Best Buy's Most Desired Customers?
September 17, 2006 7:21 AM   Subscribe

Please help me find information about the Best Buy "Centricity" market segmentation program -- specifically, the controversial new "customer profiles" used to identify Best Buy's most desirable demographics.

I'm doing some research on the profiling of customers by retail businesses, and the discrimination against ethnic minorities that may result from this profiling. From several recent articles I have read about Best Buy's "Centricity"program (for example, see these WaPo and RetailWire pieces), Best Buy has identified five core customer types that it most wants to attract to its stores. Best Buy teaches its managers and associates to recognize these customer types through the use of a representative name and profile for each type. "Barry" is a male with a six-figure income who purchases what he wants regardless of cost; "Ray" is a male who likes electronic gadgets but can't always afford what he wants; "Buzz" is a young male interested in gaming and Playstations who makes small purchases; "Jill" is Barry's wife, a stay-at-home soccer mom; and "BB4B" is a small business owner. Some, if not all, Best Buy stores are being revamped to appeal to whichever customer type is most dominent in that community.

I need copies of any of the materials Best Buy uses to explain these customer types to its associates. If I am not mistaken, all of these customer profiles are represented as affluent Caucasians. (These profiles are being cited, in a [warning: PDF] class action lawsuit, as evidence of Best Buy's white-male-centered culture.) I'd like to read Best Buy's descriptions of these customer types first-hand, and see its visual depictions of the customer profiles myself, not just through a Washington Post article or a lawsuit website.

Any ideas? There's nothing on the Best Buy corporate site, and I've Googled this question to no avail. Even sources that require payment would be welcome.

If you want to contact me directly, my email is
posted by jayder to Law & Government (22 answers total)
I'd try contacting the reporters at the WaPo and RetailWire and politely asking for copies of any BB materials they have. Hell, I'd also try asking BB directly; it doesn't seem like they're trying too hard to hide them, and I'm sure the class action lawsuit team already has them.

(btw, 'bugme' and 'not' work for a RetailWire login/password.)
posted by mediareport at 7:57 AM on September 17, 2006

The first thing you need to find out is if these profiles were derived from actual sales data. If the profiles are evidence based, the question of discrimination is moot.

what I mean is that when I go into Circle Grocery in New Orleans, I'm not looking for paté, and when I go into Whole foods, I'm not looking for turkey necks. The stores could care less the ethnicity of the people coming in, they simply stock what sells, because they can't afford to do anything else.

It wouldn't be surprising nor discriminatory to find that Big Box electronic retail stores cater to affluent white people, if and only if those people made up their primary customer base. Given that many are found in the 'burbs, this is probably the case.

So you may find a pattern in the demographics, but you've got absolutely no case for discrimination if they can prove that their profiles are derived from actual sales data.

Since I'm not a lawyer or particularly any kind of expert, this is all how I would expect it to work, and how it seems to work down here in New Orleans, But we're different down here in CC. Maybe the world is more fscked than I realize, and retailers can get in trouble for not trying to attract provably less profitable demographics.
posted by Mr. Gunn at 8:06 AM on September 17, 2006

I used to work at Best Buy when Centricity was rolling out (sorry, didn't save any paperwork on it). In relation to what Mr. Gunn said, I know that during the rollout there was a significant period of time when stores sales/performance data were being analyzed to determine what "type" (if any) each store was.

I can't remember ever seeing any actual physical representations of "Buzz" or "Jill", and it always seemed to me that our training videos were always careful to include images of associates and customers of all races/genders/body types/etc. (to the point where it was comical sometimes).

And, anecdotally, our store manager was a woman, two of the dept. supervisors were women (including the head of our Geek Squad) and two others were minority males (in an area without much diversity).

Best Buy had its problems, but discrimination was not one of them that I saw.
posted by Rock Steady at 8:29 AM on September 17, 2006

I was exposed to Best Buy's segmentation from a presentation we received as one of their vendor partners. I do remember that at least one of them were not white in the little one page summaries that had pictures. I can't remember which.

The racial background of the segments was not really highlighted or even mentioned in text. It was really about attitudinal dimensions and income. A "Barry" store may have higher end TVs in display because more people in that area have the household income to afford the TV, and that store has higher ASPs on its TVs than the other Best Buys in the area. The best example is with high-end stereos. Serious audiophiles don't go to Best Buy to get their equipment. Best Buy wants to identify pocked of stores where it can offer the higher end stuff and work with local installers for home theater installs. That isn't something Best Buy would roll out to all of its stores.

There was nothing in the Best Buy materials that suggested that ethnicity mattered in trying to figure out whether a customer was a "Buzz" or a "Barry".

One of the reasons Best Buy has been so successful is it measures everything it can and does a lot of research on its customers and prospects. Where, I live shelf tags are in English and Spanish. I'm sure that isn't the case in neighborhoods that have a smaller Latino population. Best Buy has avoided getting crushed by Wal-mart with this thinking and has helped it out-sell Circuit City and CompUSA.

