Learn / Blog / Article

Back to blog

3 reasons why customers leave Hotjar (and how we found out)

On average, for every 100$ of Monthly Recurring Revenue (MRR) being newly generated on a given day, $32 is lost one year after (which means we’re left with $68 of MRR).

Our revenue churn is therefore 32% - almost a third of our revenue - after one year, which means that our revenue retention is 68%.

Thankfully, we cover our churn losses by generating new revenue from new Hotjar customers.

But revenue churn is a parasite slowly eating us from the inside. We need to kill it now before it grows too big.

UX design and analysis
CRO

Last updated

18 Sep 2021

What’s really at stake when we talk about revenue churn?

Even though these are common problems many SaaS companies face, there isn’t a lot of information out there about how businesses like ours overcome the hurdle of revenue churn. Below are the challenges we’re working through as we grow, along with examples of how we’ve failed and what we’ve learned.

Let’s start with a little Revenue Churn 101.

The more revenue is retained month after month, year after year, the better. In fact, it’s the gift that keeps on giving.

Here’s a hypothetical example:

Company A generates $1M in new MRR every year while Company B generates twice as much ($2M). One would assume that Company B has much higher chance of success than Company A.

But what if I told you that:

  • Company A retains 90% of its new MRR after one year

  • Company B retains only 60% of it

Which one is going to win in the long-term?

Let’s look at their revenue number for the next ten years (Company A is in blue, Company B in red):

Even though Company A (blue) generates half as much revenue per year than Company B (red), its total revenue is almost identical after 5 years ($4,638 vs. $4,767). After that, Company B's growth stagnates while Company A keeps on growing.

Company A can afford to spend more to acquire new customers and hire more team members which will, in turn, allow them to grow even more.

Right, back to Hotjar.

We’re doing well in terms of acquisition, but if we don’t understand why revenue churn is happening and how to tackle it, we run the risk to follow the footpath of Company B instead of Company A.

So we started a Churn Team...

Andrew Michael, our Business Intelligence Manager, gave me the history of how we’ve handled churn at Hotjar up until now.

ANDREW MICHAEL (CENTER) IS OUR BUSINESS INTELLIGENCE MANAGER AND FOUNDED MULTIPLE STARTUPS BEFORE JOINING HOTJAR.

The first mistake was setting up a Churn Team.

Surprising, right?

But setting up a team with the purpose of tackling churn was, for us, precisely the wrong way to go about it. As Andrew puts it:

“We made quite a few mistakes in the setup of the team in the beginning—one of which was setting up the ‘team’ in the first place. Churn shouldn’t be a team’s responsibility. It should be a mindset of increasing retention KPIs and engagement that spreads across the company.

Instead of focusing those efforts and resources on adding value for our current customers, we were sidetracked by focusing on the problem of those who’d left.

So, our first mistake was creating a Churn Team whose goal was to figure out how to stop churn—rather than how to strengthen retention. But then we made a few bloopers within that effort as well.

The Churn Team was composed of employees from different teams across the company, which we hoped would lend a well-rounded, multifaceted perspective. That was a good idea, but in practice, it probably wasn’t the best use of everybody’s time. The larger issue was that these team members also had other responsibilities for their respective departments that took precedence over their Churn Team work.

Our team members were focusing on the wrong thing and sidetracked by their other duties. We also hadn’t set clear expectations for them. We basically said, “yeah, you can maybe put in an hour or two a week, or maybe dedicate one day a week to this.” But without a clear expectation of hours to devote to this project, and the metrics we’d be tracking to mark our progress, there was no real accountability like there was for the team members’ other projects. Of course, they’d put in more time with those.

Sending an exit survey

When customers leave or downgrade plans, the first natural response from any responsible business is to ask “why are you leaving?

So in Q1 of 2017, we sent an exit survey to customers who had recently stopped paying for Hotjar (using Hotjar, of course!) and used this method to analyze the open-ended questions.

Our questions included:

  • What is your current role?

  • What industry do you currently work in?

  • If you could add any functionality to Hotjar what would it be?

  • What would persuade you to become a customer again?

  • Did you replace Hotjar with another tool? If so which one?

  • What tools do you use on a weekly basis, or are critical to your business?

And the 'million-dollar' question: what are the main reasons you stopped paying for Hotjar?

Here are the top three.

Reason #1: they use Hotjar on a per-project basis

A third of our churners (34% to be exact) said their reason for canceling their paid subscription was that the project for which they used Hotjar had ended.

Currently the free plan is enough for us. We will subscribe again in the future.” - User feedback

Client didn’t need that level of detail any longer and will be fine on free plan.” - User feedback

Didn’t need to use the paid version for the current projects.” - User feedback

We work with a lot of agencies who build and optimize websites and online sales funnels for their clients, so when their client leaves or the project is done, they feel like they don’t need Hotjar anymore.

Similarly, many customers in the ecommerce or SaaS industries like to use Hotjar for research projects (to redesign their website, improve a specific metric, or to develop new features) where they collect a bunch of data, make sense of it to build a case, downgrade to a free plan once the research is done, and upgrade again once they need the features again.

