NPS is a valuable customer experience tool, and companies everywhere are using it to improve their experience and boost revenue.
But there are many misconceptions and misuses of NPS, and this piece will address those, in addition to showing you how you can actually get value from the tool.
See, when people lean on NPS like it’s a magic number, it doesn’t do much good. It doesn’t help that the original HBR article on NPS called it “the one number you need to grow.”
Unfortunately, the real world is a little more nuanced.
What’s So Special About The Net Promoter Score?
Net Promoter Score is a customer loyalty metric developed by (and a trademark of) Fred Reichheld, Bain & Company, and Satmetrix. It relies on only one question:
How likely is it that you would recommend our company/product/service to a friend or colleague?
Then, you break up responses into three chunks:
- Promoters (9-10). These are your happiest and most loyal customers that are most likely to refer you to others. Use them for testimonials, affiliates, etc.
- Passives (7-8). These customers are happy, but are unlikely to refer you to friends. They may be swayed to a competitor fairly easily.
- Detractors (0-6). Detractors are customers that are unhappy and can be dangerous for your brand, by spreading negative messages and reviews. Figure out their problems and fix them.
You calculate your Net Promoter Score by subtracting the percentage of Detractors from the percentage of Promoters.
Instead of distributing bulky customer satisfaction surveys, where even your managers don’t know what to do with the results, you have only one question, one metric to deal with.
Beauty in Simplicity
And that’s the big benefit of NPS: it’s simple.
Data is only worthwhile if you can do something with it – if it spurs action on your part. If you have a bunch of customer success metrics but don’t use them for anything, you’re wasting your time. As Fred Reichheld, creator of NPS, put it:
“Most customer satisfaction surveys aren’t very useful. They tend to be long and complicated, yielding low response rates and ambiguous implications that are difficult for operating managers to act on.”
What’s a Good NPS Score?
The range of NPS is from -100 (all detractors) to +100 (all promoters. Any NPS that is positive is usually perceived as good, and an NPS score of 50+ is considered excellent. But that’s just aggregate information, not totally useful because it’s not your specific industry.
Since NPS is a standardized market research tool, it is possible to compare it to your competitors, and within your industry as a whole.
For a simple number, there’s a bit of disagreement in how you gauge whether or not you have a ‘good’ one. But overall, I like to think of it as a metric to improve internally. As Adam Ramshaw said on a Quora thread, “a good net promoter score to have is one that is higher than you had last quarter.”
Kind of like a good conversion rate in that sense, then.
Well, What’s Wrong With NPS?
Simple and useful it may be, but the NPS doesn’t go without criticism (even on our own blog comments):
Some people clearly aren’t big fans. Prominent market researchers, usability experts, and academics have all voiced concerns over NPS.
The Downsides of Simplicity
It’s no secret that we like to think the world is simpler and more predictable than it actually is. To do that, we construct narratives and frameworks that allow us to make sense of the world.
When you take a short personality test, you get a result that lumps you with all the other ENFPs. When you’re born in September, you’re a Virgo. If you were good at drawing as a kid, you might have been labeled right-brained.
The reason I bring this up is that when you attach a business metric to “one simple question,” you’re competing with these other simplifications in order to try to fit a complex network of data into a simple number or lens.
Net Promoter Score has been shown to correlate with customer loyalty, retention and growth – but not always. But it does give you an anchor, a number that you can move up and down, and feel some progress.
You’d think from reading (sometimes sensational) blog posts that NPS is all you need. Throw away analytics and your other customer research tools, because this one simple trick will grow your business in your spare time!
It’s a bummer when you talk to survey design experts, or usability experts, or anyone who’s done a lot of market research, because they’ll tell you that there’s no way one question can tell you everything.
What NPS Can’t Tell You
According to User Testing, “NPS can tell you what your customers think of you, but not necessarily why they feel that way about your brand.”
This is true of any quantitative metric, of course. And that’s why NPS tools often ask a follow up question like, “what can we do to improve?”
There are also, specifically in regards to NPS, limitations in terms of branding and user experience measurement. Perhaps the best example I’ve heard is from Jared Spool in talk he gave:
NPS isn’t useless, but it can be dangerous because you have one data point. And you think you know your audience because you have a data point, but don’t realize how nuanced their interactions with your company actually are.
Other common criticisms of NPS hinge on its accuracy; does it actually model loyatly? Does it actually predict growth? Some claims people make:
- NPS performs worse than satisfaction in predicting growth.
- NPS uses a scale of low predictive validity
- It fails to Predict Loyalty Behaviors
- It fails to weight cultural variance
NPS and Predictive Validity
Though the original research that lead to the development of NPS showed strong evidence that NPS correlated with growth, studies since haven’t done much to support it.
In fact, there have been studies that say other indicators predict growth better and actually represent customer loyalty more accurately. Other studies show that NPS has much room for improvement in terms of validity.
For example, one study claimed that, “Recommend intention alone will not suffice as a single predictor of customers’ future loyalty behaviors. Use of multiple indicators instead of a single predictor model performs significantly better in predicting customer recommendations and retention.”
