In this small October example, start with six tickets and five on-time replies. After joining the approved targets, you should still have six rows, with each ticket ID appearing once. A duplicate October target for Alder would produce nine rows.
Check the six ticket records
October 2026 · one row per ticket| Ticket | Customer ID and name | First reply |
|---|
| T01 | C01 · Alder | On time |
|---|
| T02 | C01 · Alder | On time |
|---|
| T03 | C01 · Alder | Late |
|---|
| T04 | C02 · Birch | On time |
|---|
| T05 | C02 · Birch | On time |
|---|
| T06 | C03 · Cedar | On time |
|---|
Alder has two on-time replies out of three: 66.7%. Birch has two out of two: 100%. Cedar has one out of one: 100%, with no target to compare it against. Together, that’s five out of six, or 83.3%. Calculate from the ticket counts; averaging the three customer percentages would give a different answer.