As you do with coffee, I’ve experienced steep learning curve for the past 6 months, so steep I’m still trying to digest, and have a lot more to digest to fully grasp, apply and understand, so this post is just the tip of the ice berg of what I’ve learnt and can share right now, and I’m still in the process of learning.
I want to talk about Quality control. What is quality control? When you receive a business card and that person is in charge of QC, or you see it on a twitter handle, what does that person do? How do you control quality? I’m not pointing my fingers to anyone, because I carry out QC too, and probably, everyone will have a different methods to carry out QC, and different agenda and reasoning behind what they do. The question of ‘what is quality and how do we actually control it’ has been hurting my head for some time, if you ask 10 person what quality is, you’ll get 10 different answers, but we’re all trying to achieve the same thing though, which is consistency in quality.
Before we talk about the QC programmes that are carried out in cafe’s or roasters, we need to talk about quality in origin, because that’s where the quality starts. I was in Costa Rica few months back, and I was at a exporters office cupping unto 75 coffees in short amount of time, and by using SCAA cupping score sheet, I awarded points and evaluated the samples by tasting them. What I saw myself doing was I was only using the points scale of 7-9, out of 6-10 that was available, and the difference between scores were only like 1 or 2, the difference between lowest and highest was 6.
Coffee that scored 84.5 and 85.5 doesn’t seem that different, but it does on price, and when the mass of green adds up, we’re talking about considerable amount of money. It’s so easy giving 1 more point on SCAA cupping scoresheet, I do partially point the finger on the scoresheet and to us too. We are not a reliable taster, fact. The only few reason why we are okay with tasting is because we give fast response, and we can give qualitative and quantitative information fairly quickly, and we are quite easy to train, and learn how to taste, we pick up things quickly. Our sensory experience can be manipulated so easily, this is something everyone probably knows, but when you research deeper into this, you realise the scores that you give, the descriptors that you give, might actually not be true, and that human to human calibration is actually impossible. Variability and bias can be associated with the cuppers themselves, or the methodology, the coffee it self and of course, the environment. We can never be objective when we are tasting and when we you are tasting your 50th sample in few hours of time frame, you’re probably not scoring objectively because you’re palate is so fatigued. I’m not being sinister and saying we are useless, but we can be better tasters overall, if we take more precautions when we are cupping, even just palate cleansing and eating flavourless crackers help you so much.
Okay, its something I understand at origin, you have 100’s of sample to cup, and score, and that scoresheet has been ‘designed’ to grade the quality of the green. Then why am I seeing cafe’s and roasters using SCAA or COE cupping scoresheet or other scoresheets that has been designed for something specific? Why use them for our QC process? Yes, everyone wants delicious, but every roaster and baristas have different definition on what delicious is, and the approach to achieve that. Then why are we using a meat cleaver as a butter knife? You’d probably get the job done, but you are being super inefficient at it. If you have a specific objective, you need to have the necessary tools that suits what you are trying to achieve. Having a scoresheet or a feedback form that actually fulfils what you want, and what you are looking for, is something we need to think more about, when you really think about your own ‘philosophy’ of roasting and brewing again, you’ll find the scoresheet now you use pretty useless.
In my opinion, you can never control quality, we are the slave to quality until we actually know what we want to achieve, the term quality is so broad, and can be translated into so many meanings, everybody will not agree with you. In addition, we need further training on tasting, specifically, tasting objectively. There needs to be vigorous screening and selection tests to see where your evaluators are sitting at. Slowly, I’m losing my faith in cupping, call me stupid, or what ever, but cupping is a very flawed method to gather proper sensory data. I’m going to carry out few test to prove this, and there will be more on this, but I’m just throwing it out there.
The hardest part, where we find what we want to do, and trying to build something specific for the objective you set. We really need to think twice on the data that we collect, because data is good, but is that data actually useful? And actually, using statistic properly to analyse the data that you obtain, either from roasting or scores from the cupping table to fully interpret and magnify the result you get, and this is the exciting shit right here, when you start to obtain datas that are statistically tested, you can prove something. You reduce the guessing and trial and error. You can be a good taster, and be confident about a theory you might have, but your palate never beats figures. Everything can be better than what you’ve done yesterday, and to do that, we need the right data, the specific tool to measure what you need to change. We carry out ‘quality control’ for feed back, so we can roast and brew better and the constant procedure of control and feedback needs to be in loop, and synced. In the long term, we need to able to give useful feedback to the farmers either from the exporter or the roaster, after a year of hard work, I think they would want to hear more than ‘you’re coffee is tasty’, rethinking how we interpret and analyse the data we collect, and using the right tools to do so will I believe have an effect from the farm to the bar.
Valid data -> valuable feed back -> increase in quality.
Even though I’m studying sensory science, rather than obtaining answers, it has given me more questions I can’t answer, but I think we should really be thinking about this as an industry. We’ll never get the answers unless we tackle it, and we really need to push ourselves to raise the quality.
Oh, and I really don’t like the term Quality control any more. What’s quality again?