Form questions
Support ticket feedback questions
Ask four to six, sent when the ticket closes. Three never come out: a five point satisfaction rating, a question asking whether the problem is actually solved, and one open comment. The resolved question matters most, because a ticket marked closed by an agent and a problem fixed for a customer are two different events.
A ticket survey competes with the moment a customer finally gets their day back, so it earns about ninety seconds. These questions put the rating and the resolved check in the first two screens and treat everything after that as a bonus.
The questions
The rating
How satisfied are you with how this ticket was handled?Single choiceRequired
Five labelled points. Ask about the handling, not about whether the underlying problem was the customer's own fault.
How would you rate the outcome, separately from the service?Single choice
Worth splitting when your team often has to deliver a no. A good no still deserves a good service rating.
Was the answer you got the answer you needed?Single choice
How likely are you to contact support again if you need to?Single choice
A quiet measure of whether the experience put them off asking.
Was it resolved
Is the problem actually solved?Single choiceRequired
Fully, mostly with something left over, no but the next step is clear, no and nothing is clear. Four answers that split a closed ticket four ways.
Has it stayed solved?Single choice
Send this one a week later rather than at close.
How many times did you have to get in touch about this?Single choice
Did you have to explain it more than once?Single choice
How long did it take from your first message to the fix?Dropdown
How did the wait for a first reply feel?Single choice
How it felt, not what the clock said. The clock is already on the ticket.
The agent
The person who handled itRating grid
A five row grid on a poor to excellent scale: understood the problem, knew the product, explained things plainly, kept me updated, was straight with me.
Would you be happy for this agent to handle your next ticket?Single choice
Did anybody chase you, or did you chase them?Single choice
Were you told what would happen next?Single choice
Did the ticket change hands?Single choice
Before and after
Did you try to solve it yourself first?Single choice
If you looked in the help centre, what happened?Single choice
Found nothing, found something out of date, found something you could not follow, found it but it did not work. Four different jobs for the documentation team.
What is the one thing that would have made this easier?Long textRequired
Anything you want the agent to know?Long text
Passed on in the customer's own words, which is worth saying in the question so praise reaches the person it is about.
Which channel would you use next time?Dropdown
Which questions to make required
Three are required. The rating, because the response is not a rating without it. The resolved question, because a satisfaction score with no resolution figure beside it cannot be read: a four from somebody whose problem is still broken means something different from a four from somebody whose problem is fixed. And the open question, kept short and placed last, because a rating with no comment gives a team nothing to change.
- How satisfied are you with how this ticket was handled?
- Is the problem actually solved?
- What is the one thing that would have made this easier?
The form built from these questions
A static copy of a form built from part of the bank, exactly as it renders.
Three ways to change this bank
For an internal help desk, swap the channel questions for location and system: which office or site the person works from, and which system the ticket was about. Internal desks rarely have a channel problem and almost always have a system that generates a third of the tickets. For a shorter version, send two questions in the closing email: the rating and one comment box. That is the version most desks can actually sustain at volume, and it still supports a weekly read. For a follow up wave, write a week later to anybody who scored one or two, and ask two things: has it stayed fixed, and did anybody come back to you. A reopened problem that nobody noticed is the most expensive thing a closed ticket can hide.
Questions people ask
How short does a ticket survey have to be?
Four to six questions, and two if you want it answered at volume. The survey arrives at the moment the customer wants to move on, so every screen past the second is borrowed time. Put the questions you will actually report on first, because a partial response is not saved.
Should every closed ticket trigger one?
Every ticket, but not every customer every time. A cool down of two to four weeks per person stops your most frequent contacts being surveyed weekly, which is where survey fatigue and angry ratings come from. Exclude tickets the customer opened and closed themselves.
Should agents see their own ratings?
Yes, and quickly, or the survey becomes a management tool rather than a feedback loop. What needs care is the comment field, since comments are passed on in the customer's own words. Say so in the question, and have a lead read anything rated one or two before it goes to the agent.
Why separate the rating from whether it was resolved?
Because they move independently. A customer whose problem cannot be fixed can still rate the service highly if they were told the truth early. Collapsing both into one number leaves you unable to tell a service problem from a product problem, which are owned by different teams.
What response rate should I expect?
Ratings embedded in the closing email or shown in the chat window run well ahead of a separate survey link. Expect a wide spread by channel and a bias toward the extremes: very happy and very unhappy customers answer more. Read the middle of your distribution as underrepresented rather than absent.
How does the agent grid come out in an export?
As one CSV column with each statement and its answer joined together, rather than five separate columns. For a weekly read that is fine. If you plan to trend the five behaviours separately, ask them as five single choice questions instead and accept the extra screens.
Build this form in about a minute
The button opens the form generator with this form already described. Change the wording to match your own, generate, then edit any field.
Create a form with OneCraftOther question banks
Customer effort score questions
Effort is what a customer remembers about a support experience long after they have forgotten who helped them. These questions measure the whole journey from first message to fix, rather than rating the last person in the chain.
Delivery experience survey questions
A delivery survey has to separate three things customers blame on one another: the promise the shop made, the condition the parcel arrived in, and what the driver did. Asked as one satisfaction score they are impossible to untangle.
NPS survey questions
An NPS survey is one question plus whatever you ask next, and the second part is where nearly all the value sits. These questions give you three different follow ups so a promoter, a passive and a detractor are each asked something they can actually answer.
On writing the questions themselves, read How to write survey questions. For everything the builder can do, see the form maker. Step by step: Create a conversational form with AI, then View, export and manage your responses.
Written and checked by the OneCraft team. Last checked .