What is churn analytics?

It is a truth universally acknowledged that it’s more expensive to acquire a new user than it is to retain an existing customer. And not just a little more expensive. In fact, “to make up for the loss of one existing customer, companies have to acquire three new customers”, according to McKinsey.

Despite this, it’s all too easy for growth-focused companies to focus most of their time, energy, and efforts on acquiring new users, and are happy when those acquisition numbers climb steadily. The challenge comes when growth teams realize that their retention rates are low, and those newly-acquired users aren’t sticking around. 

The solution? Improving retention and reducing customer churn to maximize the value you get from each customer.

The first step to reducing customer churn is to understand why it’s happening and where it’s coming from. And to do that, you need churn analytics. 

Let’s talk about what churn is and how to use analytics to understand and reduce it. 

Customer churn, also known as customer attrition, is when a customer essentially stops being a customer—ie, they choose to stop using your products or services. The customer churn rate is a way to express the percentage of customers who stop doing business with you during a defined period. 

Every company experiences churn—the key is to understand why your customers are churning and decrease the proportion of those that do so. There are different ways to calculate churn for different business models, which we’ll dive into in more detail further down. 

Nearly all companies experience churn, and pretty much all of them want to avoid it. Customers don’t have to be paying customers to churn; a user forgoing Google or Facebook is considered customer churn, as is a large corporation canceling its office building lease. 

Because churn hurts company growth, many brands are (rightly) obsessed with reducing it.

Churn analytics helps companies plug the leak in their customer bucket, to borrow a common analogy. Many businesses prioritize attracting new customers over retaining existing ones and don’t notice the toll churn takes until it seriously erodes profits. By some measures, as many as 97 percent of customers who churn do so silently, without leaving feedback or clues as to why. 

As we mentioned in our 2024 Benchmarks Report, retention has been particularly challenging for companies across industries this year: “Every vertical saw double-digit drops in week one retention from the previous year.” 

If a company acquires new users at a loss—not uncommon for business or consumer apps that spend heavily on ads—and those users don’t evolve into paying customers, it’s tough for the company to make money. With churn analysis, the team there can quantify the value of customers, the price at which it makes sense to acquire them, and develop ideas to increase retention and customers’ lifetime value. 

Using a customer churn analytics tool has many advantages:

Churn analytics software helps product and growth teams track churn and understand why it occurs, and form part of the wider company tech stack. Most churn analytics platforms track individual user events to reveal the user’s journey—the steps users took before they quit—and allow teams to compare this behavior with that of retained customers to reveal what went wrong. 

There are three types of churn that companies can measure.

Subscription churn: Companies that charge a recurring fee define churn as the point at which a customer cancels or suspends the subscription. For example, when a customer calls to end their subscription to a news site, SaaS product, or fitness app.  

The formula to measure subscription churn is:

Non-subscription churn: Non-subscription churn is trickier to measure than subscription churn. Ecommerce companies and ad-supported news sites rely on a steady stream of recurring visitors and define churn as the percentage of those visitors who don’t return for a period of weeks or months. In the non-subscription model, there’s lots of what’s known as rotational churn, where visitors who were considered churned return and no longer count as churned. 

The formula to measure non-subscription churn is:

Revenue churn: Measures the percentage or dollar value of revenue that’s churned over a defined period. This is a common measure for SaaS companies.  

You can measure revenue churn with the following formula:

Churn analysis is useful to any business with many customers, or to businesses with few, high-value customers. Which is to say, nearly every company. Companies in different industries use customer churn analytics for a variety of reasons:

Each industry will have its own use cases. But as you can see from this list, churn analytics delivers useful information in pretty much every vertical. 

Churn isn’t always straightforward to calculate, especially when it’s measured based on past data. The future may resemble the past, but nothing is certain. Unforeseen events, from the emergence of new competitors to black swan market fluctuations, can prove old models wrong and lead companies to take the wrong actions. It’s also difficult for teams to apply the findings of retention analytics to individuals. 

While the law of large numbers may prove churn statistics correct for an entire population, what does it mean when an individual presents a 20 percent chance of churn? What actions should the product team take, if any? 

Finally, companies often apply customer churn analytics to datasets that are too limited, such as only reviewing the last touch customers had with the company. This rarely tells the full story. A customer who calls a support number to cancel their subscription isn’t canceling because they called support—they’re calling because they’ve accumulated grievances over many months. The call is merely a symptom. 

Teams that want to get to the bottom of their customers’ churn must view the entire customer journey, plot the low and high points, and determine its true cause. When selecting a churn analytics tool, consider:

Teams can use a churn analytics tool to view the actions users took throughout their lifespan and develop hypotheses for what led them to quit. Those insights help product teams make changes to prevent future churn. 

Follow-up surveys and questionnaires can provide critical details and suggest actions teams could have taken to prevent the churn. Teams can segment their churn data for greater clarity. No two customers quit for precisely the same reason, but classes of users, known as cohorts, often behave similarly. 

For example, users who signed up for a SaaS tool during a particular conference where they heard the CEO speak could have all shared a sense of awe that wore off after a few weeks, at which point they churned en masse. Or a group of political campaigns could have adopted the tool for the duration of an election, then summarily dumped it once the outcome was decided. 

Sometimes unprofitable or uncooperative users churn, and that’s good. 

Segmentation can help teams discover which customers are most valuable and seek to retain them above others. It can also help teams consider whether the users they’d like to keep from churning can actually be retained. In the example of political campaigns, those customers expire after the election no matter what and the churn may be inevitable. 

