My name is James Hoffmann. I write about coffee, and about the coffee business. Most of my time is spent focusing on Square Mile Coffee Roasters in London.

I also travel and talk a lot on many of the subjects I write about here. For more information or to get in touch click here.

Planning for 2014

December 27th, 2013

I’m not a New Years resolutions kind of person, but I do like a little planning. Years are as good an arbitrary unit as any other, so it seems appropriate to be looking ahead to what I want to achieve in 2014. I don’t plan on sharing every single goal but I can probably summarise things as: 2013 was a year where I talked a lot, had ideas (enough that a few might even be considered good) but I did not do. In 2014 I want to follow through on things, achieve things and feel productive. We shall see how I measure up or whether I fail publicly on that front…

As for this blog, I have some thoughts too. I know I seem to vacillate often, and have angst ridden moments about writing on here but answering the question “Why do I write on here?” can be tricky when the answer changes as often as it does.

My biggest worry is that I just use this place as somewhere to complain about things. Looking through the last year of posts this isn’t really true, but looking through my half dozen unpublished posts the fear returns.

I’m at the point where the negativity around coffee online has started to bother me, and I have to decide whether I want to contribute to the snark or be more positive. I don’t want to get to the point where I join the other cowards in the industry in starting an anonymous twitter account so I can berate people who I think are “getting away with it” when it comes to quality or the like.

This blog started as a vehicle for my own learning, and in 2014 this blog will turn 10 years old. My learning hasn’t slowed in the last 10 years, in fact it has accelerated every year. This blog doesn’t reflect that properly, which seems a shame but something that is entirely within my control to change. Let’s see how I do…

Coffee Blogs I’m Reading

December 4th, 2013

There was a question on twitter the other day asking what coffee blogs I’m reading, and I think the answer probably requires a blog post. In the past I’ve posted enormous, long lists of blogs that I subscribe to but that probably isn’t helpful if you want to add a couple more to your reading list or go back through the archives of a couple to learn some stuff.

I’m going to presume that most people are following both Sprudge and Dear Coffee, I Love You.

Edit: I should have included Roast Magazine’s Daily Coffee News in this. (An embarrassing omission!)

Cafe Owners:

I would recommend following along with Kaffeine’s blog, a cafe in London whose weekly menu blog post has evolved into an ongoing insight into the world of running a busy, successful cafe.

Colin Harmon’s blog straddles this and as well as wider industry stuff and should be followed too.

Maxwell at Colonna and Smalls also writes often about the ideas and challenges of communicating and serving great coffee, so subscribe to this.

Edit: Forgot to add the Prufrock blog to this list.

Coffee Industry:

After a brief hiatus the Coffee Lands blog is back. Subscribe. Learn.

Kevin Knox’s blog is great, because it is opinionated and challenging. Do I agree with everything he writes? Actually a surprising amount, but calling it Coffee Contrarian wouldn’t be the best name for it if everyone was supposed to agree with it.


I also enjoy the rants in between the trip reports over at TheShot, I disagree with him a lot, get exasperated and annoyed by it occasionally but what’s the point of reading things that simply confirm your views rather than challenge them?

On the review front – a different approach and a stunning blog worth following or hunting through if you’re travelling is frshgrnd. An eye for design, great photos.

Finally I’d recommend following Mike White’s tumblr, which contains simple reviews and brutal honest about the coffee he drinks each day.

What did I miss?

I miss the days of google reader, which would allow me to organise my list by how regularly they posted and then I’d unsubscribe from dormant blogs. These days my list of blogs in the RSS reader has dropped massively from the days of 150+, but it feels like there just isn’t that much being written out there.

If I missed something then tweet me a link! My only request is that you link active, not dormant, blogs that write interesting stuff. If I have missed someone painfully obvious then I will update the blog post and apologise profusely…

I’m not calling it theft

November 11th, 2013

A hypothetical situation: You’re visiting a nice coffee shop in a different city, and you get recognised as a coffee person. You’ve queued up and you’ve got your $3 in hand ready to pay for your coffee. Instead your barista, seeing as you’re industry, gives you the coffee for free. You do what you think is the honourable thing and you put your $3 in the tip jar – you’re trying to be a good person (and you don’t want to look cheap…).

