The customer experience industry is pretty well centred on one question – “On a scale from 0 to 10 how likely are you to recommend this brand?”. The proportions who give a score of 9 or10 minus those who give a score of 6 or less give us the famous Net Promoter Score – the net number of people who are likely to recommend something.
Recommendation is a great endorsement. If somebody recommends something it is almost certainly worth a look. Or is it? During the lockdown you have probably enjoyed/suffered (delete as appropriate) recommendations galore. People will have suggested books, films, TV shows and of course, the ubiquitous jokes and funny videos. The question we have for you is “how many of these recommendations were worth it?” The problem is that what you like isn't necessarily what someone else likes. If someone recommends Hillary Mantel’s new books, will they captivate you? If you recommend the box set of Mad Men, are you worried what that might say about you?
People recommend things for different reasons. Some do so because they feel that their suggestions position them as a thought leader, or at least someone who has wonderful taste. Others do so because they like to control your actions. If you are on holiday and the tour guide takes you to an incredible place that sells carpets, just check whether it is owned by their brother-in-law. The Lies Of Locke Lamora by Scott Lynch gets an NPS of 60 (a top drawer result) from over 2000 readers but you have to be into fantasy literature to like it in the first place.
For recommendations to work they have to be aligned to you and your needs. Someone who knows you well will be able to suggest a restaurant, a recipe, or a book that will grip you. But how much can we rely on the recommendations of others? Since we know that populations readily segment into different groupings, surely recommendations are only valid if they are from people who are like you.
It is therefore necessary to ask a good number of questions beyond "the recommendation question" that enable us to slice and dice the Net Promoter Score to see how the results differ within segments. If the overall NPS is high (over 60) it is likely that the brand is appealing to all segments even though there may be variations between groups. A word of caution here because we are assuming that the question is asked of a representative cross-section of target customers. In the case of the score achieved by the Lies Of Locke Lamora the NPS score has to be treated cautiously as it is provided by a self-selecting group of respondents – people who chose the book because they like that genre.
Reviewing the NPS score among different segments is especially important in business to business markets. Many business to business customers occupy a niche and it is only people within that niche whose views really matter.
So, at the end of the day what are we saying? We are saying that recommendations are important as long as they are interpreted within a context. The NPS question is sometimes the only question asked in a survey with no classification or profiling data to provide a context. We should remember that it is important to ask the right question but it is equally important to ask it of the right person.