Teams can also segment churned users by how long they were customers. Are new users churning at higher rates than medium or long-term users? Has something about the service changed to cause it? With data, teams can develop hypotheses that they can test to reduce churn.

Churn modeling, also known as predictive churn analytics, provides teams with a sense of the events that cause churn so that they can develop a model to predict it for segments of users or, ideally, for individuals, based on their demographics and behaviors. 

Churn modeling can explain whether, say, a particular zip code suggests that a user is at a greater risk of churn. Some tools allow teams to score all customers based on their likelihood to churn, known as a churn score, to allow product, marketing, sales, and customer service teams to prioritize their time. Teams can test measures for reducing churn at key points in the customer journey. 

For instance, teams can A/B test:

All the while, teams should collect qualitative data through surveys, chat widgets, session replays, and feedback buttons, and tie that data back to particular individuals or accounts.

Many of the reasons customers leave can be rectified before a user hits “cancel,” but with 97% of users churning silently, understanding their reasons can be challenging. No matter your company size or industry, your churn reduction strategy must begin with churn rate analysis, or churn analytics.

Along with customer surveys and qualitative user research, use a product analytics tool like Mixpanel to pinpoint where users drop off in a specific funnel. You can then use the data you uncover to confirm any hypotheses you may have, and test solutions in your product.

Though the reasons that customers churn will be unique to your product, they can include things like:

Once you know why users churn, begin testing solutions. Make changes, analyze their impact, and keep optimizing until you improve your product, offering, or customer journey.

For example: let’s say that your product is a subscription music streaming service. After conducting a churn analysis with the help of an analytics tool, your team identifies a cohort of subscribers who log in only a few times a month. You determine that, historically, customers who log in fewer than, say, four times per month have a higher chance of canceling their subscriptions or churning.

As a churn reduction strategy, you send that user cohort an email offering a free month of service. Then, you use cohort analysis to see how that group of users responded to outreach efforts. After a month, you’re able to determine that they were less likely to churn after receiving the promo offer. Success!

In this instance, taking the time to understand the “why” behind your customer churn rate provides actionable insights to help you recoup initial customer acquisition costs—and keep hard-earned customers engaged.

Depending on the drivers behind your churn rate, remember that churn reduction strategies can be applied at any point in the customer journey—from how your product is built or architected, to how it’s sold, marketed, and supported.

Looking at patterns of disengagement among already-churned users helps you identify which current users are at the highest risk of abandoning your product. The quickest way to find these patterns is by comparing what actions (or lack of actions) churned users took within your product compared to users who didn’t churn. 

Once you have that information, you can segment users who take those actions into cohorts of “users most likely to churn” and work to retain them proactively with tailored retention campaigns and personalized offers. 

On the flip side, you can also use churn analytics to identify the actions and behaviors of your most engaged users and nudge more users toward those actions. If users are getting value from your product, they’re less likely to leave it. 

Let’s look at a few quick examples of how Mixpanel has helped companies from three different verticals reduce churn.

Deliveroo is a London-based leading online food delivery company operating in 500+ cities worldwide. To ensure its restaurant partners are successful and get a high ROI from using the company’s services, Deliveroo looks closely at the conversion funnel from sign-up to first order and, for each market, sets targets they want partners to hit so that Deliveroo knows its reducing churn and increasing conversion.

Deliveroo used Mixpanel to analyze one- and three-week retention rates to better understand how restaurants were engaging with the platform and if they were finding value in it. Their discoveries led them to improve retention rates and reduce churn. 

Viber is Rakuten’s cross-platform instant messaging and voice over internet protocol (VoIP) application. To understand what makes Viber more fun for its users, the company used Mixpanel to get a holistic understanding of messaging patterns and how they could potentially change the product to move the needle of their most important business drivers—increasing engagement and improving retention.

SaaS-based education company codeSpark teaches young children how to code by turning programming into play. The company needed to understand how their games and lessons were contributing to user engagement and why users were churning. They also needed more insights into which free users became paying customers (and why). 

They used Mixpanel to understand their users and determine which changes would have the most impact on user engagement. With Retention reports, they were able to identify successful cohorts and build an Ideal Student Profile. By targeting those cohorts, they were able to increase subscriber retention. 

Turkish sports betting company Bilyoner has to contend with strict gambling regulations and centrally-managed odds servings and payouts. This means that the company can’t offer better odds than the competition—their only way to differentiate themselves is through customer experience. 

The company started using Mixpanel for churn analytics to better understand how customers use its app, prioritize its product roadmap, and quickly identify and fix bugs that cause churn and prevent users from placing bets. Thanks to these insights, they were able to achieve a 94% monthly retention rate. 

These are just a few examples, of course, but product analytics can be used across industries to gain a competitive advantage by gaining the deepest possible insights into what makes your customers stick or stay. 

Preventing churn is the responsibility of every team within the company. It’s up to the marketing team to properly educate new buyers, the sales team to not overset expectations, the customer success team to provide high-quality support, and the product team to build a service that continues to delight customers. Companies can improve their churn rates by sharing churn data throughout the company. The more insights each team has into how it can reduce churn, the more customers the whole company keeps.

Remember that not all churn is bad and that there’s always a chance to go back, change things, and try again. The ultimate goal is a slow but steady reduction in churn rate, and the reasons for churn will always be unique to your product. Understanding the “why” is the key, and product analytics is a powerful resource for that piece of the puzzle.