The Long-Suffering Wife vs. The Likert Scale

We’ve had the Volvo in the shop for a couple of days (nothing serious) and this always causes a debate in the Willett household.

Through circumstances beyond our control, the Volvo service department appears to be in abject terror of our car being there. This is because (not our fault!) they’re apparently required by their management to send us a customer satisfaction survey. The Long-Suffering Wife’s email address is the one in their system, so she gets it, fills it out, and returns it.

Now, let me state that as far as I can remember, the service we’ve gotten from them has always been great. I don’t know that we’ve ever been really unhappy, or even a bit peeved, with their service when the car’s in there.

But we have a serious “apples & oranges” problem.

On these customer satisfaction surveys (online or otherwise) they use a Likert scale. You may not know the term, but you know what I’m talking about. “What was your experience regarding the politeness of our staff, on a scale of one to five, five being best?” (Sometimes it’s “Strongly Disagree / Disagree / Neutral / Agree / Strongly Agree”, but it’s the same thing.)

The Long-Suffering Wife uses Likert scales at work and she has problems getting accurate data. A major problem with Likert scales is that there are several possible sources of bias. Since she has problems dealing with this bias, she tries very hard to give an accurate, unbiased response when asked to fill one out herself. Therefore, when we get satisfactory to good service, she’ll give them 3’s and 4’s. (“Apples”)

At the other end, Volvo’s management apparently (like many, many, many companies) demand that every customer get service that returns nothing but 5’s. This is ludicrous and makes the data they’re getting useless, which in my opinion makes them really useless and lousy managers, but as I said, we see this a lot. The reality is that some customers get truly outstanding service and some get truly horrendous service and most probably get decent or reasonable service. (Can you say, “Bell Curve?” Sure, I knew you could!) Defining your company’s expectations as “nothing but 5’s or we’re a failure!” is stupid. But that’s their management, not their service people. (“Oranges”)

In short, the Long-Suffering Wife gives them honest answers to the questions being asked, even though the expectations on the other end are skewed.

Then the customer service staff get yelled at and (apparently) lose bonuses because they didn’t get all 5’s. Despite the fact that it’s stupid and meaningless to expect all 5’s. Then when we bring our car in they absolutely beg us to give them all 5’s and they bust their butts to do anything for us. And The Long-Suffering Wife sees it as average or above-average service (which it is, it’s not like they’re picking the car up at home and dropping off a loaner, which they’re not supposed to to begin with…) and the cycle starts all over.

I understand the service staff’s frustration. I understand that they’re working in a broken system. I understand and agree with the Long-Suffering Wife’s objection to giving inaccurate answers.

I also understand that it’s not my float! 

I’m a pragmatist. While it looks like they’ve got a broken system, it’s not my system. While I think they’re getting useless results, they’re not my results. I think we’re getting perfectly good service and I don’t think the staff should be penalized for doing a “bad job” when they’re not. Since it’s no skin off of my nose, when the staff explains the situation and begs me to give them 5’s and I’m perfectly happy with the service received, fine, I’ll give them 5’s.

So now it’s a philosophical debate in the household. Should we give honest answers to the questions asked, or should we answer the real question and help out the perfectly competent staff?

The bigger question (which I don’t have an answer to, but would love to hear your feedback) is what to do when and if it IS my float at some point. If senior management needs accurate data, but managers and staff in the field are changing the rules and introducing bias into the results, how do you get rid of the bias?

Our solution to the Volvo customer satisfaction survey issue? The Long-Suffering Wife will now send me the link to the online survey and I’ll fill it out for us. That way we both get to hold our moral high ground and the Volvo service staff get their bonuses (based on useless data).