Now imagine a slightly different situation: this time the barista takes your money, but doesn’t ring the drink into the till and when you’re not looking just puts the cash in the tip jar.

The second one of these feels quite explicitly like theft, an employee diverted money from the business into their own pockets. Yet in both situations the business, the barista and the customer are in identical financial positions. You could argue that the intention of the barista wasn’t to pocket the cash when they comped the drink, just to offer some nice hospitality, and I completely side with the baristas that have kindly offered me drinks for free over the years. I just looked at these scenarios and could help but notice that financially the situations are pretty much identical.

The employee’s decision to comp a drink prevented the business from earning money. There is one situation in which this is ok – where there is an explicit and clear policy for comping drinks on a bar, but I don’t think nearly as many businesses have these as they do have staff who give things away.

Let me be absolutely clear: I am not calling baristas who comp drinks thieves. That would be ludicrous. I actually believe this is a flaw in the operations practices of many businesses, and just something I thought would make for a fun discussion. The only ones who deserve any blame are those who fail to set proper policies for this sort of thing within a business.

The difficult middle

November 7th, 2013

A couple of years ago I began to think quite a lot about longevity. I was seeing my own business, and many others, reach a point of maturity where it didn’t really feel fair to call them startups any more. I looked at my industry, and similar ones too, and was inevitably drawn to businesses that had been going so long that it seemed fair to call them “institutions”, for want of a better word.

What really defined these business for me wasn’t that they had simply been going for a long time, it was more that they had transcended from the requirements of a single owner. These might be businesses so old that not a single person involved in the creation of the business was still there, though the business maintained success and its principles regardless. How does a business go from being a startup to being an institution? This is what I mean by the difficult middle.

I began to read more about business longevity, and unsurprisingly there is plenty to read. One particular idea captured me, and seemed painfully applicable to the coffee industry. It was the idea that a business should be treated like a living organism, and nurtured and taught. Coffee is an industry incredible dependent on knowledge for the delivery of quality, but our focus has all too often been the education of the individual rather than the business itself.

If you’ve done any sort of barista training for a living then the following probably rings true: Training baristas is like pouring water into a bucket that has a hole in the bottom. The typical practice of most cafes is to invest reasonably heavily in training the staff, either in-house or with the assistance of a supplier such as the coffee roaster they buy from. The problem with this is simple: People leave.

This is a problem for any industry, though the relatively high turnover in coffee makes this effect more pronounced. When the “star” barista in a cafe leaves the average quality of drinks tends to drop. This is not good at all, for the business or its customers.

In any ideal world the business itself could act like a giant bathtub underneath the buckets, collecting as much of the information as possible that is being poured into the staff.

How do you teach a business?

This is a difficult question with an answer that most people won’t like: For a cafe I believe that a large part of that is structure, systems and manuals.

Every cafe should have a dialling in procedure for espresso in the morning.  That procedure shouldn’t be whatever the person dialling in learned to do in their last job. There should be an ideal method that is taught, trained and used. Having a single staff member understand a refractometer and matching software has a certain value, but having best practices available to anyone in the company as a permanent resource has a much larger value.

This does happen, but all too rarely in coffee. The places where I do see good structure tend to be the places with better than most of their market when it comes to quality, consistency and financial health.

So, will a manual fix everything? No. It is simply a piece of the puzzle, but there should be an over-arching goal of trying to retain as much knowledge as possible outside of the transient part of the business – the humans. It is about our mindset when it comes to learning, training and teaching.

Much of this applies to roasting – as we learn and explore the roasting process there needs to be more than an oral history passed down from one roaster to another. This requirement means thinking about explaining what we’re doing and, much like the process of writing this post for this blog, it forces us to unpack an idea in order to repack it neatly, with a clear narrative.

This doesn’t mean you don’t need good people, or that automation is a key to success. We have a challenge in coffee, and while we should definitely work to improve retention of staff, that isn’t a very long term solution.

