This is mostly a follow-up on a post I wrote around a year ago which explored the idea of set recipes and dialling in, specifically regarding espresso. As well as bearing that post in mind, here I'd like to explore espresso and how I feel about it at this moment in time.

A lot has happened during the past year -such is the current pace of exploration in speciality coffee- and our evolving relationship with it, particularly where espresso is concerned. An Australian barista champion joined forces with an American coffee professional and together they delivered a WBC performance and resulting events that, when you step back and think about it, have had a rather huge impact in the speciality sphere. And not just by creating an 18 week waiting list on a particular grinder. Just over a year ago I would not have contemplated brewing a lungo style shot at all. It definitely wasn’t being considered as part of our menu.

Zeitgeists in coffee have an impact which is sometimes hard to quantify, but we definitely see changes. As a multi-roaster store we are not just deciding how we would like to brew coffee, but are choosing how we would like to brew other peoples' coffee- those people who have roasted it.

The way in which we prepare and taste coffee impacts on what we will have a preference for and what we will work towards, as well as what we will work backwards from. We have to start somewhere. My post a year ago was wrestling with the idea of not exploring variables and trying to set them still. But at the same time we all have a window for brewing coffee, some narrower than others. The danger being that if we don’t have a big enough view then we may miss some of the landscape. But at the same time if it’s too large we may not be able to focus on the details and improve things.

For us at Colonna and Small’s, this window has gotten smaller and in a good way.

Our preferences have most likely changed a little bit, but I think that other things have changed as well. We have embarked on a project which has left us with a better understanding of brewing water and this information dramatically informs the brewing process. We hope to publish this research in the form of a book shortly, but we have also decided more precisely what we want from the coffees we buy and therefore the suppliers whose approach we jive with. This has meant we don’t buy coffee in the same way we did before.

Before we may have been willing to tussle with a coffee to get the best out of it, but now we are more informed about what we may struggle with, say a coffee from a roaster that has significantly different water sources or brewing practices. So I would say we are looking for the same kind of flavour results but narrowing the net we cast within which to find them. This also naturally means we are setting an end point of sorts to work towards.

For a long time the speciality coffee industry has been moving away from the double ristretto style in acknowledgement that it is almost impossible to squeeze the character of the coffee into such a small amount of liquid, in that the extraction will nearly always be low.

A ristretto is easy to write off but the range within the espresso cannon is wider and more varied.

Throughout the last couple of years the focus on extraction and TDS* and their relationship has impacted how we see espresso. The exploration of lower TDS espresso-style beverages has most likely also stretched and challenged the idea of what an espresso even is. For me there are limits on this and I feel that below a certain TDS it just doesn’t taste like espresso. Hence the Lungo we sell. With the right grinder, the right style of roast, brewed at a lower pressure and with a quicker brew time (to make sure we don’t exceed an extraction we feel tastes good) it tastes great, displaying the character of the coffee, but it's not an espresso to me.

Drinkers in our shop have often really enjoyed the lungo, with it not being as delicate and nuanced as the filters nor as intense as the espresso. However, those wanting or with a real attachment to espresso haven’t enjoyed them so much. It has been important for us in store to bill the lungo as a different way of displaying a coffee rather than as a lower TDS espresso.

What’s particularly significant for both the variables of brewing styles/goals and that of water, is that they both will impact on how the coffee is roasted. A couple of the roasters we work with will roast specifically for us, in one's case it's to explore lighter roasts that don’t necessarily suit all of his customers needs. For the other, it's to explore coffee together for competition and for water spec.
Regarding this latter roaster, together we will cup coffees that are currently available or are coming into the country from a variety of importers. We will then choose the coffees we wish to buy and the roaster will roast a variety of profiles, normally this is three distinct and different profiles to guage the direction in which we want to go with the coffee. I will brew these coffees and the process will continue in a feedback loop in which we both take part. So, in short, how I decide to test the coffee combined with the water and the grinder will all enter the feedback loop and massively impact on where we end up.

In essence we are working back from a brewing point. This notion is one that I in some ways rebuked in the set recipe post last year. At the same time it is something we are in many ways now doing by cutting out roasters whose results may not correlate with our water as well as tailoring coffees to TDS and extraction numbers -although we are still marginally flexible on this.
We are all doing this in some way. Roasters, for example, will find a set recipe style they like to work with and their coffee will be influenced by this. This kind of controlled environment makes a set recipe, or a "work backwards from brew" approach, a more linear and predictable path as the testing environment is more controlled.

At the same time, roasts vary, and water varies, even with careful supervision. And small variances in recipe are still something we use. We do this because our main goal is to serve someone the best coffee at that moment in time. We may be able to trace a negative aspect of a coffee's taste back to the roast or a change in water, but I don't want to use that as an excuse for the customer I'm preparing the drink for. Of course, tweaking recipes and parameters cannot hide massive flaws in the system, and on these instances the coffee may have to be removed from the menu altogether. But a better understanding of what our suppliers may be aiming for and understanding the nature of their feedback loop as well as our own has meant that the variance in the recipes we use has narrowed. It also follows from this that we are naturally, through such processes, pushing towards more singular recipes.

But is there an event horizon for coffee brewing or are we making one? Of course, this event horizon is most likely to be defined by TDS and Extraction combined. After all, variables need to be set to judge anything, but it's also coffee's complexity that often makes a lot of our created judging systems flawed either way, whether we set the variables or change them.

Currently our recipes in store tend between 40 and 55 percent brew ratio mark (twice as much shot weight as dry dose weight) with TDS around 9-10 percent and extractions between 19 and 20 percent and on occasion up to 22 percent (these numbers are not with the EK43, but with a K30). Sometimes though this is more the coffee than anything else. At the moment a light roast pacamara manages a much higher extraction under familiar circumstances as slightly heavier roasts of very different origin coffee, which is unusual as lighter roasts tend to be harder to extract from. The question is though,could we go back and roast the pacamara coffee in a way that means it tastes better at 20 percent rather than 22? really we'd prefer to get the others tasting good at 22percent. Or are we struggling against the very nature of this unique product. Can we have a singular recipe for all coffee? Maybe here we need to have our  rules and allow for anomolies.

And this is where the constant cycle of dilemmas between the concepts of brewing theory and flavour in coffee keeps leading me, to the space between set recipe and variability.

Let's see what happens in the next year.

 
* TDS denotes the concentration of coffee in the coffee beverage, rather than the amount of coffee extracted into the liquid, for example a ristretto is highly concentrated but its hard to get a high proportion of the coffee into the liquid. Where as 1 gram of coffee to a litre of water would be an extremely weak coffee to water ratio, but it would be much easier for the water to extract a lot form the coffee as it doesn’t have much to deal with.

I find it useful to discuss it as workload. More coffee means a higher workload for the water. It has got more work to do to get the coffee into the water, and in some cases will simple not have the capacity (ristretto) and hence the extraction is likely to be lower even though the strength of the beverage is higher. less coffee results in a lower workload for the water and therefore it can achieve a higher output(extraction). We can help the water out with things like grind and time but there are limits on the work it can do.