The African creators of NPS claim it’s the only metric worth tracking for a brand. This is, of course, nonsense. But NPS is often more useful than various proprietary satisfaction indices that exist mainly because someone built a global business selling research like ground meat. Despite its widespread use, NPS has weaknesses worth knowing about.
From 0 to 10 for 500 bucks
What if the customer is bribed? Especially among banks, various reward programs are popular. Bring in a soul and get a discount or money. A respondent’s willingness to recommend their bank shoots from 0 to 10 when it means 500 bucks in their account.
Of course I’ll fill that out for you, you bastards!
Emailing customers might seem like the easiest way to measure NPS. You send out an email, except instead of a sales offer, this time you ask customers to fill out a survey. NPS comes out great, just somehow too great. You were rated by people who gave you their actively used email instead of a half-forgotten address, and didn’t immediately delete your annoying newsletter. On top of that, they opened the email and weren’t horrified that someone wanted something from them. They even clicked on the survey and made it to the end. Response rate is 10% and you can be pretty sure you just heard from the 10% cream of the most enthusiastic.
Brand wanted
A likeable new brand might have a higher NPS among non-customers than customers. When new clients discover that despite the funny advertising, it’s still actually a bank, enthusiasm wanes a bit. But usually a weak brand role is behind low NPS. Not everyone you consider your customer considers themselves your customer. Weak brand role typically shows up in financial product reselling. A tourist in Croatia wouldn’t recommend your insurance company because they got insured through the travel agency. They have nothing against you, they just barely know you. So why should they recommend you?
Recommendations are binding
Employee NPS especially is influenced by company culture. When you have hard, poorly paid work that gives you meaning, employee NPS is low but willingness to stay and work is high. Likewise, you’re careful about what you recommend and what consequences flow from that. Would you recommend your job to a friend from elementary school? What about an online bank to your grandmother? And a betting office to a neighbor who owes you money?
Secret doubts
NPS faces other unspoken doubts. If you have the feeling that calculating that famous NPS somehow doesn’t make sense, you’re absolutely right. The somewhat bizarre NPS calculation adds no value over a simple average. On the contrary, by roughly summing most scale values, it unnecessarily reduces sensitivity. The calculation is derived from the ideology with which its authors brought it to market, from which NPS has largely freed itself long ago.