Very few businesses start out to last for 200 years. Very few startups are thinking as far as 5 years ahead in terms of planning and structure. The future is a terrifying place most of the time, but I do wish that I could give this advice to myself 5 years ago. Starting good habits early, building better systems in from the beginning, would have gotten us so much further more quickly as an operation.

Three Things I’m Working On

October 19th, 2013

As the second excruciatingly  long day of HOST draws to a close, I realise that I should probably write a little bit on here about the three things on show at the event that I’ve been involved with. In each case I should make clear that I’ve been very lucky to be involved with people and companies who’ve done the real heavy lifting on all of this stuff.

Victoria Arduino Black Eagle 0388

Over the last couple of years I’ve developed a close working relationship with Nuova Simonelli, initially as part of the grinder project (I’ll come to this in a moment), and then through the workshops we did together around the world over the last 18 months. I confess I had been nagging them about espresso machines for a little while, and pushing for an opportunity to collaborate with them on something. What we have on show at HOST is the first results of that collaboration.

The machine is a Victoria Arduino, not a Nuova Simonelli, because there was an opportunity to take that brand in a new direction. Obviously the design is a departure from their other machines, though it was the same original designer, as we pushed for a different design brief. We wanted it to be as low as possible, while still stable and serviceable, though NSF does place some restrictions on how long your legs could go. The main feature we’d been working on is what they are calling the gravimetric system. Essentially we have scales in the drip tray using live beverage mass to control and terminate the espresso shot. You can program the machine to deliver 35g of beverage and that is what it will do. If you want to pull a 40g shot then you just program the button to do that through the control screen.

If you’ve read this blog for a while you’re probably not surprised that this is the direction I want to go. In all my experiments and testing I’ve come to the conclusion that beverage mass remains incredibly important to taste, extraction and consistency. Up until now we’ve tried to control beverage mass using either time, volume in the cup or volume of water sent to the group as the control point. None of these offer the consistency I’d like, or believe we need to replicate for each customer the espresso we’re excited to serve them.

What we’re showing here is not even an alpha prototype in my eyes. The technology needs a lot more development and improvement to reliability and accuracy. HOST was a great opportunity to present what we’re working on but we’re being very clear that this is not yet a finished product. It is incredibly important that once this machine is for sale that we’re not presenting a beta testing unit to early adopters at full price. Industrialisation is a massive step in creating a new machine and one that requires patience and resources, which thankfully the Nuova Simonelli guys have. I’m excited for what the future of this machine holds. Don’t expect to see this for sale for a year or so, though you may spot a few test units in the field before then… (I’m loathe to talk too much about products that aren’t for sale, but it does seem right to explain my role).

I don’t have a good camera with me, but do head over to this Sprudge article because they have some nice ones.

The Clima Pro grinder

The grinder project was my first involvement with Nuova Simonelli, and I was a late addition to the team of Colin Harmon, Gwilym Davies and Fritz Storm. I see this as an ongoing project, and this is the first product.

Up until a year ago I was of the common mindset that heat is the enemy and should be removed at every opportunity from the grinder. Yet this grinder has a small heating element in it, which seems somewhat counter-intuitive. However, heat is inevitable when grinding coffee, and heat buildup is still going to cause the grind setting to change, as it moves from cold (at the start of the day) to warm during service, back to cool in quiet periods. By both heating and cooling this grinder we’re able to keep the grind setting more stable and have the grinder act in a more consistent way through the day.

Is it perfect? No. Is it the solution for every cafe? No. Is it a step forward in helping cafes achieve better consistency, and waste less coffee during service? I hope so.

I dialled in at 8.30 yesterday morning, before the show opened. It was a pretty busy day, but by 1pm I’d only made a single grind adjustment. It’s consistency was borne out by a machine brewing its doses to the gram and showing the shot weight and time for each brew. It felt too good to be true, and at this point if you want to dismiss this as a single anecdote and not useful data then I wouldn’t protest. The next stage is now more testing in locations around the world, and a release isn’t anticipated until early next year. If it is appropriate I will write more about it then.

