3 ms·
I'm reminded of the (perhaps apocryphal) story of how starbucks overroasts its coffee beans to achieve a consistent taste despite having multiple sources for be
by leetcrew 5y ago
I'm reminded of the (perhaps apocryphal) story of how starbucks overroasts its coffee beans to achieve a consistent taste despite having multiple sources for beans (necessary due to its scale). the implication being that they sacrifice the average quality of the finished product to reduce variance.
is there a similar tradeoff in brewing? I do notice fairly obvious differences batch-to-batch in microbrews, but I would still prefer the worst batches of my favorite local beers to any AB product I can think of. are they trading away best/average taste for consistency, or is it more a function of the price point (microbrews being quite a bit more expensive)?
- throwawayboise 5y agoYeah that's true, the Starbucks "house" beans they use in their stores are nearly soot, and the coffee tastes like it.
- tristor 5y agoFrom what I've seen, the brewers at AB are world-class, and they have a very specific and longstanding flavor profile they target. I happen to also not be a fan of Budweiser, but it's been the same for decades and has a very loyal following. They are not brewing to a price point, they spend more on process and verification than any microbrewery to ensure that consistency, although they have introduced the use of adjunct brewing (rice) to make the materials cost fit their price point, and they were able to do it without changing the resulting product taste.
- ceejayoz 5y ago> I'm reminded of the (perhaps apocryphal) story of how starbucks overroasts its coffee beans to achieve a consistent taste despite having multiple sources for beans (necessary due to its scale). the implication being that they sacrifice the average quality of the finished product to reduce variance. The version I'd heard was people associate the ashy taste and dark roast with higher caffeine, even though it's the opposite.