At Userpilot, NPS is an integral part of our product.
But let’s be honest, NPS sucks at times.
This is mainly because NPS was sold as a “quantitative” metric that is the ultimate indicator of growth—especially in product management.
The sad truth is: it’s not.
This metric ended up being adopted by companies in every industry. And now, a group of people see it as a vanity metric rather than a valuable indicator of customer sentiment.
We do, however, approach NPS in a way that’s valuable and actionable for us. And before I share it, let’s explore what NPS is and what’s wrong with it:
Get The Insights!
The fastest way to learn about Product Growth, Management & Trends.
What is Net Promoter Score (NPS)?
Net Promoter Score (NPS) is a metric that aims to quantify customer sentiment based on this survey question:
“How likely are you to recommend (our product) to a friend or colleague?”
The rating scale ranges from 0 to 10, and respondents are divided into three buckets:
- Detractors—those who answer between 0 and 6
- Passives—those who answer with 7 or 8
- Promoters—respondents that answered with 9 or 10
The score is then calculated by subtracting the percentage of detractors from the percentage of promoters.
For example: If out of 1000 responses, 42% of respondents were promoters and 17% were detractors. Then NPS is 42% – 17% = 25
What’s wrong with NPS?
The main problem with NPS is calculating it without any other metric or goal to pair it with.
It just doesn’t serve as a single indicator of “business health” like it was previously promoted.
And here’s why:
It’s yet another buzzword
The only reason companies are measuring NPS right now is for the buzz.
Everyone’s tracking NPS and comparing it with other businesses to understand what customers think of them in the context of competitors.
This is not inherently bad, but measuring NPS for the sake of benchmarking can be misleading for many businesses. Especially, when it means asking the same question without considering if it makes sense within the context of your product:
You see, NPS is mostly a relational metric (which means it’s focused on customer relationships). People choose their score for many reasons that might have nothing to do with liking your product. Depending on how you ask, a perfectly loyal user might respond negatively just because they don’t think your product could be recommended to someone who isn’t in their field.
For example, many companies make the mistake of sending the survey when users just sign up. This, without considering if new users can give a score without understanding the product first—leading to skewed results.
It’s not a satisfaction indicator
NPS is an indicator of loyalty, nothing else.
It only measures “promoters”, which are loyal customers that are so infatuated with your product that they can’t help but talk about it.
However, so many companies use it to measure satisfaction. And that’s problematic because NPS has nothing to do with customer satisfaction, or how healthy your user base is, or retention, or anything else.
If 100% of your user base is satisfied with your product but only hits a score of 8. Then your NPS would be zero.
Simply put, there are products that you can’t easily recommend to anyone even if you like them (would you recommend your ERP software at a family dinner?), which makes NPS pretty non-impactful to your business. A negative score doesn’t mean your business is doomed. A high score (although great to have) is only an indicator of how much WOM you might get.
You are adopting it as a leading metric
Unlike metrics like DAUs, MRR, churn, or sales, NPS is not valuable in itself.
Teams are not tasked to increase NPS by 20% in six months, companies don’t go to bankruptcy because their NPS was negative, and no one writes “I helped this company increase NPS” on their LinkedIn profile.
NPS rarely helps predict trends, churn, or future sales without a correlation. At most, it indicates the likelihood of receiving an unknown amount of referred customers.
And yet, a lot of companies use it as a leading metric without having any other context.
In contrast, NPS works best as a lagging metric. Especially, when comparing how NPS has changed over time and correlating it with other factors such as product updates, marketing campaigns, churn, or expansion MRR.
The science behind NPS calculation is wacky
NPS has no real science behind it.
First, it promises to be a quantitative metric for customer loyalty. But in reality, it’s just semiquantitative. It doesn’t really “quantify” customer sentiment in the same way you can quantify the revenue coming from upsells.
All this bucketing of “promoters”, “passives”, and “detractors” is simply arbitrary and is not based on any scientific method. All the math behind these scores is pretty specific for a very specific use case, yet everyone (from gyms to corporate firms) calculates it the same way.
Plus, NPS surveys commit the mistake of asking users about what they’re likely to do in the future.
This is important because if I ask you what you’re going to do on Saturday, it’s very unlikely you’ll give an accurate answer. Whereas, if I ask you what you did last Saturday, then the question becomes less subjective and more like a report.
In short: NPS is not a rigorous metric that serves as a strict measure of success, making it hard to tie it to actual business outcomes or actionable insights.
Book the Demo to see How Userpilot can Help you Extract Real Value from NPS Data
Why does NPS still matter?
Despite all its flaws, NPS can provide great value given the right context.
The main problem comes when product teams fixate on it and think of it as the holy grail of metrics.