I must disclose that I have an ongoing financial arrangement with Nuova Simonelli, though I’m sure that is pretty obvious. If you want to factor this into how I’m presenting this information then I won’t protest.


A while ago I went to Marco and begged another favour. In the past we’d worked together on what had become the Uber boiler, and I was grateful they were up for something again. In a cafe I want to brew various coffees by the cup, but I don’t want the inconsistencies or labour costs of pourover brewing or other methods. I want to get to tasty quickly, easily and repeatably. I asked Marco if they’d build me something like a batch brewer that brewed only a cup at a time. (i.e. use the same technology for consistency). This fit pretty well with a project they were already working away on in R&D nicknamed “Splurty” and at HOST they showed 2 units of the new machine. It can brew up to 400ml in up to 4 minutes (or quicker if you want) and can use existing brewers like Hario, Kalitta or Chemex. I’m excited to get the first couple soon to start playing with.

There are some great pictures again on Sprudge over here.

The common theme between all this is not that I want to automate the barista out of a job, but that I want to make getting to great coffee more easily. I’m tired of fighting coffee, I’m tired of serving coffee that we know could be better and I’m tired of drinking disappointing coffee as a customer.

This is a bit of a quick post at the end of the day, before heading out to dinner, but I wanted to write something now. I probably have a lot more to say, and to explain, about each project but I will save that for when people can actually get their hands on them. You may have questions – feel free to tweet them at me! I may not answer quickly, but I will get there… (or if you’re in Italy then come and see me on the booth tomorrow and I will make you some coffee.)

On Loyalty Cards

October 10th, 2013

This post is not going to be particularly kind to loyalty cards. Let’s start off with the first problem: we’ve got the name wrong.

Are we really creating “Loyalty”?

They aren’t really loyalty cards, they’re more accurately described as discount cards. They are a fancy and convoluted form of coupons. Great coffee shops tend not to think of themselves as the sort of places that offer discount coupons but that is what loyalty cards are.

Let’s not fool ourselves: no one is loyal because of the card. No one is going to avoid other coffee shops simply because of the fact they carry a single loyalty card in their wallet for their favourite cafe. Chances are they have several, and are just slowly working the system to earn the free drinks.

What upsets me more is the idea that a discount on coffee is the best way to create loyalty. Loyalty programs came from companies with a commoditised product who were looking for a competitive edge. The whole point of speciality coffee is to make the point that we have a differentiated product, rather than a commoditised one. If we’re earning loyalty on price, then we’re likely to lose it on price too. If we earned that loyalty through quality and through experience then it will be so much harder for a competitor to win that customer away from us.

The discounts

Coffee is sold, in most of the world, on pretty tight margins. The only way to generate considerable return is to increase your scale as typical coffee business will achieve 8-12% of their net sales as pre-tax profit. The discounts we offer on our core products through loyalty cards are pretty astonishing.

Let’s compare:

Number of drinks to fill card Discount
5 16.7%
6 14.2%
7 12.5%
8 11.1%
9 10%

Those are pretty staggering discounts on a product that doesn’t have huge margins to begin with. If you have a buy 7 and get your 8th free and a well intentioned barista throws an extra stamp on there the average discount for every drink bought using the card becomes 25%!

Now a ten percent discount only translates to a 3.8% increase in cost of goods (as a percentage of the drink), but an already squeezed business just has its life made much harder. For staff who want to be paid more, there is simply less money in the pot.

If you want to get into the detail you should probably cost out the cost of the cards, the ink and the time taken to do it (all of which do add up over a year).

Big Data

This isn’t to say that there are no redeeming features in loyalty programs, but I don’t think many independent businesses are entirely comfortable with some of the opportunities. For many larger companies a loyalty program gives access to data about the user that can be exploited for increased profitability.

The Harris and Hoole iOS app is an interesting inbetweener in all of this, and I think is worth paying attention to. I should state now that I know next to nothing about future plans for the app, or about what data is being collected and analysed. This is speculation mixed with experience.

When you download the app you’re informed you get a free coffee (essentially the digital loyalty card has already been filled once for you). The app asks you what your favourite coffee is. I picked a small, black filter coffee.

