If you’ve been reading our other blog posts on the subject of Net Promoter Scores, then by now you should be feeling pretty confident about using it as a metric. However, one question that comes up a lot, because it’s the hardest to answer, is “What is a good NPS anyway?”.
There are a few reasons why this is can be a deceptive complex question, so we’ll look at each of them, and detail some of the options available for dealing with this.
The Net Promoter Question is clearly defined and easy to understand, but fits some organisations and customers better than others.
Because the basis of Net Promoter is recommendation to friends and family, there are some organisations (and individuals) where it may not be an appropriate question. If a product or service is very specialised then the respondent may not feel that their friends or colleagues would benefit from it, so may decide they are unlikely to recommend, even if they’re satisfied. The same may apply to healthcare organisations where the service provided is of a sensitive nature or causes unavoidable discomfort.
This is not to say that NPS should not be used in these circumstances, but that low, or even negative, scores may not be “Bad”. Context is key. This leads in to the next point.
NPS is a very top-level metric and needs to be contextualised
Because NPS is designed to produce a single number, relying on it in isolation doesn’t; tell you very much. It’s always a good idea to follow up the question with an optional comment box asking for the reasons behind the score given or just for general comments (as anything entered is quite likely to illuminate the score anyway).
Context can also come from making sure that you collect scores as an ongoing programme, rather than as occasional one-off. If you make a notable change to your products or services and the scores change in a significant way, then you’re on reasonably safe ground to assume that the one caused the other.
Another way of adding context is to purchase benchmark reports from consultancies and agencies that do research into this on a per-industry basis. These are often expensive, but can offer value.
NPS is designed to magnify small swings in the underlying data in the mid-high range, and nullify it in the low range.
Because of the way NPS is designed, small changes in the actual scores being submitted can lead to differing effects on the score depending on where in the range those changes happen. In the low range, a company with an tightly-grouped average score of 1 will get a very similar (in the extreme negative) NPS to one with an average of 5. Both companies will want, and need to make improvements, though the changes required will probably be different. This effect then switches for the other half of the scale and moving from 6 range to 7-8 and then 9-10 will change the NPS a lot.
Again, this is deliberate and can have value but when interpreting your data it’s important to look at it in context, and track it over time.
The importance that’s placed on NPS in a lot of organisations does sometimes lead to them trying to stack the deck, or alter it in some way so as to get a favourable score. It’s common to get organisations use some subtle (or not so subtle) colour cueing to get respondents to score more highly. This may not be a problem if the data is still being placed into proper context and used as a means of comparison, rather than an absolute, but will of course cause issues when being compared to data that’s not been collected in the same way.
Like any business metric, the most important thing about collecting NPS is being able to use it to inform the decisions you need to make. For instance, which improvements that need to be made to improve performance. NPS works best when it forms part of a rounded programme that looks at all aspects of the business. It also works very well as a “warning light”. Because of the magnification effect in the upper range it can be excellent at highlighting situations where there’s a slow decline in average scores. While a slow and gentle decline may not be noticed as a simple average score, the NPS is likely to come to a point where it “flips” from mostly one category to another and the score changes dramatically. That can have value for busy managers.
NPS®, Net Promoter® & Net Promoter® Score are registered trademarks of Satmetrix Systems, Inc., Bain & Company and Fred Reichheld.