Starbucks: Numbers do lie. Sometimes pathologically so… | Hacker News Reader

Starbucks: Numbers do lie. Sometimes pathologically so…

gapingvoid.com

42 pointscwan22 comments

If by "Numbers do lie" the author means "Numbers don't lie, and also some models of human behavior are less useful than others", then this essay is adequately titled.

Or alternately, "Don't let accountants determine your policies."

Accountants almost always make bad decisions. Because they measure only the money. Its takes real leadership in the company to consider the import of each choice. Add another column to the ledger, so to speak, such as Quality or Customer Satisfaction.

The problem with the soundbite summary is that the analysis presented is incomplete. On the one side, 10% of customers might have noticed the drop in quality, but we have no idea how many of them would actually go somewhere else to get their coffee. On the other, while 100% would notice the higher price, we again don't know what the demand curve looks like so we don't know how many sales they would lose.

a) Story has nothing to do with its conclusion.

b) So when did Starbucks start using the awful coffee it uses nowadays? Or do they merely buy good beans and roast them badly?

Yeah, I don't see what it has to do with numbers--the explanation to the customers was the only thing that made it work; while only 10% of them may have been able to tell the difference, 100% of them would like to think they'd be able to tell the difference, and would gladly pay more to signal that ability.

I'm no coffee expert but I've heard it said that Starbucks' coffee is not so much bad as it is different and an acquired taste. Of course, these could be the words of apologists, but if so, is Starbucks entirely living on the dollars of coffee dilettantes?

I hate to start a coffee snob thread, but Starbucks right now lives off the dollars of the 90% of people who couldn't tell the difference between Starbucks and better coffee.

I get the impression that Starbucks in its early days was a coffee snob place, but at some point it transitioned into a mass-market coffee place. It would be fascinating to be able to try a coffee from the original Starbucks in the 1970s to find out whether it has got worse or whether the tastes of the US coffee snob have got more sophisticated.

I think that if you pour two cups of Starbucks Chateau du Latte Liberal 2010 vintage and tell them cup A is "regular Starbucks" and cup B is "Starbucks from the 70s -- quite exclusive, really, we're lucky we could find this, they don't make it any more since the plebians can't tell the difference" you'll find that 9 out of 10 coffee snobs agree that cup B is markedly better.

I don't think Starbucks is unaware of this.

It could be that you are coming to your coffee with preconceived notions as to its quality. I wonder what a blind tasting between Starbucks and your favorite coffee would tell us.

Hard to say, as everyone that knows how to taste coffee already knows what Starbucks tastes like. It has a distinct flavor.

Given a shot of espresso from each of the coffee places I visit regularly, I could tell you which is which just from the aroma. So there is never going to be a "blind" test.

I believe that assertion is exactly what the previous commenter proposes to test.

Starbucks is not about coffee, it's about convenience. Starbucks is right inside my building. Do I really want to walk for 20 minutes to get a better cup of coffee at Intelligentsia? Not often.

Starbucks is not great, but they are consistent and not awful. When I'm traveling in most of the US, coffee from an unknown shop is likely to be awful. I'll grab a latte from Starbucks just because I know what I'm getting and that it'll be drinkable, if not up to my usual Portland coffee-snob standards.

My theory is that they intentionally burn the coffee during roasting to make every cup consistent. Also, relatively speaking, the various espresso drinks are much better than the plain coffee. The sugar and cream do wonders to cover up the over-roasting.

I like StarBucks coffee. What do you find wrong with it? They do use stronger blends than many other coffee shops which is exactly why I like it.

Are you comparing StarBucks to other big chains, or a local place that serves coffee roasted in the past 3-10 days?

I can take Starbucks beans and make a decent cup of coffee at home (decent, not exceptional). I've never liked the brewed coffee in their stores -- they do a horrible job of brewing it. Of course, most "big tub o' coffee" brewing methods produce at best a mediocre brew.

Before their espresso bean quality really took a dive, they started introducing the first generation of their automated espresso machines. The same beans that, depending upon barista, produced a decent, sometimes very good, shot from the hand-operated espresso bars, produced an espresso approximating warm, flat cola (really) from the automated bars. I knew one store manager of 10 plus years who would dive in and tweak the automated bar until in produced espresso that was a more palatable "shadow" of a good hand-pulled shot, but his motivation was personal interest and he could not keep the machine producing at that quality -- it was a constant struggle of adjustments that could not be maintained in a commercial environment.

I've heard it suggested that they may have started "burning" the beans more in an attempt to compensate for the weakness of the espresso that the automated bars pulled. Also, my impression is that corporate attention shifted to identifying and marketing more categories of merchandise.

I hardly ever drink a Startbucks, any more, but I've heard that the quality has improved somewhat since Schultz dove back into operations and told people essentially to fix their quality problems. Apparently, the beans are not being burned as badly. They've also been rolling out a second generation of automated bar (as well as some different automated bars for locations having different volume requirements) that produces a better shot than the first generation.

Their in-store brewed coffee still sucks. Their overpriced instant coffee tastes better -- as I've experienced when they first launched it and were offering side-by-side taste test comparisons.

P.S. Their best "take home" beans/roasts, in my opinion: Sumatra, and the holiday blends (particularly end-of-year, although Thanksgiving is, I've been told, a fairly similar blend), depending on the year.

Recentness of roasting is key for coffee not tasting like gasoline.

Basically, I've heard that Starbucks does way over-roast their beans, but the main problem with them, and with most coffee places, is that they don't use freshly roasted beans. Beans go rancid within 7-10 days of roasting, and after that almost invariably produce that overpowering gasoline flavor.

Using freshly roasted beans is terrible for inventory and horribly inconvenient (transportation, etc), and people don't understand that straight coffee isn't supposed to taste like gasoline, so they do fine.

Use http://espressomap.com to find places with coffee that actually tastes good (as opposed to milk and sugar that tastes good).

(On top of this, brewing espresso is really tricky, and requires constant readjustment of grind, brew time, brew temperature, brew pressure, and brew volume (these are coupled, so not quite that many degrees of freedom) based on flavor, texture, and volume of the resulting shot. Baristas at good espresso places pull shots and taste them regularly throughout the day to recalibrate.)

I think part of the success of this option is how Star Bucks wants to represent itself to its customer base. Had Dunkin Donuts chosen to raise prices in face of an almost unnoticeable drop in quality I'm sure their customers would have been less receptive.

I don't know about that. For a while, at least, people raved about DD's coffee.

People in Boston still do. I don't really understand why.

Reply on news.ycombinator.com