Field notes
Cancellation codes that help ops teams, not just reports
Cancellation rates alone rarely tell an ops lead what to fix. “Customer cancelled” might mean the quoted wait felt too long, the price changed after estimate, or the arrival window was wrong. If your on-demand app collapses those reasons into one label, service request analytics will keep pointing at the wrong stage of the journey.
A practical code list is short enough that dispatchers will use it under pressure. We usually recommend separating wait-related cancellations, price or scope disagreements, no-show or wrong-address issues, and operator-initiated declines. Each code should map to a timestamp you already store—acceptance, assignment, or first customer message—so you can see which stage precedes the cancel.
Once codes are in place for a few weeks, compare cancellation share inside peak windows versus quiet hours. Wait-related cancels that cluster in known thin-roster windows are a capacity story. Price disagreements that sit evenly across the day are a quoting story. Mixing them into one percentage hides both.
Cleaning codes will not by itself raise fulfilment quality, but it makes a Fulfillment Bottleneck Brief far more honest about which handoff deserves attention first.