Do your subscriber churn numbers keep you up at night? You are not alone. Most SaaS businesses overinvest time and resources acquiring new customers, but leave customer retention to chance — a costly mistake, since acquisition is at least 5 times costlier than retention (even more, depending on which study you look at).

In the beginning, you might not notice how detrimental churn can be, because your customer acquisition is likely far outpacing your churn. But eventually, as acquisition stabilizes, churn starts dragging revenue down to the point where you can no longer ignore it.

In the competitive app-eat-app world of SaaS, managing churn is a constant, ongoing endeavor. In this blog post, we look at how to analyze and manage churn better, particularly through the lens of data.

Managing Customer Churn

Churn management is two-pronged: reactive and proactive. The smart thing to do is to prioritize proactive churn analysis. It is more cost-effective, more customer-centric, better for your brand, and much less hair-raising overall.

Jonah Lopin, HubSpot’s Vice President of Services, sums it up in this HBR piece: “By the time you see an increase in your churn rate, it is six or eight months after the point in time when you actually failed the customer. If churn is your only measure of customer happiness, then you’re always six months too late to influence your future.”

However, churn is inevitable, no matter how good your product is, and what you do to retain your customers. Approximately 50% of customers experience natural churn every 5 years. So it is important to have a remedial plan to fall back on, when your churn rate crosses a certain threshold. 

Plan A: Proactive Churn Management

Like we said, the best way to manage churn is before it happens. Here is where you can unleash the true power of data — by using a churn prediction model to recognize at-risk customers and proactively retain them. 

Many organizations have long been doing this. Salesforce’s Einstein Analytics uses predictive AI to help companies identify customers who might churn, and provide actionable insights for Customer Service Managers. Or look at Hubspot. Jill Avery, a senior lecturer at Harvard Business School and an author of Go To Market Tools, says: “When the economy crashed in 2008 and the company’s churn rate shot up, HubSpot delved deep into its churn data to see what it could find out about which customers were more likely to leave and when. Using that analysis, the firm targeted customers they suspected might cancel and offered services, like extra training on particular features, to convince them to stay.”

‍

Broadly speaking, this is what proactive churn management looks like.

1. Know your current churn rate. 

Depending on your subscription plan, you might want to track monthly as well as annual churn rates. As a general rule of thumb, the average monthly churn rate for small to medium businesses would be 3-7% and for larger businesses, around 1-2%. If you’re a small, fast-growing company, aim to keep this under 5%. 

It’s also a good idea to look at revenue churn side-by-side with customer churn. For instance, the churn among high-value customers can be prioritized over that among other customers.

2. Gather all relevant data.

Here are some numbers that could serve as inputs to your churn prediction model. Once you decide what your input variables are, establish a process to continuously monitor these numbers.

Usage Metrics: How customers are using your product can help identify early signs of dissatisfaction or disinterest. For software applications, this could be login frequency, feature usage, or session duration. 

Transaction metrics: These are data points such as purchase history, payment methods, transaction frequency etc. Analyzing this data can unveil spending patterns and shine the light on future churn triggers.

Customer Feedback and NPS (Net Promoter Scores): Customer feedback from reviews and support interactions can help identify issues or areas for improvement. NPS measures customer satisfaction and loyalty by asking customers how likely they are to recommend your product to others. A waning NPS score could possibly be an early indicator of churn. 

Other tailored early warning Indicators: Establishing proactive thresholds or triggers based on key metrics (e.g., usage frequency, in-app activity, payment failures) helps flag customers who may be at risk of churn. 

3. Build a churn prediction model.

With predictive modeling, all of this data can be converted into a 360-degree view of a company’s churn rate. A churn prediction model typically uses supervised ML to segment customers into various risk categories, based on weightings of input variables (the data points we discussed above).

Some AI systems are also capable of offering suggestions for the “next best action” (NBA)—what the business should do to woo back at-risk customers.

4. Assess churn risk scores and segment customers.

Let’s say your predictive model assigns a churn risk score between 1-100 to each customer. You could then segment your customers into risk groups. For example:

1. High Churn Risk: 76-100
2. Medium Churn Risk: 51-75
3. Low Churn Risk: 0-50

Note: These numbers are indicative and vary by industry and sector.