When you arrive at, or near, a location you check in and can send your order to the till – you don’t have to be inside, just close. By the time I got to the ordering point the barista asked if I was James. She then asked if I wanted to use my free drink, which I did. She then told me it was already being made and I could collect it from the second of their two coffee stations. Impressive.

Interestingly they brew a specific coffee for black filter coffee drinker – I didn’t realise this until I asked someone afterwards, but the fact they gave me a naturally processed coffee was interesting. Knowing I wasn’t going to add milk allowed them to showcase a more interesting and risky coffee than they otherwise might have. The experience was fluid, and once they add payment then they’ve removed a great deal of hassle from the entire experience. For locations with time constrained people (i.e. the financial district), I could see this making a lot of sense.

As I said, I don’t know what data they’re collecting. Simply putting a digital, location aware loyalty card in your customer’s pockets yields interesting opportunities. As Harris and Hoole are connected to Tesco I hope they’re doing more with the data. I certainly would want to!

I’d want to look for trends in customer behaviour and ordering patterns. Do they visit the same store every time? Do they always order the same thing? Do more customers order black coffee drinks on Mondays, allowing me to drop my milk order a little bit – or be prepared if people tend to drink larger beverages midweek? If I alert a customer to a newly opened store when they walk past it, coupled with a reminder that they’ve nearly hit their free drink, can I capture a sale that might have been lost to a competitor?

I’m sure there are huge limitations on what we can currently do with data, and how accurate we can be, but there is no doubt that the collection of information about us and our habits is hugely important for many businesses. Whether we approve or knowingly consent to this data collection is a large, but separate topic.

Get rid of loyalty cards?

I want to make this clear: I am not criticising people who have those programs in place. If I had opened a cafe in the last 5 years I would undoubtedly have thought long and hard about doing one. Opening a business is terrifying and it is natural to look for every opportunity for success.

Nor am I saying that people should get rid of them. Taking something away is incredibly difficult to do, because it is hard to reward a customer for a behaviour and then tell them you think this behaviour is bad. This makes people confused and angry. We don’t want that.

I would urge people to consider them carefully, if only from a fiscal perspective, and be fully prepared for their impact on margins and opportunities for lost revenue. I don’t want us to think about price and discounts driving loyalty, because it takes away focus from what we should be good at.

I expect lots of people won’t like this post. Lots of people will accuse me of criticising their business, which I’m not, or tell me that their program is very successful (which, on the condition that they’re ok with the discount, I hope it is).

I wanted to write this because Simon from Electric Coffee wrote to me about them, and got my brain thinking (thank you) and I thought I’d share those thoughts here.

Handing Over Control

September 26th, 2013

I was talking about Rene Redzepi’s talk at the NBC last night and had a realisation, that perhaps is not staggering or new but was probably worth writing down.

Rene remarks that people have no objections to being fed live ants, but get very angry when you stop them from having milk in their coffee. From the outside it does look a little crazy, and you would think that the best restaurant in the world should be able to serve everything exactly as they want it.

What I think is missing from this dynamic is the idea of control. When you eat at a restaurant like Noma your meal begins when you hand over control to the chefs. This is the glorious part of it – you are entirely in their hands, to experience food as they see it or as they want you to see it. You’ll be adventurous because you the experience leads you to believe that your trust in them won’t be abused and they won’t serve you something that will make you unhappy.

Then there is a moment in the meal that a restaurant hands back control to the guest. Typically the food has finished and the experience is coming to a close. It is usually signified with something along the lines of:

“Would you like anything else? Perhaps some coffee?”

From this point on the guest is in control of their own destiny and making decisions, and when we deny them milk with their coffee we are implying that we think they have made a poor decision. We are criticising them. They tend to react badly when they feel criticised for asking for something they habitually enjoy and consider to be good.

This isn’t really a post about how Noma should just make coffee a course, not an optional addition – though I think they would encounter way less resistance in getting people to experience the coffee exactly as they want.

