Skip to ContentReforge logo

You’re (Probably) Using NPS Wrong

As more and more companies find success through “customer-centric” strategizing, nearly every product or service now asks about your likelihood to recommend it — usually using the industry-standard metric of a Net Promoter Score (NPS).

Despite becoming the poster child for measuring how customers feel about a product, most companies misuse NPS, either by measuring it incorrectly (often at the wrong time), or using it to measure something it can’t possibly measure.

The reality is: Most of us don’t go around sharing our favorite enterprise software, and more problematically, NPS doesn’t directly tell you how well you are meeting customer needs.

Rather than embracing what NPS does best – measure social viral acquisition loops (aka “Word of Mouth” or WOM loops) — people have turned it into an inactionable vanity metric.

In this post, we’ll discuss:

Net Promoter Score Graph

About the Author

Behzod Sirjani

Behzod Sirjani

Behzod is a Program Partner at Reforge, where he built the User Insights for Product Decisions program. Outside of Reforge, he runs Yet Another Studio and is a venture partner at El Cap. Previously, he led Research Operations at Slack and was a Senior UX Researcher at Meta.

What is NPS?

NPS is a single question survey instrument developed by Fred Reichfeld that asks respondents to rate the likelihood that they would recommend a company, product, or service to a friend or colleague.

Respondents choose a number from 0 to 10, where responses of 9 or 10 are considered “promoters,” 7 or 8 are considered “passives,” and 0 through 6 are considered “detractors.”

NPS scores are calculated by subtracting the proportion of detractors from the proportion of promoters (i.e. if you had 65% promoters, 10% passives, and 30% detractors, your NPS score would be 35).

2 Common Mistakes People Make Measuring NPS

Because NPS asks about whether or not you’re likely to recommend something, people often assume that NPS measures satisfaction, usefulness, and value all in one. It also seems to be the most familiar survey question in the English-speaking world, so it’s many people’s default when it comes to gathering feedback from customers. But just because something is common, doesn’t mean it’s good – or that it’s being used correctly.

People make two common (but fixable) mistakes when it comes to NPS:

  1. Using NPS to measure something it doesn’t
  2. Measuring NPS at the wrong place or time in a customer journey

Let’s dive into each…

Feature Product Strategy Lesson

Mistake 1: Using NPS to Measure Something It Doesn’t

NPS is a measure of evangelism. But evangelism isn’t directly correlated with the success of your product as much as loyalty, retention, or satisfaction — so you shouldn’t use NPS as a stand-in for those measurements in your strategy.

NPS Doesn’t Equal Retention

It’s important to separate the concept of evangelism from loyalty and retention, because high-scoring NPS customers are not necessarily highly retained customers (and vice versa).

Just because I was highly delighted with the product at one point does not mean I will continue feeling so. Also, I may be highly retained because using the product is critical to my job, even if I would love to find an alternative – or in a B2B situation, I have no influence in the decision about whether or not we use this product.Even further, organizations often fail to look at NPS scores by audience segment. If your highest-scoring NPS customers fall outside your core persona or your core frequency of use (over-indexed on power or casual users), those highly-delighted customers won’t be representative of the user base that drives your main retention.

Rather than use NPS to measure retention, you should be focusing on your retention curves — the visual representations of your product’s user cohorts that show you the shape and trend of how users retain over time. (Reforge goes deep on retention visualizations in our Retention + Engagement program.)

A standard retention curve, showing 50% of users drop off after week 5.

NPS Doesn’t Equal Satisfaction

Evangelism isn’t directly correlated with satisfaction, either. At the most basic level, people can be satisfied with a product they would not refer to others, and they also can refer a product they are not satisfied with because it’s the best tool for the job, even if it’s not a great tool.

Instead of measuring NPS in place of satisfaction, you should measure overall satisfaction directly using a question like “How satisfied or dissatisfied are you with X?”

In some companies, satisfaction is measured at feature, product, and overall levels. In these cases, overall satisfaction is best thought of as a composite metric that represents the holistic product experience — and one potential contributor to NPS.

