Something I learned super late was the difference the bean makes in making coffee.
I had heard and tried some single-origin from the supermarket, but it was really only after I ordered my first 250g bag at one of the local specialty roasters that I noticed the difference.
Be careful. It will ruin your taste. Once you’ve tasted the difference, you’ll have a hard time coming back.
If you think about it, it’s obvious. The good stuff is produced in perfect locations with ideal climates, then processed carefully and roasted quite differently from most supermarket coffee.
I’d say the gear and recipe when brewing coffee play a role, but good coffee beans are much more important.
Problem is good coffee is expensive.
To not break the bank, I started playing around with data that one can get from coffee roasters.
Most shops these days use Shopify and publish the product catalogue in JSON format, making it easily readable by a script.
I plugged in my favourite roaster and collected all the different beans they offer to quickly compare prices, see where a particular coffee lies compared to others, or simply convert from different package sizes to a uniformly comparable metric of $/100g.
Additionally, I wanted to know when prices or products changed.
The little script turned into an app that collects, enriches, and filters coffees from more than 200 roasters, totalling more than 3000 coffees.
I probably went a little overboard here, collecting all that. Right now, I only use it for New Zealand to compare prices of different coffees.
Here is how I buy beans
I learned from Reddit that freezing beans works quite well, and I tried it out and can confirm it works.
That unlocks some good savings potential. Buy 1kg, dice it up into 250g ziplock bags and freeze it.
With the data I collected, the comparison is easy.
Shopify gives me the price per 250g, 500g and 1kg for most beans, so with a little math, one can unlock bulk shopping discounts.
If you enjoy expensive beans and drink a couple of cups a day, that’s pretty decent.
Alternatively, a search through all roasters and their coffees can reveal some interesting variety.
Without collecting all this data, I wouldn’t have been able to filter by coffees that have “peach”-tasting notes, filter by “filter”-roasts or sort by price.
Another aspect one can look at is a regional price comparison of coffees in a country.
Say I want to know how expensive this particular Guatemalan coffee is compared to other Guatemalan coffees from a New Zealand roaster.
Price is not everything, so I make sure to taste a new bag first in a smaller quantity rather than buying bulk straight away.
More cool stuff to do with coffee data
I’m sure there is much more potential for analytics here, but here are a few things I found interesting in crunching through the data I collected.
There are decent price differences between coffees from different regions. That doesn’t mean they necessarily taste better.
Why some regions are cheaper than others, I don’t know. It could have to do with local labour costs, the varieties that are grown, and the available data here.
When I collect this data, I run a job that gets the Shopify catalogue data and then updates the price. I keep that update to compare price changes.
Now this is all just a little side project, but I love the idea of having all this data available and frequently updating it to gain insights, even if that’s just me buying better coffee and brewing tastier cups.
Ultimately, this might be my excuse for finally buying that expensive bag of beans I held myself back from.
It’s not the $40 I’m worried about; it’s more so that I’m worried it will ruin my taste and raise the bar for what I consider good coffee.
You can try out this little app here:
https://coffee.learnedlate.com/
PS.: I’ve recently been enjoying this V60 recipe. It’s not so much a recipe but more a guide on how react to parameters when your coffee is brewing
I highly recommend it as i’ve been upping my coffee game thanks to it.