Illustrated artwork showing laptops (yes laptops) in conversation with each other and holding up numerical rating cards

From vanity to value: How Intercom conducts NPS surveys

Main illustration: CACHETEJACK

There’s a good reason that the seemingly simple Net Promoter Score (NPS) has become a ubiquitous, revered statistic in modern business – this single number is viewed as a measure of customer loyalty, a way to benchmark competitive performance, and has been proven to correlate with revenue growth (hence the popularity).

But all too often the way the underlying data is collected makes NPS little more than a vanity metric, easily rigged to give the impression of positive customer sentiment, leading to shallow insights and false assurances.

At Intercom, we’re on a journey to do NPS the right way. Here, I’ll share why we decided to revamp our NPS strategy, what survey mechanics proved most effective, and how we’re delivering value by providing actionable customer insights across our company.

What is Net Promoter Score?

Let’s give some context, starting with a simple definition – Net Promoter Score or NPS is a benchmark of customer loyalty and is measured by asking a range of customers a simple question: “How likely are you to recommend this product or service to a colleague or friend?”

“First developed in 2001 by Fred Reichheld, NPS quickly became a key metric in all sorts of industries, an agreed measure of customer loyalty”

First developed in 2001 by management consultant Fred Reichheld, NPS quickly became a key metric in all sorts of industries, an agreed measure of customer loyalty. It’s different from a customer satisfaction score (CSAT) in that it measures broad brand sentiment, rather than reflecting the impression of a single interaction.

Revamping how we do NPS surveys

NPS surveys aren’t new to Intercom – we’ve been running them in one form or another for years. However, while the results and feedback were being surfaced, there was no single team taking full responsibility for consistently running the surveys, so the program lacked strategic focus and measurable impact.

Intercom’s Customer Advocacy team formed in 2019 with a mission to turn every Intercom customer into an advocate. One of our first foundational projects was to refresh Intercom’s NPS program as it can provide daily information on the wants, needs, and expectations of our customers – in short, a sign-post leading the way to passionate fans.

This formed the basis of our NPS strategy: to build a balanced picture of how our customers feel about Intercom, identify opportunities for improvement, and enable action to provide value for both the customer and the company.

Having a dedicated team responsible for NPS ensures the program can adhere to this clear vision and also be held accountable for its success.

Formula for Net Promoter Score

Even if you haven’t previously heard the term NPS, you’re probably familiar with the survey. It’s a simple format asking you to rate how likely you would be to recommend a company/service on a scale of 0 to 10, with 10 being the most enthusiastic endorsement of the product or service.

“While both the survey question and NPS formula used are fairly standard from business to business, there are plenty of small differences in how it’s designed and deployed that can hugely influence the ultimate score”

The overall NPS score is calculated by subtracting the percentage of Detractors (people who answer from 0-6) from Promoters (people who reply with 9 or 10). People who respond with 7 or 8 are called Passives, and those results don’t feed into the total NPS.

This gives a result somewhere between -100 and +100, which is your overall Net Promoter Score.

While both the survey question and NPS formula used are fairly standard from business to business, there are plenty of small differences in how it’s designed and deployed that can hugely influence the ultimate score.

Defining which customers to survey

It sounds obvious, but who you survey has one of the greatest impacts on your overall NPS score. Some businesses restrict the survey to their “ideal” customer segments, those who spend the most money, or even those who have already left high CSAT scores.

In line with our NPS strategy, we believe that true value can only be found when every customer has a voice, irrespective of their spend level, customer segment or satisfaction. This approach gives us more data to build a fuller picture of how our customers feel about Intercom, while also having the ability to deep dive on specific sub-segments to provide relevant insights where needed.

When to do an NPS survey

When it comes to survey timing, again companies are faced with a myriad of options. Do you survey your customers quarterly? Or after specific events occur such as purchases or business reviews? Or keep it randomized? And how do you treat people in the same company? Should colleagues get the survey at the same time or dispersed throughout the survey period?

For us, we again made a decision on what would provide the most valuable and actionable insights so we chose to survey customers at specific milestones after signing up (at two months, six months and every six months from then on). This helps us understand the journey of our customers, from their first experience getting to grips with our tools, right through to their “expert” years. To enable this, our chosen survey partners at InMoment (formerly Wootric) developed a feature allowing us to analyze responses by the number of days that have passed since the respondent’s sign-up date.