At Slack, in addition to measuring overall satisfaction in our quarterly tracking surveys, we also measured things that we believed to be “drivers of satisfaction” like product completeness, fit for your team’s needs, performance, quality, and reliability. We ran regressions on these metrics to understand the degree to which the changes we made in a given quarter impacted these various drivers and overall satisfaction.

Mistake 2: Measuring NPS at the Wrong Time and Place

NPS is a measure of evangelism, which is helpful to understanding one way your product can grow. But if you’re measuring NPS incorrectly — at the wrong time or wrong level — you’re gathering data you shouldn’t use or using data to inform decisions that don’t exist.

Measuring NPS at the Wrong Time

Getting a pop up notification or email asking for an NPS rating just minutes after using a product for the first time is annoying and ineffective. In most cases, a few minutes or hours (or even days) may not be enough time to adequately assess whether or not you’d refer a product or service to someone else.

This is because there is rarely enough time for a user to complete one habit loop — the full series of events required to establish your product as a habit (something Reforge covers in depth in our Growth Series program).

Ideally, you should survey your customer about NPS after they’ve completed at least one habit loop: This is the moment when they have sufficient surface area to understand and potentially recommend your product.

The habit loop: The trigger inspires an action, than inspires a reward, which inspires another trigger.

Measuring NPS at the Wrong Level

NPS asks users about their likelihood to recommend a product or service, but I often see NPS surveys used at more granular, feature-specific levels where word-of-mouth isn’t useful or accurate.

When the part of the product or feature is not something that you could recommend on its own, NPS falls short.

One example of this would be the “snap receipt” feature within QuickBooks. Even if a user thought that feature was valuable and delightful, it isn’t representative of the product overall. A high NPS score for the snap receipt feature isn’t an accurate indicator of growth. Asking for NPS data at this granularity is a futile exercise — there’s nothing of value your team can do with it.

A screenshot of Quickbooks software, showing a menu of options, including "Snap receipt"

What NPS Really Measures: Word-of-Mouth Loops

Because NPS helps you quantify how many people love your product enough that they want to evangelize it (or show others that they're using it), it’s an ideal metric to measure the health of your word-of-mouth (WOM) growth loops.

A Word-of-Mouth growth loop: New users become engaged in the product, extracting value that entices the user to share the product offline or via social media to attract new users.

WOM loops rely on people spreading awareness about a product without direct utility or financial gain. And they can be incredibly powerful for many reasons:

  • They bring new people into the product for little to no cost
  • They increase your passive brand awareness for little to no cost
  • They signify market fit, even if only with a small audience

They shift the burden of growth from saturated external channels (like Google Search and Facebook ads), to owned channels (like making your product really great).

What Else You Need Besides NPS

NPS is a lens or starting point for you to get a view of which of your customers are most likely to be product evangelists – the ones who are most likely to drive WOM loops.

WOM loops can help fuel organic acquisition, but a healthy organization tracks its complete growth system — acquisition, retention & engagement, and monetization — as well as self-reported metrics like satisfaction and NPS. In that way, NPS can be a part of — but not a replacement for — measuring your product’s growth.

In many organizations, NPS is one of the longest-running, consistent sources of data from customers, and in that way it can act as a meaningful yardstick to benchmark other metric changes and correlations against.

If you're leading growth at an organization that over-relies on NPS or getting started in measuring growth, there are a couple of straightforward steps you can take to build out a robust system:

  • First, identify and build out the right measurement instruments for your customer acquisition, engagement, and retention, so that you’re not using NPS as a proxy metric.
  • Second, right-size the usage of NPS around what it does best – measuring WOM loops.
  • Third, start gathering feedback around customer satisfaction, so that you can complete your core measurement suite and have visibility into not just what people are doing, but how they feel as well.

Regardless of whether or not you’re using NPS, I’d encourage you to be clear about the decisions you’re making and gather the best evidence you can in service of making those decisions, not just the evidence that’s most convenient. For a deeper dive on evidence-based decision making, check out the User Insights for Product Decisions program or Data for Product Managers.

Behzod Sirjani