This explains in part why our re-activation rates are high: around 30% of revenue generated every month comes from users who have re-activated their paid plans.

For this reason, they’re not really “churned” in the strictest sense of the word. It’s more like they’re dormant.

Reason #2: they are overwhelmed with data

The second largest segment of churners said there was just too much information and they didn’t have the time or knowledge to analyze it—data overload.

“We were just collecting data for the sake of it, rather than specifically capturing answers to questions that we had. This meant the company was wasting time watching recordings and setting up heatmaps with no clear hypothesis or reason for doing so. Basically, the unlimited tiers meant that we were not using Hotjar analytically, but just as a deluge of data that became counter-productive.” - User feedback

“Awesome data, but wasn’t quite sure how to pull actionable insights from it.” - User feedback

“Time. We don’t have time for analyzing Hotjar results.” - User feedback

Some Hotjar users feel like they are facing a sea of data and do not know how to face it. Photo by Ruslan Valeev on Unsplash.

The customers who do best with Hotjar are those who are professional marketers and designers who specialize in the user experience—they already know how to parse through quantitative and qualitative data to pull actionable insights. We have to make the wealth of information our product has to offer more accessible.

David Darmanin, our CEO and one of our founders, has a vision for how this could work:

“The way we always wanted it to work is for Hotjar to be able to segment by site type, so our users can navigate Hotjar in the most logical way to reach their goals. For e-commerce, for example, the primary goal is more orders. So we’d love Hotjar to enable users to define their goal and automatically show them the metrics they need to achieve it.”

And once someone completes the goal, David says he’d like to ask them “How was this experience? How difficult was it?” and optimize, or create FAQ pages and other resources, from there.

Why haven’t we made that vision happen yet? If you’re frustrated, you’re not alone. So is David.

“I actually stopped talking about the vision because it became so frustrating over two years of just talking about this stuff and nothing happened. We shifted to talking much more about solving the most immediate scale problems before we can get to the vision.”

We knew when we launched the beta in 2014 that it was limited in term of functionalities.

THE TIMELINE OF HOTJAR'S BETA LAUNCH: DID WE LAUNCH TOO FAST?

We had to make a choice between launching fast and learning as we went, or launching more slowly by improving our feature set while still in beta. Our customers effectively made that decision for us by adopting Hotjar ‘as-is’ quickly, forcing us to scale and put perfecting every part of Hotjar on the backburner. As David recalls,

“Early on, you don’t see these problems coming. So at that stage, you say ‘Yeah, this is working. Okay, churn is not so good, but we’ll sort that out.’ And then you start thinking ‘Okay, if we spend X more on acquisition, we get Y amount back, which means we can hire five more people, and that will make it easier to fix churn, right?’ But then you get into this cycle of hiring and training which takes a lot of time. It took us two years to get the organization into position to fix the problem.”

Which leads us to the third reason...

Reason #3: they don’t get enough value and are frustrated by usability issues

The third largest segment of churners was frustrated by usability issues and bugs which we haven’t solved yet.

Your recordings never worked. I was tired of getting a canned response that you’re working on the software, but it’s been over a year.” - User feedback

My client did not see the value in it. Personally, I would love to continue using it for him - I think it gives us great insights.” - User feedback

Interestingly, our main competition—what our customers leave us for—offers even less value.

What is our main competition, you might ask?

Doing nothing.

When customers stop paying for Hotjar, many of them just stop monitoring their websites. They literally stop understanding their user behavior or collecting feedback. Which means our biggest enemy is inertia.

It’s easier to do nothing than to try something else. And the more effort we ask of our customers, to figure out how to use our product and analyze the results, the more inertia wins.

I just want to install it (on Wordpress) and track where visitors are going across my site. I don't want to have to create separate pages for Hotjar to monitor; I just wanted to follow and monitor where people enter my site, what they click on and where they leave. It wasn't easy to do that with Hotjar.” - User feedback

David says half of the concerns and wishlist items customers tell us about are already part of Hotjar’s vision; we just haven’t been able to ship them yet. Which means that the value we’d like to offer is not completely accessible.

That’s on us.

And in a future post, I’ll tell you how we are going to address each of these issues.

How are you tackling revenue retention?

Key takeaways

  1. We have a revenue retention problem at Hotjar: on average, we lose 32% of new Monthly Recurring Revenue (MRR) after one year

  2. To tackle this issue, we originally created a Churn Team composed of employees from different teams across the company but it didn’t work out because of competing priorities

  3. Instead, we should have created a company-wide mindset of retention and engagement

  4. Our customers stop paying for Hotjar for 3 main reasons:

    1. Their projects ended and they go dormant

    2. They didn’t know (or didn’t have time to figure out) how to analyze and use all of the data

    3. They didn’t see enough value or saw far too many usability issues or bugs

--

Full transparency: idea, structure, data, and recording for this post were provided by the author (Louis Grenier) while the piece was written by our copywriter Lauren.