How to Measure Recontact Rate
Recontact rate is the percentage of customers who come back about the same issue within a set window, most commonly 72 hours, after a ticket was marked resolved. It is the lie detector for false resolution: a ticket the customer returns to within three days was not actually solved, regardless of what the close status said. Measured well, recontact rate is the truest check on whether your resolution numbers are real.
How to calculate it
Take the tickets marked resolved in a period, and count how many of those same customers contacted again about the same issue within the window. Divide recontacts by resolved tickets to get the rate. The two design choices that matter are the window length (72 hours is a common, defensible default) and same-issue matching, so a new, unrelated question does not count against a genuine resolution.
Why the window matters
A resolution metric measured at close time rewards aggressive closing. Adding a recontact window moves the measurement to after the fact, where premature and wrong closes get caught. Segmenting recontact rate to AI-handled tickets tells you specifically whether the automation is durable, which a blended number would hide. This is the same signal reopen rate captures, viewed from the customer-return side.
How Resolve uses it
Resolve counts a ticket as verified resolved only when the customer confirms the fix or does not recontact within the window, and it treats the recontact and reopen signal as the north-star honesty metric shown beside auto-resolution. When recontacts cluster on a topic, that surfaces as a knowledge base gap to fix rather than a number to bury, so the metric drives improvement instead of just grading.