But if you approach NPS as a baseline metric and leverage the qualitative feedback to get valuable insights, then there’s a lot you can do with it, such as:
- Having a basic idea of your customer sentiment levels and tracking it over time to see its trends.
- Using it as a churn signal. So if NPS declines for no apparent reason, it could indicate a churn problem that you need to investigate.
- Segmenting NPS responses by different groups of customers (and finding potential groups where detractors are concentrated).
- Tracking the in-app behavior of detractors to understand their experience with your product.
Plus, it’s easy to implement. You don’t need a data engineering team to send NPS surveys and calculate them. Software like Userpilot (and many others, too) can do it without requiring any coding skills.
And speaking of implementing NPS successfully:
How does Userpilot make NPS make sense?
At Userpilot, we mostly use NPS for qualitative insights. Trying to “science it out” is simply not worth it.
We calculate NPS just as a lagging metric, and then we use the NPS responses to find opportunities to target specific strategies.
This involves:
- Sending transactional NPS surveys with follow-ups to understand what converts users into promoters or detractors.
- Filtering NPS responses by keywords to identify common issues among detractors.
- Segmenting users to see if the majority of detractors come from a specific group.
- Analyzing the in-app behavior of NPS detractors to capture their issues.
Let’s explore each of these NPS tactics:
Enable sending contextual NPS surveys
An NPS question alone tends to work very poorly as a transactional survey. It tends to work best with relational NPS surveys.
Unless you add a follow-up question.
With Userpilot, you can trigger NPS surveys based on multiple conditions such as target audience, date, journey stage, and demographics—and add a follow-up question.
But this process is not only to measure NPS but to correlate overall NPS scores with specific product experiences.
This makes insights more actionable and helps you focus on what matters. For example, if NPS from activated users is higher, you can find out through follow-up questions:
- The most valuable features for them.
- The JTBDs they’re trying to fulfill.
As a result, you’ll be able to optimize and refine your product’s core features to fulfill these JTBDs instead of following assumptions.
Tag responses to identify common issues among detractors
With Userpilot, you can tag qualitative responses to filter them based on specific keywords.
It’s extremely helpful to identify common issues among detractors and find ways to solve them.
For example, you might find out that active users who scored 0 or 1 are often mentioning that there’s a key feature that doesn’t work properly.
With this data point, you can investigate the bug further and find out it’s a bug that only happens to users with the latest macOS version. Then, you can fix it and prevent other users from becoming detractors in the future.
Segment users to understand where your detractors are concentrated
Another useful way to analyze NPS survey results is through segmentation.
With Userpilot, there are multiple ways to compare and narrow down NPS survey results, for example:
- You can segment your users based on user personas—for instance—and watch how NPS changes over time among them.
- It’s possible to see the NPS trends of different user segments and figure out if one of them is at risk of churn.
- Userpilot lets you see the profiles of both users and companies, including their NPS. This way, for instance, you can find that there’s a big account with low NPS and ask the customer success team to fix their unique problems.
- Segment NPS promoters and analyze if there’s a common characteristic among them.
Overall, you can play with your NPS survey responses as you see fit and find insights without having to write code.
Find root causes of low NPS scores with product usage and behavioral data
Userpilot isn’t just an NPS survey tool, it’s an all-in-one platform that also includes product analytics.
This means you can easily track the in-app behavior of both promoters and detractors to understand their experiences. Think of finding the root cause of a common problem, defining the happy path of a loyal user, and more.
For this, here are the tools you can use with Userpilot, including:
- Autocapture. Which lets you perform retroactive analysis and see the behavioral data of a detractor. For instance, maybe all detractors tend to show a decline in product usage over time because they didn’t find value in it.
- Session replays. Where you can literally sit and watch how both promoters and detractors interact with your product.
- Path analysis. Which draws the path of specific segments (be it promoters or detractors) with your product. Perfect for defining the happy path.
With these, it’s completely possible to get actionable insights if you pair NPS with more data.
Conclusion
NPS has its challenges, but when used thoughtfully, it can provide actionable insights that drive real value for your business.
As we showed you, features like contextual surveys, response tagging, and advanced segmentation make Userpilot one of the best no-code tools for tracking NPS.
Want to see how Userpilot can help you extract real value from NPS data? Book a demo today to see how you can do this without code!
Userpilot NPS FAQs
What does Userpilot do?
Userpilot is an all-in-one product growth platform that enables businesses to deliver personalized, in-app experiences to drive user engagement, retention, and adoption. It offers tools for user onboarding, product analytics, and NPS surveys to help you create data-driven strategies for growth.
How much is Userpilot startup?
Userpilot pricing starts at $249/month for up to 2,000 MAUs. It includes both a free trial and a free demo for those who want to experience its value.
Book the Demo to see How Userpilot can Help you Extract Real Value from NPS Data
About the author
