Four hidden dangers of NPS

Customer surveys are everywhere, and they are getting mighty tedious, aren’t they?  They’re also getting more formulaic.

These days I brace for the question I know is coming: 

How likely is it that you’d recommend us to a friend or colleague?

Can you spell NPS?

NPS defined

Net Promoter Score is a measure of customer loyalty.  Based on a 0-10 scale, it sorts people into three categories based on their response to the “would you recommend” question.  Promoters (those answering 9-10) are deemed brand enthusiasts.  Passives (7-8) are considered satisfied but not especially loyal.  Detractors (0-6) are unhappy campers who could harm your brand through negative word-of-mouth.

NPS is a simple calculation:  NPS = % Promoters  –  % Detractors

Developed by loyalty guru Fred Reichheld at Bain & Company, it was introduced in a Harvard Business Review article in 2003.  NPS has become the darling of the left-brain intelligentsia.  Touted as the Ultimate Question and The Only Number You Need to Know, it has become a virtual industry.  Simplicity is a huge part of its appeal.

Critics of NPS contend that NPS is either ineffective or not as good as other measures.  But everyone seems to think it’s harmless.

I disagree.

Why NPS is dangerous

1. It makes your brand look clueless

Asking if we would recommend something to a friend or colleague assumes that your brand or company would actually come up in conversation.  That’s a steep hurdle.

Most brands and experiences simply don’t merit conversation.  Sorry.  Even if I like your brand, I’m probably not going to talk about you.  Would I recommend Citibank after a brief chat with a call-center rep?  What are you smoking?  Asking after such a non-event makes me think you have no idea what real people’s lives are like.  Not good for your brand.

2. It may backfire

When you ask if we’d recommend a brand or company, you remind us how perfectly unexceptional our experience was.  (Most are.  Exceptional experiences are, by definition, rare.)

Associating that mundane or second-rate experience with a binary yes/no question (even if it’s spread over a 10-point scale) may reinforce our disenchantment.  Before being asked, we might have forgotten the unmemorable experience.  Now you’ve made us realize: Nah, we really wouldn’t recommend you.  Is that what you want?

3. It may cause good people harm

A wonderful front-line worker may get a lousy rating associated with his/her work.  Is that fair?  Case in point: after a lesson on my new smartphone with a Device Specialist, I was asked whether I’d recommend my wireless provider.  How relevant is that?  The specialist was great (10), but my experience with that company is complex—and decidedly mixed (5).  Hanging the whole brand on one guy’s slender shoulders is unjust.

I’ve checked: yes, these ratings are often tied to the employees (and to their managers).   How deeply unfair.  My guy and his manager are not responsible for high prices, inaccurate and impenetrable bills, or other assorted horrors and annoyances.

4. A special danger for big companies

Big companies may be most at risk when using NPS.  First, they’re most likely to use it.  They live and breathe numerical data.  (And, frankly, they’re most able to afford Bain’s pricey counsel.)

More importantly, big corporations are also most likely to:

• Be perceived as out of touch with real people’s lives
• Offer the blandest brands and experiences
• Risk being seen as mindless, punitive bureaucracies

Last week I called two major credit card companies to see if my cards would work when I travel abroad (I’d heard some cards won’t because of new technologies there).  Both immediately sent me surveys using NPS.

Both call-center experiences were fine, fleeting, and unremarkable.  The reps did their job.  Would I recommend these companies to a friend or colleague?  No.  Why should I?  Because their employees answered a question?  But neither do I want to slam perfectly okay employees—especially those toiling in huge data-driven companies with little sense of warmth, flexibility, and humanity.

Fast, cheap and out of control

The impulses behind NPS (and customer surveys generally) are laudable.  Yes, companies should want to know how they’re doing.  Yes, tracking this in some simple way would be great.  And in some instances, the would-you-recommend question is entirely appropriate.  For example, when someone has performed a genuine service that you might speak of to someone else because recommendations are a vital way to find good resources.  Those situations, however, are rare relative to the volume of NPS surveys out there.  Instead, NPS is being used as a default for even the most trivial customer experiences.

Not every customer interaction warrants a survey.  Sending us one every time we interact with your brand or company is foolish and counterproductive.

And it NPS is not harmless.  You may decide the risks are worth it—and they may well be.  But please don’t assume it’s risk-free, just because it’s easy and popular.

We are busy, thoughtful people.  We do not exist to give you data.   No, I would not recommend using NPS to a friend or colleague.  Call me a Detractor.