My first encounter with the brilliant yet dubious nature of Net Promoter Score (NPS) came shortly after its “invention” by Fred Reichheld of Bain & Co. I’ll say it was the year 2004, since that’s the year after NPS made its debut, but I’m only guessing at this point.
For some time we had been running a customer satisfaction survey which ended the same as most do by asking our customers to rate their overall satisfaction from “highly satisfied” to “highly dissatisfied.” The results of other questions on the survey were then correlated to overall satisfaction in order to learn what factors were impacting satisfaction the most.
My boss at the time thought it would be a great idea to add the NPS question to the survey, he didn’t want to replace our overall satisfaction question, but simply add the NPS question on at the end. It seemed obvious to me that one or the other would be the way to go, but he was my boss so I gladly accommodated his wishes.
Of course the results between the two questions were highly correlated and both produced roughly the same correlation to other variables in the survey. So we had successfully lengthened the survey with no actual benefit. But at least it got me thinking. Is this question really better than our overall satisfaction question, and if so why? Since that time I have had multiple encounters with NPS across several companies, products, services, and customer types. Here’s my take on the pros and cons of NPS.
PROS
It’s Simple. C-level management loves NPS and its simplicity is the primary reason. It provides a quick read of customer happiness. The question itself is straightforward as well, leading to relatively little confusion from survey takers.
It’s Easy to Understand. Closely related to simplicity are the candid results: “40% of your customers are Promoters, 35% are Passives, and 25% are Detractors…your NPS is 15% (Promoters minus Detractors).” There’s no complicated statistical analysis to explain, yet the involvement of some sort of number crunching, even though it’s extremely rudimentary, for some reason helps lend validity to it for those that want to see something beyond top line results.
Benchmarking. Because NPS is so popular, hundreds of companies ask the question to their customers on a regular basis. Several of these companies have chosen to publish their results. Provided you are not in some obscure industry, you can easily see how your company’s score stacks up to several of your competitors.
It Measures What’s Important. If you are going to recommend something you are putting your reputation on the line. If someone is willing to take that risk, it’s a really good indication of their loyalty and thus likelihood that they will continue to support your company with their wallet.
Research Supports Its Value. In the development of the NPS question Fred Reichheld wanted to determine which of several similar questions had the highest correlation with repeat purchase behavior. Of the questions examined, the NPS question came out on top. Its ability to predict the likelihood that someone will purchase from you again is solid.
CONS
It’s Relationship Focused. NPS is focused on the relationship between the company and the customer rather than the transaction. It assumes that the likelihood someone will recommend the company is the same thing as how much they actually like the company, not the product or service. Someone may in fact have a good transaction but dislike the company as a whole or the other way around. So if the objective is to improve aspects related specifically to the transaction, NPS can miss the mark.
Some Just Don’t Recommend. I nearly always follow up the NPS question by asking why they would or would not recommend the company. In doing so I have found that there is always a healthy number who say they would not recommend the company because the “just don’t recommend things t people.” They may actually love the company, but for them that’s not the point. They take the question very literally focusing on the “recommend” part.
Limited Markets. If the product has a limited market, they really may not be able to think of anyone that they could possibly recommend the company to, since from their perspective, they may be the only one of their friends or colleagues who would want the product. Although it lessens the validity of benchmarking results, in order to reduce this problem, on occasion I have modified the question from “How likely are you to recommend [COMPANY NAME] to a friend or colleague?”, to “How likely are you to recommend [COMPANY NAME] to a friend or colleague who is in the market for [PRODUCT OR SERVICE]”
Picking “5.” A lot of people tend to pick “5” because they are indifferent. This is especially the case if they don’t have a lot of experience with or are not passionate about the company. The result is an over calculation of the percentage of Detractors. Someone who is a “5” or a “6” typically feels very different about the company than someone who is a “0” or a “1.” Yet they are all counted as Detractors. Hypothetically, what if 50% of the customers for one company were to select “5” and for another company 50% were to select “0,” with the remaining 50% distributed equally in both cases. Is customer sentiment for the two companies the same? It think not. Yet the NPS scores would be identical.
It’s Just One Question. The simplicity of NPS is both a strength and a weakness. Perhaps if you could just ask one question of your customers this might be it. Not that I am a promoter of long surveys, but to really understand your customers this question by itself comes up considerably short. It’s like a doctor who takes your weight as the only measure of your health. Of course your weight may be a good indicator of your health, but there is so much more that should be measured. Unfortunately some companies use NPS as a standalone measure in a one or two question survey and then assume that they understand their customers. In order to really understand how the customer feels about your company you really need to go a lot deeper.
Best of the Worst. This past November, most of us voted for what we perceived to be the lesser of two evils for president. Most weren’t passionate about who they voted for, it’s just that they perceived the other choice as being worse. This can sometimes be the case with a company that is in an industry that no one really likes, such as your subscription TV service, internet service provider, health insurance provider, or airline. Companies in these industries have very few raving fans, most customers who would recommend them do so because the competition is perceived as worse. When this is the case, NPS can give the false impression that your customers like you when in fact they don’t but are simply “settling” for you as the least detested option.
The reality is NPS is a good measure of customer satisfaction. The “cons” I listed aren’t so much cons as they are things to be aware of. It’s important to understand when NPS is appropriate and its limitations. What is the likelihood that I would recommend NPS to a friend or colleague? I don’t know…I’d probably pick “5.”