Back to Parative
Customer churn is a natural part of the SaaS sales cycle. And because retaining customers is less costly than finding new ones, businesses need to find ways to manage and reduce customer churn. One of the best ways to do that is to identify the factors that lead customers down the path to non-renewal.
In this post, we’ll look at how to predict customer churn and strategies for preventing it. Then we’ll share insights from SaaS leaders about their experience with predicting customer churn.
Customer churn is calculated as a percentage — it’s the number of customers lost during a specific period, divided by the number of customers at the beginning of that period. So, if you had 1,000 subscribers at the beginning of Q1 and lost 30 of them in that quarter, your churn rate would be 3%.
Churn prediction is determining which customers are likely to churn, based on historical data and their usage of the software. For example, let’s say that based on your CRM data, you know that customers you’ve lost in the past were those who rarely used several key elements of your software. That tells you that any current customers who aren’t fully using your software might decide to not renew their subscription.
Sometimes, churn may be a result of factors beyond your customer’s control, such as the hiring of a new manager with different priorities, or an unforeseen budget cut. Generally, though, these are the three most common reasons customers churn:
This is a common theme — that the cost of the software outweighs its benefits. This perception may arise when a SaaS product is useful for a certain team, but not an entire organization. Or it may stem from a lack of reporting features that could prove the software’s value to management.
Busy teams may not have much time for onboarding, and if your company doesn’t offer assistance during implementation, your customers might not be able to fully engage with your product.
Low adoption may also occur when your customers’ priorities change. For example, they might’ve purchased a software solution for managing content production but stopped using it when they began outsourcing content.
Businesses buy software to make their jobs easier, but if a SaaS product doesn’t integrate with the tools they already use, their workflows can become much more complicated.
Even though most communication between SaaS providers and their customers is digital, customers still want the ability to speak to a human being, should they need assistance. A dedicated customer success representative or a support desk that offers live chat could help improve customer satisfaction and reduce churn.
So, data can help you understand and predict customer churn, but what metrics should you be reviewing? These are the key metrics to watch:
To know when a customer’s usage of your product is below average, you need to establish a usage baseline that incorporates the following metrics
“Activation” means that your customer has achieved the desired outcome using your product — for example, sunsetting an old platform and using yours to manage all digital marketing assets.
Activation Rate = Activated Users / Acquired Users
Your feature adoption rate can be calculated in several ways. You might want to look at what percentage of your customers use all of your product’s primary features so you can identify any correlations between feature usage and churn. Or, you could look at which features customers tend to use when they first acquire your product and see if usage of those features declines over time.
Your retention rate is the inverse of your churn rate — so, if your churn rate is 5%, your retention rate is 95%. Review this metric regularly to see if it changes over time.
To get an accurate picture of your churn rate, you’ll need to collect data for two months. Here’s the baseline churn formula:
(# of Customers on Month 1, Day 1 + # of Churned Customers on Month 2, Day 1) / Total Customers in Month 1
Review your customer data and look for trends. What do your former customers have in common? Were churn rates higher for customers with the lowest service tiers? Did business size have any impact on customer churn? Answering these questions will help you identify your current customers that are a churn risk.
Also be aware of telltale signs that a customer is about to churn. For example, you may notice that your former customers exported data and moved other files out of your platform before they churned.
There are a few ways to prevent customer churn, and they all involve increasing the frequency and quality of communication with your customers.
Regular interaction with your company can help build customer loyalty. A standing monthly meeting — even if that’s a Zoom call — gives customers an opportunity to ask questions, request new features, and tell you how they’re feeling about your product.
A feature request is a form of feedback — it means your software is missing some functionality that a customer wants. Ideally, you should segment your customers into groups, based on their feature requests, acknowledge their requests, and provide frequent status updates. When and if you do roll out the requested feature, the users who requested it should be the first to know.
Micro-surveys help you collect important, ongoing feedback without encroaching on your customers’ time. These are the most common micro-surveys:
Now, let’s see what our SaaS experts told us about predicting customer churn!
Split Dragon | Fractional CMO
Aside from setting up automated trigger-based email campaigns to nudge inactive users, we go the extra mile. Our Customer Success Team monitors app usage patterns of our users (especially the VIP ones). We segment them upon sign up and the VIP list gets extra attention from the team. We then engage with them and throw in a complementary account set up to facilitate product adoption. This way, we identify churn risks before they happen.
SPOTIO | Head of Customer Experience
Monitoring customer activity metrics over time has allowed us to understand each customers adoption baseline. We have implemented triggers that alert the CSM when a customer is showing significant drops in activity. This allows us to connect with our customers to course correct whatever they are experiencing & ward off churn proactively.
Qualaroo | Product Marketing Manager
Customer churn has been a major area of concern, and we have increasingly diverted our attention to this issue. Gathering first-hand feedback from the customers was our primary objective. We have been conducting Net Promoter Score (NPS) surveys on our website’s landing and product pages. By doing so, we can gauge the overall customer satisfaction in real-time and determine how likely the customers are to recommend our brand and products to their friends, family, and colleagues. You can use tools like Qualaroo to create NPS surveys & measure customer satisfaction.
The NPS rating also helps us determine the overall brand perception and any drop in the ratings allows us to forecast the number of customers on the verge of churning.
By surveying our website visitors, we are able to uncover elements that are performing well on our website along with areas that might be driving the customers away. The AI-powered Sentiment Analysis engine also helped us go deeper into analyzing the overall customer sentiment and identify the number of users who are most likely to churn.
Jitbit | Founder
We monitor average feature usage, then cross-reference that with features that are statistically most used by paying customers. Then we reach out with semi-automated "have you tried this?" emails.
Backblaze | Senior Director of Marketing
We do in-depth analysis of who has already churned and build models around when in their journey a costumer is likely to churn (broken out by SKU) then we run preemptive programs to keep them engaged.
Dryrun | CEO
Of course, we track usage patterns but the key for us is to take action proactively based on the data. Users are graded on a scale of success, through a set of criteria we've established, so that we are able to quickly identify those at highest risk of churning. The number of communication channels, frequency and messaging are all based on their position on the scale. Those at highest risk are elevated to direct intervention by our success team.
Callbell | Growth and Marketing
We generally implement a form for our users to fill out if they want to leave the service.
Another very important thing that we do is that our support team analyzes customers who do not open the application for a certain period of time and contact them to ask why they do not use the service frequently in order to recover customers at risk of abandonment. .
JSL Marketing & Web Design | VP of Sales & Marketing
One way that our team works to identify customers who may be at risk of churning is by having a monthly consultation that goes along with the monthly report our team sends out. This consultation is always done over Google Hangouts so that we can also talk to the client face-to-face. This allows us to gauge their response to the most recent numbers from the report and also gives us the ability to explain the report further if there is any confusion or misunderstanding.