This is about looking for moments when a guest places their trust in us, places control of their experience in our hands. It is also about realising that most of the time people feel (quite rightly) that they are in control and our objections are heard as criticisms and judgement of who they are.

The Productivity of Online Discussion

September 23rd, 2013

As I mentioned in my reader survey results post – I’ve been planning to write something about the viability and worth of online discussion. I’ve been through the “why I removed the comments” thing a few times. I’ve also experimented with things like Branch to see if that worked better.

People generally bemoan the state of online discussion, and often the blame seems to fall on twitter. We all spend our time microblogging (this is a stupid word) and our conversations on there are fast paced, global but ultimately shallow and often somewhat smug.

I’ve thought a lot about how online discussion could work better, and I’ve come to a conclusion that many people won’t really like: It isn’t possible.

Useful, interesting, engaging discussion simply doesn’t suit the form and fit of the internet. It suits small groups of people being in the same room at the same time. In my limited experience of life I would say the smaller the number the better. Discussions that have shaped my thinking, inspired me, astonished me or left me in awe have usually been with a 1 to 3 other people. This is the nature of genuine conversation.

The internet doesn’t work well for that. It is full of voices, and its democracy both helps and hinders. On the one hand, all voices are equal so having a certain level of reputation or kudos is not required to get your voice heard. On the other hand, all voices are equal and intelligent, experienced opinion can be drowned out by noise and nonsense. This happens all too often.

“Ah,” you interject, “but what about those really great discussions you had in your comments section in the past? What about the Naturals Debate blog post?”

I would argue that that wasn’t really a great discussion. Some very smart people wrote short essays on a subject, and kindly allowed me to host them. (Those in favour of moderation as a solution to comments should remember that I could easily have prevented any of those from being shared based on my own whims and beliefs. I don’t think that possibility really belongs in an environment for great discussion).

While the part of me that enjoys the bickering aspect of debates, I generally thought the whole thing got more confusing and less helpful as it tried to turn into a actual conversational discussion. Reading through them again it feels quite fragmented outside of the main back and forth between Peter Giuliano and Geoff Watts. There are people trying to pull the debate in all sorts of directions, those how just want to chime in with their own opinions (sometimes of dubious relevance). Questions are asked that are inevitably lost, because they are not asked to anyone in particular. When you ask that semi-rhetorical question in a conversation you generally will end up working towards an actual answer. Yes – great information was shared, but I would argue that Geoff should have written a blog post on a site he or his company owned, to have full control over his own words and opinions. Finding and read it would not have been much more difficult than scrolling through dozens of comments to find it.

Does this all mean that I see this blog as a place to broadcast my ideas from, to have them go unchallenged? I don’t think so. I think the internet is incredibly good for sharing ideas, and when an idea is good the internet will do an incredibly good job at spreading it to an enormous number of people.

I would say the back and forth on the subject of the EK43 was great when it turned into long form essays from various participants. Points could be communicated clearly, and there was none of the fuzzy confusion of the discussion on the same subject happening on twitter.

I believe you’ll learn more from discussing or arguing about one of Colin Harmon’s blog posts with a friend or colleague in the real world, than you would from reading comments about it online. (Note: Colin is used as a hypothetical example here, but you get the idea…)

Importantly, I don’t think this is particularly unique to coffee. I see the same issues in the other industries that I’m interested in. The internet is a vast archive of ideas and information, categorised and impressively searchable. It isn’t a great vehicle to drive discussion about those ideas. I believe this is an opportunity for trade associations like the SCAA, BGA or SCAE to step in. I go to SCAA each year for a lot of reasons, but top of the list is the half dozen truly great conversations I will have when I am there. That is worth the price of admission and travel to me. After all – a trade association should probably help those of us in the trade associate…

I don’t think this is going to be a terribly popular blog post, and by my own logic it is going to difficult to communicate why you think I’m wrong back to me. Perhaps you’ll write a blog post about it (which I’d happily link to on our here and on twitter). I do hope, however, that there will be an opportunity in the future where we might be able to sit down over a cup of coffee, or a glass of something inebriating, to actually talk about it in person. I think it would be a worthwhile discussion.

Popular Threads

Other business