Another similarly stated, “given the present state of evidence, it cannot be recommended to use the NPI as a predictor of growth nor financial performance.”
While it makes intuitive sense that NPS correlates with growth (people are more pleased with the company, they tell their friends, more people buy stuff, etc), sometimes it’s not so simple and clear cut.
In fairness, even Reichheld said that NPS doesn’t always predict growth:
But it’s not just in fringe cases and industries. NPS has been shown to have low ‘predictive validity,” meaning the level of value a scale has in predicting future business performance. A study showed that NPS actually had the lowest predictive validity of 4 scales tested.
Usability guru Jared Spool also believes that, “Net Promoter Score is an ineffective instrument for measuring how your customers feel about you.”
Therefore, it’s pertinent to figure out correlative metrics that work specific to your business. Maybe it’s NPS, maybe it’s something else. But simply relying on NPS like you would a crystal ball is no way to forecast future growth.
NPS and Cultural Variance
CustomerGauge wrote about this, dubbing it the “Dutch Effect” (the Dutch won’t give you a nine or ten, I guess). Though they suggest that this has nothing to do with cultural perceptions of score ratings, but rather that the Dutch are used to getting poor service so they rate poorly.
I’m not sure that’s the case. I think it’s highly possible that there are cultural variations in how people perceive ratings.
How To Get Actual Value From NPS
Net Promoter Score, for any criticism it receives, is still very useful. As a dormant number or an industry comparison, it doesn’t do much for you. But as part of a fluid and constant operational management process, NPS brings great value.
So if you’re looking to implement NPS, here are some best practices for getting the most value from it:
- Ask Follow Up Questions
- Combine it With User Research
- Find and Fix Issues
- Market to Promoters
1. Ask a Follow Up Question
Seeing an improvement in your aggregate NPS is nice, but the real value comes in the feedback from follow up questions. At least that’s how Sachin Rekhi put it, writing on Andrew Chen’s blog:
Wootric also emphasizes having an automated follow up question to get some qualitative feedback along with your quantitative NPS. They suggest the following three follow ups:
- Promoters: What’s your favorite part about our product/service?
- Passives: What would make you love us?
- Detractors: What could we do to improve your experience?
2. Combine it with User Research
As mentioned before, NPS isn’t enough data to base decisions on. Adding a follow up question will open the door to valuable qualitative feedback and voice of customer research. But you also can’t neglect user research, including usability testing and other common conversion research techniques.
This will help you find some specific problems that correlate with low scores on the NPS. As UserTesting said, “You may discover that although your NPS is high, nearly every customer was having the same issue with your navigation, or that the copy on your pricing page doesn’t clearly explain what’s included in the price.”
Some techniques that can help pinpoint these usability issues:
- User testing
- Customer surveys
- Live chat
- Heat maps
3. Find and Fix Issues
NPS, though commonly used specifically as a customer success tool, can also be used by other teams, including your optimization team. Combining the overall score with the follow up question as well as the rest of your conversion research will help you build and prioritize test ideas.
While NPS, depending on the time you ask the question, doesn’t necessarily correlate with on-site user experience (NPS could and should include more of the whole experience, customer service included, optimization teams can work on specific issues that customers voice on the survey.
There’s a huge opportunity to turn detractors into promoters. As Reichheld said, “Every detractor represents a missed opportunity to add a promoter to the customer population, one more unpaid salesperson to market your product or service and generate growth.”
Marketizator also wrote about NPS and how you should deal with detractors:
“The best way to go with detractors is to investigate why they had an unpleased experience and why they will not recommend you. Knowing the answer, you can solve and improve that certain aspect and, in this way, transforming a detractor into a passive customer or even a promoter. Make them an offer they can’t refuse!”
4. Market to Promoters
Segmenting your audience into detractors, passives and promoters can help with your marketing efforts as well. One of the most popular ways to do that is to turn promoters into advocates. As Wootric put it, “when a customer rates your company at nine or 10, it’s the perfect time to ask them to refer friends.”
Lincoln Murphy, Chief Customer Evangelist at Gainsight, talks about using NPS to evangelize your best customers. Just because they say they’d recommend you doesn’t mean they jump up and do so. You’ve gotta nudge them…
Even if it’s not explicitly asking for a recommendation or giving away free stuff to promoters, there are ways to analyze your NPS data and find correlative actions that may predict success. As Sachin Rekhi put it:
Lincoln Murphy also recommends using NPS as a growth hacking tool, basically triggering Consistency Bias to reinforce positive beliefs:
While there are many blog posts and consulting firms that will tell you NPS is the be-all end-all metric, know this: there is no single survey question that can predict your company’s success.
There are also studies that have suggested NPS isn’t as predictive of growth, retention, or virality that the initial research claimed.
Either way, NPS isn’t all bad. If anything, it triggers an organization-wide attitude towards improving the customer experience. That’s not a bad thing at all.
And used in conjunction with user research, analytics, and A/B testing, NPS can be a solid addition to the conversion research arsenal.