The Statistics That Count

I’m sick of companies that send me reminders and requests to rate them or the products/services that I’ve purchased. I’m also more than a little tired of polls that guesstimate how the public feels about this or that issue or politician, almost on a day-by-day basis, and particularly the pseudo-polls sent by both parties that misrepresent the issues in order to beg for contributions. Not to mention companies that legally misrepresent prices and interest rates… and, of course, politicians who declare that the election was stolen when recounts, investigations, and audits show that it wasn’t.

I purchased the product or service. Whether I purchase more is what really counts for the company. What I especially hate are those questions asking how the company could improve its products and services. If a manufacturer or service provider doesn’t already know its shortcomings, the odds are that they won’t take my suggestions anyway. And why should I provide free market research that they’ll ignore? The bottom line isn’t what I think, but whether the product/service is good and the company sells enough to remain in business.

While good polls can reveal what those polled feel at the moment, what such polls don’t reveal is, in most cases, more important than what they do. Today, everyone is most concerned about inflation. It takes a poll to verify that? The more important question is why they blame the current administration, when it has only a minor part in creating the inflation. This isn’t an apology for Biden; it’s been a problem for decades, if not longer. The factors that influence the economy have long lead times, and whoever’s in office now gets the credit or blame for the acts of his predecessor. The polls just focus the blame/credit on the wrong person… and most of the public is either too stupid to understand or doesn’t care, because they want someone to blame.

And far too many companies misrepresent prices, like the replacement window company that offers your second window at forty percent off, provided that you buy four, which means that you get the first four windows at ten percent off, but the number that sticks in most people’s mind is forty percent. Or the car dealers or others who advertise no payments for the first year, but don’t mention that the payments after that include interest on the entire amount for that first year.

As for the Republicans who insist the election was stolen, the bottom line is the final authenticated vote count, and, interestingly enough, the bottom line in Kansas on abortion was that sixty percent of the voters voting [which was a record turnout for a primary election] didn’t want abortion banned, no matter what the right-to-lifers claim.

6 thoughts on “The Statistics That Count”

  1. Frank Hamsher says:

    I have a similar reaction and response to such surveys and polls. However, I think there is a deeper issue here. It ties in with your posts about how many in modern society are content with sound bites and easy answers that do not reflect real solutions to real problems. The companies generating these auto-polls are part of the more significant problem.

    When I first moved to Australia in the mid-80s, we would hear, “Oh, they chucked it in the too hard basket.”

    Now, unless you care enough to do the work, everything goes in the too-hard basket.

    Incidentally, I just received my Kindle copy of Councilor this morning. I’m tempted to opt for full radio silence mode until I finish it.

  2. R. Hamilton says:

    A stream of information of questionable value (i.e. could one really get statistically valid and useful information from the usual product review questionnaires?) is something some managers use to claim success or justify what they want to do anyway. I wouldn’t completely rule out that the information might get used more constructively, but I suspect that’s rare. That’s a problem in other fields too; IT tends to love metrics, but not necessarily make good use of them.

    Take one large hardware store as an example: they ask for a followup survey on every purchase (given that I get a nice discount from them, they know I made the purchase). Mostly I like them, but my usual problem with them is that their location information (what aisle and bay has a product) is frequently wrong (or not promptly updated when items are rearranged). Given the size of the stores, that’s a nuisance, esp. since few of the workers know where more obscure items actually are; hiking through the entire store to check the multiple locations some item MIGHT be gets more difficult as the years go by. Over a couple of years or more, I’ve pointed that out repeatedly as their biggest fail in review responses, but have yet to see any evidence of improvement.

    Contrast that with a northeastern regional chain of very large upscale grocery stores, which is MUCH better about keeping product location information up to date, and where their app will arrange your shopping list in such a way as to make an efficient pass through the store, and not send you back and forth. Oh, and at all levels, they at least try to answer even unusual questions (like where they get their music playlist – I find myself using a song recognition app on a couple of songs to add to my library every time I go there). So doing a good job of showing accurate product locations _is_ possible. I think a lot of it comes down to employee competence and diligence (and training?) at least as much as the systems used; the grocery store’s employees all seem happy to be there and almost all seem knowledgable; while at least half the big hardware store’s employees, while not as zombie-like as those of a busy Walmart near a big city, seem rather indifferent at best.

    (the only beef, if beef it is, I have with the above grocery store is that every location’s design is different, and so the layouts and some optional goodies (prepared food/in-store restaurants) vary by store, meaning their app’s ability to locate items is just about a necessity, compared to say a much smaller store (Sprouts) where the layout is relatively consistent and you can see practically everything from some positions anyway.)

  3. Postagoras says:

    The annoying part is that the data usually fuels a completely bogus management function. Companies evaluate their performance and often peg bonuses for senior management to the measurements from these polls. That’s without trying to frame the sample within any real statistical analysis.

    Or even dumber, when the cable company that has a monopoly asks if you would recommend them to a friend. As if you had a choice.

    These data are the Velveeta of management, processed until palatable . See Goodhart’s Law.

  4. Monica says:

    I have worked in retail. A lot of these surveys are used to rate a company’s sales team.
    It’s a management tool used to incite performance. If you get a bad survey, the manager talks to you about it. Upper management also has benchmarks of how many good surveys a store or team should get in a month and they use it as leverage over the store staff.
    I have always thought that using these surveys is a ridiculous method to go by, but I have seen various companies use them.

  5. JakeB says:

    I have found that I’m willing to answer these surveys as long as they don’t take more than a minute or so to answer. One, maybe two, pages with 2 or 3 multiple-choice answers each, sure. More than that, my response is F. U. and I close the page (or browser, if necessary).

    I think of it in terms of the value of my own time (which I value highly). I hadn’t gone as far as thinking about whether my responses will even be analyzed until I read your observations here. But given your points, I think I will be inclined to be even less tolerant of marketers asking for my time.

    Two years ago, I had been thinking about how much I was enjoying being able to commute at 80-85 mph. Others were going even faster — sometimes, even at 80, I was passed like I was standing still. I figured that here in the Bay Area, people were trying to get back all the hours they had lost to commuting in the past. Really, even driving at the speed of light would not be too fast. I’m now thinking about all the times my family answered the phone back in the day before answering machines much less cell phones because those dickbag marketers knew that if they called at 6-7 people would almost always be home, and would answer the phone, and thinking I should be bothering even less to answer any kind of requested response from marketing groups of any stripe. To get back my own, to whatever degree I can.

  6. Michael Creek says:

    Asking current customers is not really that useful. Asking long term customers that no longer use the companies’ services why might actually result in important information. One recent example might be the F/SF magazine from a recent blog entry.

Leave a Reply

Your email address will not be published.