market ready solution to various problems. We are excited. We believe that we have access to a higher tier of cup quality, with a few caveats and challenges along the way.

If you aren’t ready to suffer through alpha and beta testing: please wait.  Doing this without an an understanding of what you’re trying to achieve will decrease your quality and consistency.

I will welcome any rebuttals, arguments or otherwise.  I would like to thank Dale for starting a conversation that I hope continues.  I like that he writes in long form often, publishes independent thoughts, and has his own voice and opinions. I look forward to the continuing discussions he will doubtless generate.

10 Comments

  1. Pingback: UK Models
  2. Pingback: Vincent Malfitano
  3. Pingback: Jeff Halevy
  4. Pingback: Laura Glading APFA
  5. Pingback: book publishing
  6. Pingback: Kion Kashefi

Comments are closed.