Internal Help Desk Satisfaction Surveys That Tell You Something
How to measure satisfaction with an internal help desk without annoying everyone — one question after resolution, a follow-up only on bad scores, response rates, CSAT vs effort vs NPS, and turning answers into changes.
An internal help desk satisfaction survey works best as one question asked right after a request is resolved — "How did we do?" with a simple scale — plus an optional comment box that appears only for low scores. Report the share of satisfied responses per category and month with the response rate, read every low-score comment, and pair it with metrics like resolution time so you know why scores move. Short, timely surveys beat long annual ones.
Why measure satisfaction at all
Resolution times and SLA attainment tell you whether the desk is fast. They don't tell you whether people felt helped: a request closed quickly with a curt reply, or "fixed" by a workaround the person hated, looks fine in the numbers. A short satisfaction question catches what the timestamps miss — and gives requesters a voice, which internal teams rarely get.
One question, right after resolution
Timing and length matter more than anything else:
- Ask when the request is resolved, while the experience is fresh — in the resolution notification or on the request page.
- One question. "How satisfied were you with how this request was handled?" with a 1–5 scale or three faces (good / okay / bad).
- A comment box only for low scores: "What could we have done better?" People who had a fine experience shouldn't be asked to write.
- One click to answer from the notification, without signing in again.
Every additional question lowers the response rate and the value of the answers you do get.
LetRelay's own version is about as small as it gets: on a resolved request, the requester can give a thumbs up or down and optionally add a comment. Only the person who filed the request can rate it — enforced in the database, so ratings can't be stuffed by anyone else — and staff see the results in Analytics.
CSAT, effort or NPS?
Three common measures, and which fits an internal desk:
| Measure | Question | Good for |
|---|---|---|
| CSAT (customer satisfaction) | "How satisfied were you with…?" | Per-request feedback — the best fit for a help desk |
| Effort | "How easy was it to get this resolved?" | Finding friction in the process (forms, handoffs) |
| NPS (Net Promoter Score) | "How likely are you to recommend…?" | Brand loyalty — a poor fit internally, since employees can't choose another IT team |
Use CSAT per request. Consider an occasional effort question when you change the process (a new form, a new intake channel). Skip NPS for internal services.
How to report it
- Satisfaction rate: the share of responses that were satisfied (4–5 on a 5-point scale), per category and per month. A single number like "91%" is easy to track; averages of 1–5 scores hide the unhappy tail.
- Response rate: responses ÷ resolved requests. A satisfaction rate from 3% of requests is anecdote; watch that the rate is high enough to mean something, and be wary when it drops.
- Low-score comments: read every one. They're more useful than the number.
- Alongside operational metrics: first response, resolution time, reopen rate for the same period. If satisfaction drops as resolution time rises, you know why. See help desk metrics that matter.
Beware the biases
- Self-selection. People with strong feelings respond more; very happy and very unhappy voices are over-represented.
- Scoring the outcome, not the service. A person denied access they wanted may score the desk low even if it handled the request perfectly. Read comments before judging.
- Agent gaming. If individual scores feed performance reviews, agents may nudge ("please rate us 5!") or avoid difficult requests. Use scores to improve the service, and discuss individual feedback privately and constructively.
- Survey fatigue. Don't survey the same person more than once a week; skip surveys for automatic resolutions and duplicates.
Turn answers into changes
A survey without action teaches people not to answer. Close the loop:
- Follow up on low scores within a couple of days: a short, human reply — "Sorry the fix didn't hold; we've reopened it." Often the follow-up alone turns the experience around.
- Group comments monthly by theme: slow response, unclear communication, problem not really fixed, process friction.
- Pick one theme to fix each month — a knowledge-base article, a clearer template reply, a shorter form, a target change.
- Tell people what changed because of their feedback. "You told us access requests were slow; approvals now go straight to the data owner."
A worked example
An IT desk resolves about 250 requests a month and adds a one-question survey to its resolution notifications.
- Month one: 92 responses (37% response rate). 84% satisfied. Eleven low scores; nine comments.
- Reading the comments: five say some version of "it was closed but it isn't fixed"; three mention not hearing anything for days; one is about a denied software request.
- Cross-checking operational data: the "closed but not fixed" comments all come from the Something's broken or slow category, which also has the highest reopen rate. The "didn't hear anything" comments are requests that waited on another team.
- Changes: agents now ask "can you confirm it's working?" before resolving broken-or-slow requests; requests waiting on another team get an automatic status note to the requester.
- Month three: 88% satisfied, the reopen rate for that category halved, and the comments shifted to smaller things — a sign the big problems were addressed.
The satisfaction number moved, but the value came from the nine comments and from reading them next to the reopen and handoff data.
Designing the question well
Small wording choices change answers:
- Ask about the service, not the person. "How satisfied were you with how this request was handled?" rather than "How did Sam do?".
- Label the scale's ends in words ("Very dissatisfied" … "Very satisfied"), so everyone reads it the same way.
- Keep the scale the same forever. Changing from 5 points to 3 breaks the trend line.
- Make the comment box optional and short. A single line invites an honest sentence; a large empty box invites nothing.
Annual surveys: short and specific
A yearly survey of all staff can complement per-request feedback for things per-request questions can't capture: "Do you know where to get help?", "Is the help desk available when you need it?", "What's one thing we should change?". Keep it to five questions or fewer, report the results openly, and tie each answer to a planned change.
Beyond surveys
Satisfaction shows up in behaviour too:
- Side-channel requests — if people still message agents directly instead of using the desk, something about the desk isn't working for them. See onboarding a team to a new help desk tool.
- Repeat questions — the same question asked every week suggests the knowledge base isn't findable. See ticket deflection.
- Reopened requests — fixes that didn't hold.
FAQ
What is CSAT for an internal help desk?
A per-request customer satisfaction score: after a request is resolved, the requester answers one question about how satisfied they were, and you report the share of satisfied responses.
When should a help desk send a satisfaction survey?
Right after a request is resolved, in the resolution notification, with a single question and a comment box only for low scores.
Is NPS useful for an internal IT help desk?
Rarely. Employees can't choose another internal IT team, so "likelihood to recommend" doesn't measure much. Per-request satisfaction and occasional effort questions are more useful.
What's a good response rate for help desk surveys?
High enough that the results represent your requesters; the lower it is, the more cautiously you should read the score. Track it next to the satisfaction rate and investigate drops.
Sources
Building Relay — the internal request desk with AI triage and SLA tracking.
Run your internal requests on LetRelay
AI triage, SLA-tracked queues, and bottleneck analytics — the help desk your team actually likes. Free to start.
Keep reading
Facilities Request Management: From Broken Chairs to Keys
How to run facilities requests like a proper service — categories, location on every request, safety issues that skip the queue, vendor work, recurring maintenance and the numbers that show where the building needs attention.
HR Service Desk: Handling Employee Requests Without the Inbox
How an HR service desk works — categories for pay, leave, contracts and benefits, privacy that's enforced not promised, targets that respect deadlines like payroll, self-service for policy questions, and the metrics HR can use.
Shared Inbox vs Help Desk: When to Switch
A shared inbox is fine until two people answer the same email and a third goes unanswered. What a shared inbox does well, where it breaks, what a help desk adds, and the signals that it's time to switch.