Support metrics glossary
Resolution rate. The first number on every support dashboard.
Resolution rate is the share of incoming tickets your team resolves in a period, and the easiest metric in support to inflate without answering a single extra customer. Here is the formula, the worked math, and the fine print.
Last updated 14 August 2026
What resolution rate means
Resolution rate is tickets resolved divided by tickets received in the same period, times 100. If 1,000 tickets arrive in a week and the team resolves 920 of them, the rate for that week is 92%. Nothing more complicated than that.
Two definitions decide everything, and most teams never write them down. What counts as resolved: an agent pressing solved, a timer closing a silent ticket, or the customer confirming the fix. And what period you measure. Change either one and the same team prints a different number.
The metric that looks honest and lies politely
Here is the flat version. It is the most honest-looking liar in support metrics. A problem fixed on the first reply and a ticket a timer closed after three days of silence both move the numerator by one. Only one of them was service.
Counting closed tickets as resolved is like counting dishes out of the sink without checking whether they are clean. If the helpdesk auto-closes stale tickets, or agents mark solved to keep the queue tidy, the number measures closing discipline. Not support.
The arithmetic, worked through
The formula is simple. The backlog is what makes it interesting. One team, four weeks, every number below arithmetic rather than benchmark.
| Week | Tickets received | Tickets resolved | Resolution rate | Backlog carried out |
|---|---|---|---|---|
| Week 1 | 1,000 | 920 | 92.0% | 80 |
| Week 2 | 1,050 | 1,000 | 95.2% | 130 |
| Week 3 | 980 | 900 | 91.8% | 210 |
| Week 4 | 1,100 | 1,160 | 105.5% | 150 |
Week 4 is the one to stare at. The team resolved 1,160 tickets while only 1,100 arrived, because it worked through 60 old tickets on top, so the week prints 105.5%. That looks impossible until you remember what the formula counts: work finished in the period, divided by work that arrived in it. Nothing says the work has to be from the same period.
Weeks 1 to 3 show the quieter failure. Every week prints above 91%, and the backlog still nearly triples. A metric can look healthy every single week while the unresolved pile doubles underneath it. Track the backlog next to the rate, or the rate will hide it from you.
Resolution rate vs first contact resolution
These two get merged constantly, and they measure different things. One asks whether the ticket got resolved at all. The other asks a harder question of the ones that did: was a single contact enough, or did the customer have to come back.
A ticket closed after four replies and two re-opens raises one metric and lowers the other.
First contact resolution has the full re-open window math, and the two pages should be read together.
How to read yours without fooling yourself
Three habits keep the number honest. Define resolved as the customer confirming the fix, and report auto-closed tickets as their own line. Segment by category, because a blended 95% can hide one category rotting at 60%. And read it next to re-open count and satisfaction scores, because a team can close fast, close wrong, and still print a beautiful number.
Notice what the denominator leaves out. A customer who got an answer from self-service or an AI agent never became a ticket at all, so that volume never enters the formula.
Deflection rate measures exactly that volume, and it changes what your queue even contains.
WISMO is the largest ticket category in ecommerce and the most reliably resolvable one, so automating it is the fastest honest way to lift the number.
Resolution rate questions, answered straight
What is resolution rate in customer support?
The share of incoming tickets resolved in a period: tickets resolved divided by tickets received, times 100. A week with 1,000 arrivals and 920 resolutions prints 92%. It measures whether the queue gets cleared, not how fast or how well.
How do you calculate it?
Pick a period. Count the tickets that arrived in it, count the tickets resolved in it, divide the second number by the first, multiply by 100. That quotient is the resolution rate. Keep one definition of resolved across periods, or the trend means nothing.
What is a good resolution rate?
There is no honest universal resolution rate, because the definition of resolved varies more than the work does. A team that auto-closes silent tickets always outscores a team that waits for the customer to confirm. Judge yours against your own trend and your own backlog.
Can resolution rate go above 100%?
Yes. The formula divides work finished in the period by work that arrived in it, and finished work can include old tickets. Receive 1,100 and resolve 1,160 and you print 105.5%. Sustained readings above 100% mean the backlog is shrinking.
How is it different from first contact resolution?
Resolution rate counts every resolved ticket, however many contacts it took. First contact resolution counts only the ones resolved in a single contact, with no follow-up and no re-open inside the agreed window. A ticket solved on the fourth reply helps the first metric and hurts the second.
Why does a high resolution rate not guarantee good support?
Because it ignores speed, effort and correctness. A ticket closed in thirty seconds with the wrong answer counts the same as a careful fix, and an auto-closed ticket counts the same as both. Read it next to first response time, re-open count and satisfaction scores before believing it.