Before doing the gig I do now, I worked for a company that employed a similar segmentation scheme. We didn't ask -- nor did we use in purchased lists for mailing -- for any racial information. We didn't train sales reps to look for a particular type of person by race. The training materials we had with example questions had to do with how the customer was going to use the product, who in the family was going to use it, etc. Race didn't matter in CE purchases. Gadget freaks come from every ethnicity.
posted by birdherder at 9:06 AM on September 17, 2006

Thanks for all these great answers. Let me clarify my thinking a little.

My thinking is not so much that Best Buy overtly encourages the marginalization of minorities, but that as a result of these profiles (if they are, in fact, white), associates have been trained to give better treatment to the more affluent. In some communities, race is linked in the minds of many with being poor (whether justly or not), and so may be interpreted by associates to mean, "Oh, this guy isn't part of our target demographic," and result in poor service because there's an assumption that they are less desirable.

Under public accommodations/human rights laws enacted in many states, retail establishments are not allowed to give lesser service based on the ethnicity of a customer. If it can be shown that race is taken as shorthand for "not our target demographic," I think discrimination may be the result of these segmentation programs.

Thanks, everyone, and please keep the information coming.
posted by jayder at 10:04 AM on September 17, 2006

One key thing to consider with both CC and BB is that they move a lot of their stuff on razor margins or take losses (after expenses). They make their profitability numbers using two main services: extended warranties and financing for big ticket purchases. One way to think of them is less as retailers and more as credit providers for the FICO-challenged.
posted by meehawl at 10:19 AM on September 17, 2006 [1 favorite]

I agree with Rock Steady - In any materials that were distributed to blue shirts, you would never see such visible evidence of discriminatory thinking.
I remember an online training module teaching how to sell PSPs (extended warranties) and when to back off. Two examples of customers who were buying products but didn't want the warranty were, no kidding, the Queen of England and a female astronaut.
Now as to whether there is discrimination at the store level, of course there is. As a female, I was hired and offered a position in appliances, because "women like to buy this sort of stuff from other women." I ended up in the computer department, as the first female employee they'd had in that department in the history of the store. This wasn't some brand-new store, either, it was the direct descendant of the Sound of Music store destroyed by a tornado, which became the FIRST BBY store.
Email me if you have questions.
posted by Coffeemate at 10:47 AM on September 17, 2006

Per your further comments:

Centricity was never intended to help associates identify (segment) customers. It was intended to change the way individual stores were merchandised/staffed/stocked. The point wasn't "Is this customer a Barry?" but "Is our store a Barry?"

We were often told by management not to worry about which type of store we were, as we had no control over it. All we could do was get our key metrics up to the levels that were required to undergo "Centricity".

Obviously I can't say that it never trickled down to the individual employees/customers in the way you are thinking, but you certainly won't find anything like that in official training materials.
posted by Rock Steady at 11:14 AM on September 17, 2006

Clearly the other people in this thread have more immediately pertinent information than I do, I just want to be sure you're distinguishing cause from effect.

If preferential treatment is caused by underlying racial bias, then that's bad, but if preferential treatment is caused by knowledge of established factual demographic buying patterns, then that's good, even if it creates a sales strategy that looks like ethnic bias to an outside observer. A retailer must stick to the data, and make the business decisions the data indicate, because they're making a extremely thin margin on most things. BTW, AV cables are a notable exception to this, where they make a ridiculous markup.

Before you can make ANY kind of statement about your premise, you have to show that the inferior treatment is in spite of the business intelligence, and not simply a result of making a race-agnostic business decision. I wish you well in your study, but as in any study, you have to be careful that you haven't decided what you want your results to be before the data have been collected. If you don't think there could be any legitimate reason that certain customers get preferential treatment over others, and therefore don't even control for the possibility, then you're doomed to a biased study before you even start.
posted by Mr. Gunn at 11:53 AM on September 17, 2006

For further research, the key concepts you want to read about are causation and correlation, specifically how they differ from each other.
posted by Mr. Gunn at 11:56 AM on September 17, 2006

Thanks, Mr. Gunn, but I already have a sound grasp of social science methodologies.

You may want to re-read this paragraph from my clarifying comment to see that I am not mistaking correlation for causation:

If it can be shown that race is taken as shorthand for "not our target demographic," I think discrimination may be the result of these segmentation programs.
posted by jayder at 12:01 PM on September 17, 2006

So, what visual cues _would_ a Best Buy sales associate use to determine the desirability of a customer? Now I'm curious. I don't think it would have to be race, but is there a disguise I can use (in addition to looking male or motherly)?
posted by amtho at 12:19 PM on September 17, 2006

I mean of course if such visual cues are included in any training materials. Are they?

Sorry if this seems a derail; I started my first comment a while ago.
posted by amtho at 12:23 PM on September 17, 2006

Upon reflection, I think Mr. Gunn and I have different understandings of "race discrimination." I'll spell out how I think our understandings differ.

Under Mr. Gunn's definition, a behavior is not race discrimination if you have a rational and articulable reason, other than inherent dislike for a certain race, for preferring not to deal with that race (i.e., if your economic studies show you that African-Americans don't spend as much as Caucasians). You could treat Caucasian customers better than African-Americans because you think they will spend more in your store, and this couldn't be characterized as race discrimination, just rational economic behavior.

