Churn Analysis: Unpacking Which Customers Are Leaving

One of the most important — and potentially daunting — metrics for SaaS businesses is churn. Customer churn, otherwise known as customer attrition, is a key performance indicator (KPI) that measures the number of customers who cancel their subscription with your company.

Though the definition of churn seems simple enough, analyzing why your churn is what it is — and fixing it — is more complex. For the average SaaS business, the benchmark for churn rate is around 5% and a good churn rate is considered 3% or lower, according to ProfitWell. Even a small improvement can supercharge your retention and revenue, but first you’ve got to know where to start.

In this blog, we’re going to help you navigate performing a churn analysis for your company and identify a churn issue before it gets out of hand.

What Is Churn Analysis and Why Is It So Important?

Customer churn analysis is a way of understanding why and when customers are leaving your product. More specifically, it refers to the process of reviewing your company’s churn data to identify commonalities or patterns among customers that are churning.

For many SaaS startups, churn can be a silent killer. If you have a product that isn’t working smoothly yet, or you haven’t found the correct product-market fit, you’ll know right away with the lack of new customers or sign-ups. With churn, however, it could take months before you notice that something is wrong — if you’re not actively monitoring it. Here’s how a churn analysis lets you turn things around:

It Helps You Understand Who Is Churning and Target Them Effectively

There are numerous reasons why customers may choose to leave your SaaS company or switch to a competitor. Here are some of the most common causes of churn:

• The customer wasn’t properly onboarded
• The product isn’t meeting or exceeding the customer’s expectations
• The customer fails to see the long-term value proposition of your product
• The customer’s needs have changed
• There’s poor customer service or low response rates
• The customer has budgeting/financial issues
• A new stakeholder wants to switch products/services

Even from this short list, it’s apparent that each customer will have their own unique reason for leaving your product. A one-size-fits-all approach won’t work here.

A churn analysis allows you to dive deep into these causes and segment your audiences accordingly. You’ll understand which types of buyers are most likely to churn, and you can then target them with the right retention and loyalty strategies.

In other words, this analysis lets you go beyond your customer churn rate and avoid the trap of simply applying the same solution(s) to all your users. For example, if your company has churned customers due to a poor pricing strategy, you won’t be able to adopt the same strategy as a team that’s losing customers due to a lackluster or confusing onboarding process.

Recommended reading: A Guide to High-Touch B2B SaaS Onboarding

It Tells You Where To Make Changes in the Customer Lifecycle

Based on the timeframes of when customers are churning, you can start to identify whether you have a sales problem, an onboarding problem, a customer success problem, or something else entirely.

Here are two quick examples:

1. If your customers are churning in the first 60 days, it’s likely an onboarding problem since it’s relatively early in the journey. We recommend tracking and analyzing specific onboarding metrics to measure the effectiveness of your onboarding program and optimize your overall customer experience. For instance, maybe you need to switch from low-touch onboarding to high-touch onboarding.
2. If customers are churning at 11 months or near the common renewal or upgrade phase, you might have a customer success problem in which customers aren’t experiencing enough value by the time they need to recommit to your SaaS tool. That’s why your team needs to focus on reducing every customer’s time to value (TTV).

It Gives Your Team a Set of Data To Monitor Progress

Many SaaS teams are already great about collecting data, so using it to build a powerful churn analysis is a no-brainer. As you discover where changes need to be made and implement actions to curb your churn, you’ll have a set of metrics and KPIs to continuously reference. This will help your team stay on track and adjust your strategy as needed.

Although analyzing churn is the first step to reducing it, the truth is that churn is extremely hard to prevent altogether. According to Toward Data Science, the reason for this is simple:

“In order to prevent churn in a long term and reliable way, a service must actually improve either the benefit delivered by the service or reduce the cost incurred from using the service.”

There’s no silver bullet to reduce churn, and that’s why homing in on a solid churn analysis process — and sticking with it — is so important.

It Lets You Identify Lost Revenue and What You’re Leaving on the Table

You’ve probably heard before that customer acquisition costs are always higher than the cost of retaining your existing customers.

So, it’s no surprise that another huge advantage of performing a churn analysis is that you’ll learn how much money you’re losing at any given point in the buyer’s journey — and concrete opportunities to get that money back. And as you lower your churn rate, your customer lifetime value (LTV) will increase, contributing to higher profits.

In addition, analyzing the hard numbers will allow you to better understand whether tools and services to prevent churn or optimize your customer journey (like Arrows for better onboarding) are worth the cost.

When Is the Right Time To Analyze Customer Churn? (And How Often Should You Do It?)

Less is more! In most cases, it isn’t necessary to analyze churn every week or even every month. Depending on the thoroughness of your analysis, it could be performed every six months to a year.

