CLV (Customer Lifetime Value)
CLV is not a synonym for historical LTV. It's its predictive version, calculated customer by customer to decide where to invest before the value even materializes.
“An average LTV tells you what your typical customer is worth. A predictive CLV tells you what every single customer you have today is worth.”
Otto GTM ObservatoryFrom average value to predictive per-account value
While LTV is historical and aggregate, CLV uses real-time behavioral signals (usage frequency, feature adoption, support tickets, sentiment) to estimate the expected future value of each active account, allowing intervention before a high-risk customer churns.
This approach transforms customer success from a reactive function into a predictive one: accounts with declining CLV are automatically flagged for targeted intervention, well before the signal becomes visible in a quarterly churn report.
Anti-patterns
Common mistake: using CLV and LTV as interchangeable synonyms, losing the distinction between an aggregate historical figure and a predictive model at the individual customer level useful for immediate operational decisions.
Second mistake: building a predictive CLV model without updating it with fresh behavioral data, effectively turning it into a renamed historical LTV, lacking the real predictive power that justifies it.
Practical Application
An account with €24,000/year ARPA and positive behavioral signals (weekly usage, 3 advanced features adopted) receives a predictive CLV of €156,000 over a 5-year horizon. An account with the same ARPA but usage down 40% in the last 90 days and doubled support tickets receives a predictive CLV of only €41,000: the same nominal contract value, a completely different churn risk.