Whether Fred Reichheld intended this or not, the greatest weakness of NPS is how it makes enemies of our customers who only wanted a good experience.
- The good experience that was promised to them by our advertising, which we spent thousands of dollars on.
- The good experience that we spent thousands of dollars training our staff to deliver.
They had a bad experience, and we label them Detractors. If the experience was so-so, then they are Passive. And thus, the blame shifts (as it often does) on the customer, and not our Standard Operating Procedures, or lack thereof.
And because NPS is often tied to the executive bonus, those customers that we have failed, somehow feel like our enemies.
Too many Detractors, and we will not be promoted.
The problem with NPS is that it completely ignores the subjectivity of interpretation.
First things first: eliciting and understanding what people say is hard.
How hard? Well, let us say very hard.
Or extremely hard.
Or, very difficult.
Or, extremely difficult.
Depends on what scale you are using.
And this is a huge part of the problem with NPS: putting a numeric value on an emotional response which differs from person to person.
Marc Antony: ‘Friends, Romans, countrymen, lend me your ears; I come to bury Caesar, not to praise him. The evil that men do lives after them; the good is oft interred with their bones. So let it be with Caesar.
Interviewer: ‘Erm, we have a scale here, so would that be more like very happy with Caesar or very unhappy?’
Like any other scale available, NPS is most useful for indicating trends or directions…for example, if we are seeing lots of people moving from fair to good and upwards, there is a directional thing happening.
That is the role of analysis, the feeling.
Where NPS goes wrong as a scale, is that puts too much focus on the score, not the customer.
Whether using NPS or not, my advice is to try to understand what you are not doing, that discourages advocacy. What are the memorable moments that your brand is known for? Good ones or bad ones?
Yes, it really is that simple.
The problem with NPS is that it is demonstrably not the ultimate question, it is not always even the right question!
We have all been there.
That needy message from an App or an installed program:
Do you love me? Won’t you recommend me?
Checking in regularly with customers is always a good idea. It mirrors what happens in REAL LIFE situations, where a waiter or salesperson will check if ‘everything is okay with you guys?’
The art of the non-intrusive intrusion.
When designing any customer experience research, just make sure to ask the right question.
Do not try to measure likely recommendation of something that: People. Are. Highly. Unlikely. To. Recommend.
Or the Calculator App.
Or complicated Life Insurance policies.
It is perfectly acceptable and normal that people can be very satisfied with something, but do not feel inclined to recommend it. More commonly, we tend to wait until someone asks us for our recommendation.
And this is such a huge problem with NPS: we do not ask people if:
- They have actually recommended our brand/products/services?
- If they recommend any brands/products/services?
- What tends to encourage recommendation?
The results lack context. All we know is that our score is either positive or negative, and if we have too many Detractors means it is unlikely we will be Promoted.
Qual eye for the quant guy
While NPS is a sturdy way to measure satisfaction and advocacy, we get the ‘golden nuggets’ from the verbatim, and that is where the focus should lie.
As an organization, focus on the question of: will you advocate for us?
From an operational perspective, that means trawling your way through all the verbatim to get the ‘magic’ on what you are doing well.
This is hard work, no doubt, and should not be boiled down to sentiment analysis. That defeats the purpose.
Think of verbatim as effective ‘case studies’ of what you can adopt and adapt as best practice.
I strongly recommend it.
Carl is also a member of EGN Singapore Marketing Leaders peer group.