While churn risk groups are the easiest way to segment your customers, you can also slice and dice in more focused, finely-tuned ways. For instance, you can group customers likely to churn according to behavioral traits, such as infrequent product usage or negative customer service interactions. Your retention strategies can then be targeted especially to the pain points faced by these customers.

Another kind of segmentation could be based on customer value. Some groups of customers are high value, purchasing your products and services repeatedly. You do not want to lose these customers, and your retention strategies for this segment ought to be more time and resource-intensive. This tiered approach can help you make the best use of your team’s limited bandwidth.

Ask Brian Dean, CEO of SaaS business Exploding Topics. When the churn rate of the business went up over 10%, the team made a list of their top ten customers - the ones on their highest plan, the ones who had stayed the longest. They then dove headlong into researching these customers’ needs and gave them features designed especially for them! The churn rate, he says, dropped to 2.7%. Is this always scalable? No. But is it impossible? Also, no.

5. Implement proactive retention strategies.

Once you’ve got a thorough understanding of your churn risk, here are some strategies you can employ to win back your customers.

• Targeted Communications: Tailor reengagement campaigns to  customers who haven’t interacted with the brand recently.
• Subscription Renewal Reminders: Even something as simple as an automated reminder about renewal can increase the likelihood of your customer continuing with you.
• Personalized offers: You can create customized offers tailored to woo back individual customers, and time them strategically to re-engage them.
• Product Enhancements: Analyzing usage data, customer feedback, and reviews gives you a wealth of information about how to improve your product features. Use this data to make your product sticky, to effectively make it so tailored to your customers’ needs that it’s hard to replace.
• Loyalty Programs: Analytics can help you pinpoint high-value customers and their preferences. You can create loyalty programs that reward frequent patrons and incentivize them to stay with you for the long haul.
  

Plan B: Reactive Churn Management

Perhaps you were busy putting out some other fire while your churn rate slowly crept up. Perhaps you have implemented a few proactive strategies to manage churn, but you’re facing a problematic churn rate anyway. Here are some measures you might find useful in pulling your business out of troubled waters. Some of them are data-centric; others are more behavioral and process-related.

1. Analyze churn when it happens.

This might seem kind of obvious. What we mean is, staring down the barrel at a double-digit churn rate is uncomfortable—but it is also a golden opportunity for you to figure out what customers really need out of your product.

When are customers most frequently churning? Is it 30, 60, 90 or more days after they subscribe? Does churn happen if customers don’t use the product or service for ‘x’ number of days? And so on.

If it is not apparent why they churned (as is common), ask them.‍

2. Seek extensive feedback from customers.

Reach out to at-risk or lost customers for feedback. Make your feedback system more robust by requesting your customers to fill out an exit survey, and then design your survey such that they’ll actually do so. Keep it short and to-the-point. Include an open-ended question for them to tell you what didn’t work for them (or to vent, if they need to). 

Analyze this feedback (using text analytics and NLP to extract insights from voluminous data) to figure out specific pain points driving customer attrition. You can then address these issues to prevent future churn. 

3. Lean into your most valuable customers. 

We spoke about investing in high-value customers proactively to reduce the likelihood of them churning. The same applies when they are on the brink of canceling their subscription. 

Sunil Gupta, the Edward W. Carter Professor of Business Administration at Harvard Business School, recommends that businesses focus their attention on the most profitable customers on the brink of churning—instead of attending to every single customer who is likely to leave. 

He says, “If I offer an incentive to customers most likely to churn, they may not leave the company, but will it be profitable for me? The traditional method is focused on reducing churn, but we contend the goal should be maximizing profits, rather than only reducing churn.”

In the end, it comes down to this. Churn is a fact of life at a Saas business. It can merely be a bunch of harmless breakups that don’t impact your bottom line. Or, well, it can make your stomach churn. The difference is in how thoroughly you analyze it, both proactively and reactively. And what you make of the insights so garnered.

Churn analysis is well worth the effort that goes into it. In fact, a landmark study by Bain & Company, along with Earl Sasser of the Harvard Business School, has shown that even a 5 percent increase in customer retention can lead to an increase in profits of between 25 and 95 percent.

What measures do you take to manage churn? Has it helped you stay competitive in your arena? We’d love to hear from you.