NPS is a widely used metric in the SaaS world, and with good reason. In a business that is both dependent on high customer satisfaction as well as on fairly high costs to acquire new customers the ability to turn customers into promoters can be a huge boost to the success of a company. NPS can give you a good sense of how happy your customers are with your service.
Yet at the same time, NPS is not without controversy. Some people think that it can conceal as much as it reveals, that it is perhaps an overly abstracted statistic that lets companies hide real problems behind a seemingly benign two digit number.
Lately, some experts in the entrepreneurial blogosphere have been attempting to quantify how useful NPS really is, and, perhaps equally importantly, how it can be refined to present a clearer picture of the customers’ views of a company. David Skok at For Entrepreneurs, Brandon Hickie at openview, Sean Works at Kissmetrics, and Jason Lemkin at Saastr have all shared some useful thoughts on the utility of NPS.
NPS is calculated by asking customers a simple question: On a scale of 1 to 10, how likely is it that you would recommend [brand] to a friend or colleague? Respondents are then grouped into categories: people who answered 9-10 are called Promoters, people who answered 7-8 are called Passives and people who answered 0-6 are called Detractors. To calculate the Net Promoter Score, you subtract the number of Detractors from the number of Promoters.
NPS and Operational Rhythm
David Skok sees several inherent benefits to NPS. Firstly, he sees NPS as a good predictor of churn. Happy customers tend to stay. Dissatisfied ones tend not to. NPS gives a general sense of this with one number, and because it’s a standardized metric, it’s easy to compare your score to those of the other companies in your field and see how you compete. Additionally, Skok thinks there are great advantages integrating a Net Promoter program into the whole company’s strategy. Sales, marketing, product, service, and every area of the business can bring the data into their operational rhythm and use it to drive ‘continuous improvement and innovation.’
The Predictive Power of NPS
Jason Lemkin was initially skeptical about NPS, unconvinced that it was really all that useful a statistic. Firstly, he saw it as backward looking and essentially non-predictive when it comes to future prospects or even recent customers. Additionally, he was concerned that NPS is ‘abstracted away from revenue’, and too far removed from the things that really affect revenue like upsales and churn (a common complaint about non standard metrics). Thirdly, he saw NPS as a slow-changing number that allowed companies to distract from present mediocrity. “I saw big companies celebrate their NPS scores,” he writes, “even knowing their products were on the way to obsolescence as new entrants were displacing them.”
Yet in the end, Lemkin is an unabashed lover of NPS. For large accounts, he believes it functions as a predicative of net negative churn as well as future growth: a high NPS mean that the upgrades and expansion deals are going to come in. Likewise, a low NPS means that churn is coming. As long as NPS is high, the company is on the right track, even if the current finances don’t necessarily reflect that. And it keeps an SaaS company honest.
While different factions in the company (different teams, different executives, different departments) might have disputes over product quality and/or the direction to take, NPS acts as the voice of the customer: a high NPS means that the customers are happy, even if there are feature gaps. Likewise, if the NPS is low, there are serious issues with your product, even if some are convinced otherwise. Lemkin also likes that NPS acts as a confidence booster for startups. If you’re struggling to move past a certain level of ARR, you can take comfort knowing that a high NPS means that you’re doing something right. Eventually the lean times will end.
Brandon Hickie at OpenView warns of the potential for misusing NPS by failing to distinguish between promoters and detractors adequately enough. He points out that a company with forty promoters, sixty neutrals and zero detractors has the same score as a company with sixty promoters, twenty neutrals, and twenty detractors. Yet in reality, the company with twenty percent of its customers recommending against adoption of the company’s services is in a far different situation from the company with no detractors. He strongly recommends tracking promoter and detractor scores separately from the overall NPS to avoid this conflation. Hickey is also concerned by companies neglecting the customer tracking process when they switch to an NPS system, “even though the data that they were collecting in their previous tracking studies could be the perfect data for evaluating why customer sentiment is changing”. At the very least, the legacy system can be a good starting ground.
Hickie also cautions against failing to link NPS to the CRM database, as NPS ‘contains a significant amount of explanatory power in customer satisfaction levels’, and can be critical in determining where shifts in customer satisfaction levels are happening. This demographic information is key in remedying any problems. Without this crucial link, you’re in the ocean without an engine.
The Origins of NPS
Over at Kissmetrics, Sean Works has the NPS scores of some popular brands: USAA Banking is pulling in at a strong 87% , Costco at 77% , and Apple at a very respectable 72%. For comparison, average companies usually see numbers in the high single digits. “Stellar companies”, Sean Works writes, “operate at NPS efficiency ratings of 50 to 80%.” Sean Works also details the origins of NPS: Fred Reichheld debuted the metric in the Harvard Business Review after being inspired by Enterprise Rent a Car’s survey methodology. Enterprise asked customers only two questions, and only recorded the most enthusiastic responses. “By doing so,” Sean Works writes, “Enterprise could quickly keep tabs on the best performing locations and spread the lessons learned to other branches.”
NPS is certainly a valuable statistic for quantifying customer satisfaction. Although some may find it simplistic, its very simplicity is a virtue: sometimes you just want a yes or no answer, and in a very real sense, NPS is that answer. The question: are my customers satisfied?