Charles Rosenblatt, CEO of Butter Payments, explores how machine learning (ML) and retention data can help optimise every stage of the customer journey to maximise value while keeping customers pleased.
When customers rein in their spending or experience dips in disposable income, subscriptions are often the first expenses to go. Regardless of what’s happening in the economy, the best way to retain your customers is to ensure your product remains valuable. You can’t control whether they’ll continue to have the money for the ongoing expense, but the decision to offer value is always in your court.
That’s why driving growth is no longer solely marketing’s domain. Today, it’s a strategic priority that directly impacts revenue, lifetime value (LTV), and the overall health of your business.
Make retention a data-driven competitive advantage
Along the customer journey, there tends to be a far greater application of resources and strategy at the acquisition stage than throughout retention. This can be a costly misstep if enough customers leave before you can see a return on your investment. That’s where savvier account management and retention strategies play pivotal roles.
You’re left with two choices: hire internally to study your limited retention data or get outside technological help and access to a broader scope of data. ML models can spot patterns and trends across billions of anonymous industry and customer data points. While your teams can use observation, feedback, and common sense to identify reasons you might be losing customers, ML can uncover other insights people might miss.
Operationalise ML for revenue impact
ML insights can predict whether a customer will be able to make their upcoming subscription payments, whether a customer might be engaging in fraudulent behavior, or which customers are most likely to return the greatest LTV. You can even use ML insights to determine what type of customer is leaving (e.g., age range, use case, or acquisition source), to then cross-reference that information with how you’ve communicated with those customers in the past. Collectively, this information can paint a clearer picture of why people are leaving; without it, most retention strategies would be nothing more than a shot in the dark.
For example, imagine a subscription-based company is experiencing 30% attrition. Their product team is worried. They work on substantial updates to their product or even replace it entirely. The company invests considerable time and money in solving the issue, only to later realise it was a technical payment leakage issue unrelated to their product.
A mistake like this could cost millions. It also speaks to two pitfalls associated with going the manual data analysis route: working with a limited sample set and leading with a firm hypothesis that won’t let the data speak without bias.
You can’t control the market – but you can control retention
When businesses have access to the right data – and the ability to analyse it objectively – ML helps separate real customer churn from avoidable issues like payment failures or engagement gaps. That clarity allows teams to focus their time, capital, and energy on changes that actually move the needle, rather than reacting to assumptions or incomplete signals. Instead of guessing, businesses gain a clearer understanding of what keeps customers loyal and engaged – and paying.
You may not be able to control market volatility, but you can control how well you understand and respond to your customers. Using retention data intelligently isn’t just a defensive tactic – it’s a strategic growth lever that delivers value long after the market stabilises.
Subscribe!
Our editor carefully curates two InternetRetailing newsletters a week filled with up-to-date news, analysis and research. In addition to this, there is a dedicated newsletter focusing on the subscription economy every second Wednesday – click here to subscribe for free. And why not follow us on LinkedIn to receive the latest updates on our research and analysis?