Under my definition of race discrimination, anytime race is used to distinguish desirable customers from not-so-desirable customers, with the more desirable customers receiving a higher level of service from an establishment, this constitutes race discrimination --- even if you have a rational and articulable economic reason for doing so. So, if a company trains its employees that "Caucasians are the affluent, big spenders we want, and African-Americans aren't," and employees as a result routinely give Caucasians better treatment than African-Americans, this is race discrimination.

I'm pretty sure that federal and state civil rights laws resoundingly support my definition of race discrimination.

My theory --- it's just a theory at this point --- is that Best Buy's Centricity program may result in race discrimination, by subtly inculcating a view of minorities as less desirable customers and resulting in poorer service.
posted by jayder at 12:45 PM on September 17, 2006

I have no strong opinion here either way, but I'll echo Mr. Gunn's comment:

" have to be careful that you haven't decided what you want your results to be before the data have been collected."

It sounds to me like you've already made up your mind, and you're looking for data to support your theory.

It may be true and it may not, but from your framing of the question and your subsequent statements, you sound like you're already convinced of 'the truth', and anything that vaguely supports that view in any way will be taken as evidence in favor. So I would tread most carefully; this is the kind of idea that can result in a lot of damage to the company if it gets traction, even if it's not true.

I don't like Best Buy at all, and I avoid it whenever possible. That said, it strikes me that your study could easily turn into a hatchet job. Given how sensitive this area is, please make SURE you're right before drawing any public conclusions.
posted by Malor at 1:05 PM on September 17, 2006

I already have a sound grasp of social science methodologies

If you did, you would understand the cardinal rule of statistical testing methodology is to decide to accept your hypothesis after you collect data, not before.
posted by Blazecock Pileon at 1:13 PM on September 17, 2006

Even if you get the data you're after and it shows exactly what you hope it does, it won't be very good evidence for your theory.

To support your theory that Centricity causes race discrimination, you'd have to show not only that Centricity has racial components but also that racially discriminatory treatment at Best Buy got worse after Centricity was introduced. Not that it existed afterwards, but that afterwards it was worse than it was before.

If you want to think of it that way, a racial component to Centricity is necessary but not sufficient for your theory.

I assume you've contacted the law firm involved with the class-action suit about getting the relevant information?
posted by ROU_Xenophobe at 1:36 PM on September 17, 2006

Fortune magazine had an article about Best Buy's new approach recently.

It really didn't come across as sinister or racist to me.

Isn't it racist of you, jaydar, to assume that any depiction of a middle or upper glass person must be of a white person?
posted by croutonsupafreak at 2:26 PM on September 17, 2006

Isn't it racist of you, jaydar, to assume that any depiction of a middle or upper glass person must be of a white person?


You must not have read my question very closely: I said, "If I am not mistaken, all of these customer profiles are represented as affluent Caucasians." Does that sound like the statement of someone who is just assuming something without basis?

Read the PDF press release issued by the law firm; it says all the profiles are white.
posted by jayder at 2:48 PM on September 17, 2006

If I am not mistaken, all of these customer profiles are represented as affluent Caucasians.

You are mistaken.

The whole point of this restructuring was to fight against marketing by demographic (race, affluence, etc) and begin structuring stores around psychographics. Why? Because Best Buy seemed to realize that segmenting by race/income/education was missing a critical element in deeply appealing to people. In fact, these segments seem to come up because demographics were very poorly related to what people were looking for in their Best Buy.

In all my time working with Best Buy, I never saw images attached to the stories, nor did the profiles ever seem to be final.

A lot of stores took intiative to try to round out the profiles, get employees to understand them, etc. I can imagine that images or similar might have been attached to them, but I'd be surprised if they came from Best Buy corporate.
posted by Gucky at 4:05 PM on September 17, 2006

I'm not familiar with the details of Best Buy's profiles, but based on my experience developing and using these profiles in my work as a product designer, Gucky is exactly right -- if Best Buy included race in their profiles, they were doing them wrong. The whole point of these types of profiles is to get at psychological and behavioral characteristics that don't map to demographic segments.
posted by jjg at 5:56 AM on September 18, 2006

I work for a large retailer, and we also use a similar type of customer segmentation. Pretty much all of the sample photos are of white people, plus a young asian guy. Why? Because the majority of our customers are white. Why? Because we are an old company that many people shop out of childhood nostalgia; people who have immigrated in the last generation or two don't have the same loyalty to our store, and therefore don't shop there as much. (Most immigration to Canada prior to a 2 generations ago was of the non-visible-minority variety.)

The segmentation isn't used to train store associates - they probably don't even know that the segmentation exists. The information is used to figure out what types of products to sell, how to merchandise them effectively based on shopping habits, which stores to stock them in, etc.

In short, Best Buy probably also isn't using their segmentation info to train associates to racially profile their customers.
posted by Kololo at 8:44 AM on March 14, 2007

« Older iTrouble with iTunes   |   When does sex stop being painful (and start being... Newer »
This thread is closed to new comments.