Where to present an NPS survey

In order to provide true value you need a robust and reliable data set. In other words, you need as many customers to respond to your survey as possible.

As a messenger-first company we know the power of using in-app communication alongside traditional email, so it was important we found a tool that could survey people directly in the Intercom messenger. Luckily, there are a number of great partners in our app store who integrate with Intercom for surveys and feedback.

Did we make the right choice? Since our new survey launched last September we’ve seen in-app response rates 123% higher than email for our NPS survey.

Choosing the NPS survey sender

As with the mechanics, there are plenty of small differences in design of your survey message that can have a huge impact on outcome.

Most NPS commentators advise companies to send the survey from a real person, and this mirrors Intercom’s own ethos that personal messaging helps build relationships. All of our communications have historically come from a person; eg. Phil Byrne helps you get onboarded, Bobby Stapleton sends you billing notifications, and so on.

“Would customers be less likely to give us the honest, candid feedback we need if they felt like they were giving it to a real person?”

But we wondered for NPS, could this personal approach actually hinder what we were trying to achieve? Would customers be less likely to give us the honest, candid feedback we need if they felt like they were giving it to a real person? There was only one way to find out – we ran an A/B test, and the results were extremely revealing.

We created a new, generic, Intercom NPS identity, and tested this against sending it from a person (me in this case). While the in-app surveys saw similar response rates, customers who received an email from our corporate NPS identity were more than twice as likely to respond than those who received the survey from me. I try not to take this personally 😅.

What else to ask in the survey

As mentioned, the NPS survey is traditionally a rating between 0 and 10. Once someone leaves you a rating, the NPS part of the survey is essentially done. But while the numerical score gives us a high-level measure of how we’re doing as a business, the optional questions that follow, often called “driver” questions, are how we learn what we need to do to improve. This critical extra information tells us why the user gave us the rating they did, and it’s where the true value in NPS lies. It’s what helps us make a better product and experience for Intercom customers.

Some companies go granular here, providing multiple choice options to help direct and categorize user feedback, but we feel it’s better to start broad to ensure you don’t unintentionally bias responses in any way.

For customers who leave us a rating between 0 and 8 (the Detractors and Passives) we simply ask “How can Intercom do better in future?” However, for customers who give us a 9 or 10 rating (aka Promoters), we ask what their favorite thing about Intercom is. This is to add balance to feedback from Detractors and Passives. One person might not recommend Intercom for the exact reason another might advocate for us, and so it’s important to record both viewpoints to aid strategic decision making.


Consider your follow-up message

One of the reasons NPS surveys are so popular is because they’re quick to complete. While that’s true, it’s still important to recognize that folks have taken time out of their day to help us, tell them how we plan to use the information for their benefit, and set response expectations in case they need urgent assistance.

Just like the follow-up question, these “thank you” messages are tailored depending on the rating a customer has given, and are sent using our recently released event-based messaging feature. There’s value in doing this as it strengthens the relationship with your customers and creates a better survey experience which will make them more likely to respond in future.

Delivering action with your NPS results

The last piece of the puzzle is to listen to your customers and put their feedback to good use. At Intercom, we do this in a few ways.

Firstly, for account-managed clients, ratings and comments are surfaced to their Relationship Managers, giving them access to the honest feedback needed to maintain a transparent and constructive relationship. They can then take personal action to relieve pain points or celebrate success.

Alongside this, all NPS comments are tagged based on their area of feedback and fed into a filterable NPS dashboard to surface findings to our internal teams ahead of critical planning milestones. This essentially gives the customer a seat at the table when it comes to our future roadmap for both product and service-focused teams.

Finally, for the wider business we create a monthly report detailing our overall NPS score, a breakdown of scores by different customer segments, and any top line trends we’re seeing. The slide deck as well as a short video talking through the highlights are sent to all levels of the company, keeping each and every Intercomrade connected to the people we’re all working for – our customers.

Advocating for NPS insights

While we don’t believe that NPS should be the only input a company considers, the work that the Advocacy team has undertaken in the last 12 months has transformed it into a truly valuable metric – it serves to shine a light on the opportunities ahead for Intercom and it’s customers.

We’ll continue to experiment, learn, and adapt this program while always staying true to our mission to turn every Intercom customer into an advocate.

Meet Interconnected – Intercom's new community forum