Sunday 25 September 2011

Thoughts on Customer Satisfaction Surveys

"Close to the Customer" has become a mantra of many companies ever since Tom Peters and Bob Waterman introduced the phrase in their book, In Search of Excellence. And often, our attempts at being close to the customer involve taking their temperature on a regular basis via customer satisfaction surveys.

That's a good thing. However, at some point, some executive takes it one step further and thinks: "We'll hold our (employees, partners) accountable for customer satisfaction by basing part of their (compensation, promotion, renewal) on the results on these surveys."

I think that's the wrong move, for several reasons.

It skews the results, and in the totally wrong direction. People don't like to get bad news anyway, but if their salary or bonus or other compensation depends on how little bad news they get, they'll do everything that can to avoid negative feedback.  "Good," you might say, "They'll please our customers so they won't get negative feedback." No, they'll game the system.  For example:
  • They'll hint or even downright tell customers how to respond.  "If my service isn't a perfect 10, let me know so I can fix it."
  • They'll cherry pick, only sending surveys to customers whose projects went well and "forgetting" to send the survey when they went bad.
  • They'll avoid risky engagements, such as new products and sensitive customers, delegating them to less experienced colleagues. So paradoxically, the most experienced people withdraw from the places they're most needed.
  • They'll avoid being heroes and parachuting into existing projects that are not going well, because their name will be associated with the low-satisfaction project.
  • They'll prefer to do the safe and reduced-to-practice services that they know well, instead of innovating or being willing to take on tasks they haven't done before. This leads to stagnation and boredom.
  • In rare cases, they'll even cheat and file bogus surveys themselves to raise their scores.
On the customer side, the problem with customer surveys is that customers get so many of them already, and when survey scores influence compensation the company has to send one for every interaction, not just periodically. Even though a survey may claim to require only "a few minutes of your time," most are far too long. This discourages response rate, or worse, encourages down the middle lukewarm responses that are easy to give but reveal little.

The net effect is to reduce corporate learning. It's unfortunate, but true: you learn from the bad news. It's hard enough for even the very professional to seek out bad news over compliments because they know it will help them improve. Add a monetary disincentive, and even saints can turn into sinners.

A lot of this is unconscious, of course. Though a few bad apples will play the system consciously, customer be damned,  I believe most folks really do care. The unconscious motivations are corrosive.

The bottom line: if you want to get feedback from customers, great! Encourage them to not hold back and tell you everything you could have done to make it better. Then act on that feedback. Just don't hold a gun to people's heads and hold their paycheck hostage.