What changed
Your save rate now reflects every customer session where a cancellation was averted, including customers who opened the survey and walked away without canceling. This gives you a fuller, more accurate picture of your cancel survey's impact.
How save rate used to work
The previous save rate was powered by Retention Engine, a machine learning model focused on optimizing save offer treatments. Under that model, a "save" was counted only when a customer was shown a save offer and actively accepted it, such as clicking to take a discount or swap their product.
This was a reasonable measure for the time. The prevailing thinking in subscription retention was that offers — discounts, swaps, pauses — were the primary lever for preventing churn. The model was designed to get more people to accept those offers, and the save rate reflected exactly that.
How save rate works now
The new cancel survey calculates save rate at the "session" level. Any time a subscriber clicks the cancel button and does not complete their cancellation, whether they accepted an offer, read an educational card, or simply closed the survey, that cancellation attempt or "session" is counted as a save.
Does your save rate look higher?
If you're seeing a higher save rate after switching to the new cancel survey, that's expected, and it's a good sign. Your survey was already saving more customers than Retention Engine was measuring. Now those wins are reflected in your metrics.
Why abandons count as saves
The way merchants think about cancel surveys has evolved significantly. Early retention tooling was built around a simple idea: present an offer, get an acceptance. But merchant and subscriber behavior has shifted.
Today, many of the most effective cancel survey experiences lean on education; explaining value, surfacing product details, addressing common hesitations. Customers who read that content and close the survey without canceling have been retained. Their intent to cancel was interrupted, and that's exactly what the survey is designed to do.
Counting only offer acceptances was leaving a significant portion of successful saves unmeasured. The new save rate closes that gap.
FAQs
What if a customer abandons the survey and comes back to cancel later?
Each cancellation attempt is evaluated independently. If a customer later returns and completes their cancellation, that subsequent attempt is recorded separately as a churn event.
Should I redesign my survey now that abandons count?
This is a great opportunity to think about your survey holistically. Educational content, clear value communication, and well-structured flows can now contribute directly to your save rate.
Can I still see offer-specific performance?
Yes, and even better than before. Your in-flow cancel survey analytics includes a breakdown by cancellation reason, treatment type, and outcome, so you can still evaluate offer performance independently.