The reality is that a churn analysis done the right way takes a lot of work and a lot of data, especially for a small or growing team. (One way to cut down on the required time and effort would be to pull in data science professionals to assist you.)

If you’re in a position where you have localized teams, they could perform smaller churn analyses more frequently to understand their own effectiveness in their parts of the customer journey.

If your customer churn is happening more frequently (for example, if the average customer churns after three months), it might be necessary to perform a monthly churn analysis.

All that said, you should be regularly checking your churn rate to maintain a high-level view and notice any large spikes in either direction. This is especially true if you’re fundraising or working with venture capital firms.

Churn rate is a major variable in how investors perceive your business. In fact, venture capital firms consider customer churn one of the most critical SaaS metrics for raising capital, according to Jeron Paul, CEO and founder of Capshare.

A High-Level Overview of How To Do a Churn Analysis

Now comes the interesting part: Performing the churn analysis. These four steps will make the process manageable and organized, helping you draw out key insights about your customer churn without getting lost in the weeds.

Step 1: Map Out Your Customer Journey in Epic Detail

The first and most important step in your churn analysis is mapping out your customer journey in as much detail as possible. You should be able to see all of the milestones and steps a customer should reach, the interactions they should have at each stage based on their segment, and the value they should receive at each of the customer lifecycle stages:

1. Marketing and Lead Generation
2. Selling and Purchasing Process
3. Customer Onboarding
4. Receiving Value and Product Adoption
5. Achieving Retention
6. Further Expansion

Check out our guide to creating a customer journey map to see how to navigate this process, step by step.

Step 2: Collect Customer Journey Data for a Year (if Possible)

After mapping out your customer journey, you’ll want to gather as many datasets from the journey as you can for a year or as long as possible. This is where it can get a bit confusing if you’re not utilizing analytics tools.

We highly recommend you find an analytics tool that works for your company to ensure you have valid ways of measuring statistical significance, understanding what to measure from each part of the process and which teams influence which data points, segmenting customers, etc. Once your data is all in one place, you can modify the data to extract meaningful actionable insights as you see fit.

Step 3: Review Your Churn Data To Get Insights

Next, identify when customers churn (the number of months, days, weeks, years after sign-up) and which stage is losing a significant number of customers. This includes determining which steps and/or actions they successfully took and which steps they didn’t.

You should draw out insights fairly quickly. For instance, maybe the customers who logged in at least eight times in two weeks were more likely to stick around than those who only interacted with your product three times in two weeks.

Once you’ve done this, then you can start to plan and deploy strategies to curb customer churn. Fighting churn is a joint effort among numerous team members — it isn’t something that can solely be done by AI, machine learning, or data science, though each can help.

Think about tapping into your product managers, software engineers, content creators, marketers, customer support representatives, and account managers. Each can play a vital role in reducing churn.

Step 4: Monitor Analytics by Customer Segment

It’s important to understand the patterns among “like” or similar customer segments and cohorts. This allows you to make more accurate conclusions when customers in the same segment behaved differently.

For example, maybe you discover that the customers were onboarded in different ways. Maybe one customer received guidance and proper onboarding and the other attempted to entirely self-onboard without going through the typical program.

This process of customer segment is also referred to as creating churn profiles. A churn profile consists of common characteristics in your customer base, grouping customers by those common characteristics, patterns, and churn rates for each group.

One way you can organize your customer base is by the following:

• Financial similarities: Try to group by Monthly Recurring Revenue (MRR) or Annual Recurring Revenue (ARR) and then organize them into tiers based on those metrics.
• Geographic/demographic similarities: Customers who are in similar geographies may be driven by certain market or industry forces and follow some of the same patterns.
• Your customer’s evolutionary similarities: As your customers evolve, they will likely face common hurdles regardless of industry. For example, startups who recently had a Series A funding round may quickly look for a wider feature set or technical support and churn accordingly. The same may be true for those companies who have just raised Series B, C, D, and so on.
• Your own evolutionary similarities: Growing and progressing your business will inevitably include both good and not-so-good phases. When things are in a “not-so-good” phase, you may notice more customers who signed up during that time ultimately churned. Understanding when customers came on board with your company, and which phase it was, can be more helpful than some financial, geographic, or demographic attributes.

Are customers churning during onboarding? Arrows can help.

After conducting a thorough customer churn analysis, you’ll have a clearer picture of where in your customer lifecycle your subscribers are dropping off. If you’re realizing that the issues are arising from your onboarding process, Arrows can help.

Arrows’ onboarding platform helps you set your customers up for success, allowing your team to easily create personalized onboarding plans by individual or cohort, optimize your onboarding process, and stop churn at the source.

Ready to see what Arrows can do for your team? Schedule a demo.

‍