You pull last week's closed-tickets report by technician. One tech closed 22 a day. Another closed 9. The easy read is that the second tech is slow. Then you open the tickets: the first one took password resets and "can't print" all week, and split two calls into three tickets each. The second one did two new-user setups, a mailbox migration and every escalation Level 1 couldn't finish. And neither number counts the calls either of them fixed without opening a ticket.
That's why "how many tickets should a help desk tech close per day" has no honest single answer. The right number for your desk comes from two things you can measure: how many minutes a tech actually has for ticket work, and how long each type of ticket takes. Divide the first by the second, by ticket type, and you have a daily range for that tech and that queue. A flat target like "20 a day" rewards whoever gets the easy tickets and punishes whoever takes the hard ones.
The short answer
For a service desk tech working mostly short, phone-resolved tickets, a daily count in the high teens or more can be normal. For a tech doing installs, onboardings, projects or escalations, single digits can be a full day. Both can be right. The number only means something next to the ticket mix and the handle time behind it.
Published benchmarks show how wide the spread is. In an HDI column on tickets per technician per month, MetricNet's Jeff Rumburg reported desktop support desks ranging from 30 to 198 tickets per technician per month, and argued that staffing "should be based upon workload, not user population." It's older data (2012) and it's desktop support, not a phone-first service desk, so treat it as proof of the spread rather than a target.
Do the math from handle time
Handle time is the input that makes the count mean something. HDI's metric of the month on ticket handle time defines it as the time an agent spends on a ticket, including talk time and the wrap-up work after the call. Its own example shows how fast the count moves with it: at a 10-minute handle time an agent handles about 400 tickets a month, and at 15 minutes, about 270. Five extra minutes per ticket, roughly a third fewer tickets.
Here is the calculation for one tech. The numbers are illustrative; swap in your own.
- Start with the shift. 8 hours, minus an hour for lunch and breaks, minus 30 minutes for the queue huddle and admin: 390 minutes.
- Take out time that isn't ticket work. Nobody spends every available minute on tickets. Assume 80%: about 312 minutes.
- Price the mix. Say a typical day is 6 password or MFA resets at 10 minutes, 4 Outlook or printer problems at 25 minutes, and 1 new-user setup at 75 minutes. That's 11 tickets and 235 minutes.
- Leave room for the rest. The remaining 77 minutes go to follow-ups, callbacks and the ticket that runs long. Realistic capacity: 11 to 13 tickets.
Now give a second tech nothing but 10-minute resets. Same 312 minutes, about 31 tickets. Same effort, almost three times the count. If your scorecard compares those two numbers directly, it's measuring the dispatcher, not the techs.
Why "tickets closed per day" lies
Even with the math done, the raw count has failure points every service manager runs into:
- Ticket splitting. One call becomes three tickets. The count goes up and nothing extra got done.
- Closing early. A ticket closed before the user confirms the fix comes back as a reopen or a new ticket. Watch reopen rate next to the closed count.
- Fast escalation looks productive. A Level 1 tech who escalates in four minutes posts a better count than one who spends twenty minutes and resolves it. HDI notes that handle time rises as first contact resolution rises, so the tech resolving more on first contact can look slower on handle time.
- Phone fixes that never become tickets. A tech who answers the phone all day and fixes things in five-minute calls can show the lowest count on the team, because half that work never reached the PSA. That work has no ticket, no note and no time entry. Our post on MSP time tracking covers how to measure that gap.
What to track instead
Keep tickets closed on the scorecard, but never alone. Read it next to:
- Closed tickets by type, so a day of onboardings isn't compared with a day of resets.
- Average handle time by ticket type, from time entries rather than open-to-close duration, which includes waiting.
- Reopen rate per tech, to catch early closes.
- Logged hours versus hours on shift. If a tech worked 8 hours and logged 4, the count is missing work, not effort.
- Inbound calls versus tickets touched, from your phone system and PSA. A big gap means calls are being handled off the books.
You can get most of this from a time entry export in ConnectWise PSA, Autotask, HaloPSA or Syncro, grouped by technician and by the ticket's type or category. Our MSP KPIs scorecard goes through how each of these numbers can mislead, and the IT staffing ratios post uses the same handle-time math to decide when to hire.
Where AI changes the number, and where it doesn't
AI doesn't make a printer driver install go faster. What it can take off a tech's plate is the work around the fix: writing the note, logging the time, and chasing a user for details the ticket arrived without.
That changes two inputs in the math above. Wrap-up time drops when the note and time entry are drafted from the call, and the count gets more honest because phone-only fixes show up in the PSA. Vight does this for MSPs on ConnectWise, Autotask, HaloPSA and Syncro: it pulls each recorded support call from the phone system, matches it to the client and ticket, posts a summary of the issue, troubleshooting and resolution as a ticket note, and logs a time entry for the time worked. On ConnectWise and HaloPSA, ticket triage emails the requester for missing details before a tech picks the ticket up. Write-ups like these are still worth a tech's review, and they make better ticket notes the default rather than the exception.
Common questions
Is 20 tickets a day a lot for a help desk tech?
It depends on the mix. Twenty 10-minute resets fit in a day with room to spare. Twenty tickets that each need 30 minutes don't fit in an 8-hour shift. Ask what the 20 were before you judge the number.
Should tickets closed per day be on a technician's scorecard?
Yes, as one input, read alongside handle time by ticket type and reopen rate. On its own it rewards splitting tickets, closing early and cherry-picking easy work.
How do I set a fair daily ticket target?
Measure average handle time by ticket type from time entries, take the minutes a tech really has for ticket work, and divide. Set a range per role, Level 1 and escalation separately, and revisit it when the ticket mix changes.
Why does my best tech close the fewest tickets?
Often because they get the hardest work: escalations, projects and the clients nobody else wants. Check their ticket types and handle times before reading the count as a performance problem.
If you want to see how many of your calls never become ticket notes or time entries, book a Vight demo and see how calls turn into ticket notes and time entries in your PSA.
