Is Your Net Promoter Score (NPS) Junk?

A couple of recent conversations with clients, and one hilarious interaction at a popular chain restaurant, have finally convinced me to break my silence on the issue of NPS Scores, how they’re being used in #pbsa and #studenthousing, and whether anyone should pay any attention to them. So here’s what I really think…

Routinely deployed by everyone from telecoms giants to your local dentist, the Net Promoter Score question is very familiar to everyone - "On a scale of 0 - 10, how likely are to recommend [your dental hygienist*] to friends or family?" *substitute as appropriate.

The history of NPS and it's original purpose is for another time, but it's fair to say that it is revered and cherished by Boards, CEOs and marketers the world over. The attraction is obvious. One simple question to rule them all.

But if you ask a professional researcher about NPS, prepare for an altogether more sceptical response. A raised eyebrow, perhaps combined with a light tut or a deeper sigh. Why such a lack of reverence for this hero metric of marketing and customer service? The answers are both strategic and practical.

First, and most importantly, NPS has too often been adopted as an organisational KPI by default, with no consideration given to whether it is actually the right fit for the business' strategy. NPS may be the right customer KPI for some brands in some industries. But it's not a panacea and I'd argue it's probably not the best choice for most. This is because the unique way NPS scores are calculated only rewards improved customer performance at the top end of the scale. Moving customers from scoring 6 to scoring 7 is a boost to NPS. Moving customers from an 8 to a 9 is also a boost. Convert your most ardent critic from a zero to a 6 and you'll see no impact at all on your NPS score. NPS doesn’t give a ****.

Just think for a moment about applying that logic to, for example, a school, or a hospital, or... accommodation for students living away from home for the first time in their lives.

Pick the wrong KPI – one that doesn’t properly align with your values or your commercial strategy - and you can quickly confuse and demotivate your team, waste time and resources on the wrong priorities, and ultimately do more harm than good.

You may be thinking "Ok, but we can see the breakdown of scores so we can still push for improvements among those customers who are really struggling." And of course you'd be right, in principal. The problem is that too often, organisations choose NPS without a strategy, and before they realise it, NPS has BECOME their strategy. Staff are monitored and Boards are remunerated based on this single, handy, all-in-one score. And all anyone is focused on is converting 6's to 7's and 8's to 9's. You might well end up trying to avoid having zero-to-six scorers (aka 'Detractors' in the lingo) as customers in the first place! Who wants that miserable lot in their school / hospital / accommodation after all?

And this leads to the second big problem with NPS. Whisper it quietly now... it's simply far too easy to cheat. Partly that's inherent in a KPI based on a single question, but the design of the NPS scoring system combined with the fact that customers and staff alike know full well how it works, makes it particularly vulnerable. Often it's whole companies, not rogue individuals, who are deploying ludicrous tactics in a desperate and nihilistic effort to move their scores ever northward.

Coloured answer options, inverted scales, strategically placed frowning/smiling emojis, helpful 'tweaks' to the question wording, and little pre-question hints are all commonly deployed. I've even seen versions with a big arrow pointing to the 'right' answer! Recent pop psychology manipulations I've been subjected to by customer service staff include "A nine would be fine!" and "An eight is average but I hope you'll agree my service has been better than that!".

?

Honestly, who are we kidding with this nonsense? It’s not a real NPS score.

?

Don't get me wrong, it is possible to do NPS properly, if that’s your cup of tea, and I think we do a pretty darn good job of it in the Global Student Living Index. For example, although we can't control all the client promotional activity that goes on, we do actively monitor it and provide ‘guidance’ back to clients where necessary. We also hold the question back until midway through the survey (further from external influence), and we're very strict about the wording and presentation of the question itself. We never, ever use partial response data (there are GDPR and ethical issues with that, as well as data quality ones), and in fact we actively discard around 20% of all the survey responses that are submitted because they don't pass our response quality algorithms. That probably sounds like madness! Until you come to realise that having a lower volume of good quality data is far, far better than having a lot of poor quality data. This is true for medical research, it's true for cancer trial data, it's true for survey data, and it's true for NPS too.

I'm not kidding myself that NPS is going to disappear overnight, or that clients don't want the biggest response rate they can get (provided it's good quality). And I'd stand by our NPS scores as the most robust in the business. But I do hope more organisations will think actively about the right KPIs for their particular brand and strategy. Better alternatives are available, including (of course I would say this) the GSL Index Score. A composite KPI designed specifically for PBSA, that separates Operational and Structural factors so that when there’s a disappointing score your teams know for sure if it’s them or really the building that’s the problem. The Index Score is based on multiple factors that tell you why, not just what your score is. It’s stable, trackable, benchmarked and almost impossible to manipulate. We're pretty proud of what we've built there and I'd be happy to tell you more if you’re interested.

In the meantime I hope you've found my service today to be more than satisfactory and if you don't think I deserve a 10, I'll still appreciate a 9, I'll learn from it and do better next time for sure. Ok, thanks, bye!