Event Feedback Survey Template
A 12-question event feedback survey you can take right now, for a conference, a webinar or a one-day event. Some organisers call it a post-event survey. Eight ratings, one question about when the answer was given, two open answers and one number to trend. Read the event survey questions below, copy them, or open the template in the editor.
- 12questions
- About 3 minto complete
- In the roomthe moment to ask
- 7 daysthen close it
When to send it: the event clock and the inbox clock
Two clocks start when your event ends, and the usual advice runs them together. One decides what an answer means. The other decides how many answers you get and when you can close the link.
Clock one, event to ask. This is the one that changes the data: the nearer the ask sits to the thing being rated, the more the answer is about the experience and the less it is about a memory of it. In an experience-sampling study, a rating given in the same moment as the feeling tracked it closely, and one given three and a half hours later barely tracked it at all.[2] That was 98 students rating ordinary days: a direction, not a rate.
Clock two, ask to reply. Once the invitation is out, replies arrive fast and then stop. Across 500,000 email invitations, about 4 in 10 responses came back within a day and 8 in 10 within the week, the second week adding roughly a tenth and the third almost nothing.[1] That is one platform's email collector from 2009 and 2010, so read the shape, not the rate: it says when to close, not when to send.
Why the advice everywhere is wrong. The rule you keep reading, send it before people forget, comes with a number that is never sourced. What has been measured is the second clock, quietly attached to the first. Mixing them turns a rule about closing a survey into a rule about running an event.
| Your event | Ask for the whole-event rating | Ask for the detail |
|---|---|---|
| A webinar | In the last minute, on the closing screen, while the call is still open. | In the same session. People close the tab and the event stops existing. |
| A one-day event | On the closing slide, with a printed code on the tables. | A same-day message, before the evening. |
| A conference over two or more days | At the end of each day, while that day is still separable. | One whole-event ask on the last day, then close the link a week later. |
One survey, three trigger points. Splitting by audience, one survey each for attendees, speakers, sponsors and volunteers, gives you four sends that never happen. Splitting by moment changes what you learn, and a conference has a Tuesday morning that nobody can rate on Friday.
Three things get called an event and are not this survey. For a course, the post-training feedback survey asks whether people learned it and will use it. For one facilitated session with no agenda, the general workshop survey fits that room. A hackathon survey adds the judging and team dynamics that format lives on.
Build it before the event, not after it. All twelve questions are written, so open them in the free survey creator, change any wording, and have the link on a slide before anyone packs up.
The 12 event feedback survey questions, and what to swap for a webinar
Eight ratings, one question about when the answer was given, two open answers, one number to trend. Each item says what a bad answer to it looks like. Copy one block or all twelve.
The whole event (ask these two in the room)
If somebody answers nothing else, these two still tell you whether to run it again.
- Overall, this event was worth the time it took.The answer to protect. High here with a low programme score means people came for the room.
- I would come to this event again.Intent, which is harder to be polite about. Below the overall rating, something specific put people off.
The programme
What was said, separated from how it was run.
- The sessions gave me something I can use.Low here with a high speaker score means the content was enjoyable and useless.
- The speakers were worth listening to.The whole bill, not a scorecard per person. One weak slot rarely moves it; three do.
- The programme ran to time.The item most often low at a first event. Overrunning eats the breaks people came for.
The practical side
Whether people could get at the programme at all.
- Getting to the sessions I wanted was straightforward.Signs, room changes and joining links. A low answer usually means one broken moment, not the venue.
- I could take part in every session I attended.Access, in one construct: sight lines, sound, captions, a room somebody can get into. Under a 4 belongs with the comments, because the people it failed rarely complain.
- There was enough time to talk to other people.The reason a lot of people bought the ticket. Often the lowest item at a packed conference.
When the answer was given
One tap that makes every other answer readable.
- When are you answering this?A later group rating the practical items lower is remembering one queue, not reporting a venue that failed.
In their own words
Two open answers, before the end, so people still have a reason to type.
- Which part of the event was the most valuable to you?The high point, named. What comes up again and again belongs on the next agenda.
- What one thing should we change next time?One thing, not any comments. It gets a specific answer instead of a shrug.
The number to trend
A rating is the cheapest thing to answer once someone is nearly finished, so it closes the survey.
- How likely are you to recommend this event to a colleague?Means little after one event and a lot after four. Keep the wording identical or the trend breaks.
Nothing here asks for a name. A form that ends on name, email and phone loses answers when it is nearly finished, which is why the last question here is a rating. If you need a contact detail, ask in the middle, optional.
Swap-ins, not extras. Three event shapes want different wording. Take the replacement, leave the rest of the twelve alone; the copy button above still hands over the survey as it stands.
| If your event was | Take out | Ask instead |
|---|---|---|
| A webinar | Questions 6 and 8, about getting to sessions and time to talk. | The chat and the questions at the end were worth taking part in. The platform did not get in the way. I would watch this again as a recording. |
| A conference over two or more days | Nothing. | Which day are you answering about? Which sessions did you attend? Ask both at the end of a day, not the end of the week. |
| Ticketed, catered, or with an exhibition | Nothing. | The ticket was fair for what I got. The food and drink held up across the day. The exhibition was worth walking around. Only the ones that existed. |
Two jobs this survey hands over. Question 4 rates the speakers as a bill, not one by one, because a per-person scorecard carries a confidentiality problem this one does not: use the facilitator feedback survey for that. When the venue is a hotel and the subject is the stay rather than the programme, the guest satisfaction survey asks it.
Your event finished on Tuesday: what a late answer is still worth
Most people reach a page like this after the event. A late answer is not a weaker version of an early one; it is an answer about different things, which changes what to ask.
What a memory of an event is made of. Days later, nobody averages a two-day agenda. A retrospective rating of an episode leans on its most intense moment and on its ending.[4] That was measured on pain and anxiety rather than on conferences, so take the shape and not a rate. It tells you what you are holding: the closing keynote and the worst queue, wearing the label of the whole event.
So change what you ask, rather than giving up. Late, the whole-event rating and the two open answers are still worth having. The per-session detail is not: nobody can separate Tuesday's breakout from Wednesday's four days later. Ask less, ask about the parts that stood out, and let question 9 record that the answer came late.
The in-room ask has its own tilt. The people still in the room are the people who stayed, and being asked face to face pulls answers up: in a hospital study every headline item scored higher when a volunteer asked in person than when the same questions went by post.[3] In-room answers run warm, late answers run peaky, and neither is the truth alone.
Forty answers, mostly fours and fives: how to read them
An average with nothing to compare it to is a number, not a finding. Event feedback has no standard instrument and no norms table, and the guides that ask what a good score is never answer it. Five reference points instead of a benchmark.
| What you have | How to read it | What it is not |
|---|---|---|
| One rating, averaged | Report the share who chose 4 or 5, not the mean. A share moves when opinion moves; a mean hides a split room. | Not a mark out of five. Nobody publishes a pass mark for an event. |
| Eight ratings, side by side | Rank your own items against each other. The lowest is the thing to fix, whatever it looks like in absolute terms. | Not a comparison with another organiser, whose people paid, travelled and expected different things. |
| This event against your last one | Same questions, same wording, same point in the day. The gap between two of your own events is the comparison that carries. | Not valid if you changed a question or asked at a different moment. |
| In-room answers against later ones | Split on question 9 before you average anything, then report the groups side by side: they were given at different moments and are not two versions of one number. | Not a reason to bin the late ones. They record what survived the week. |
| The two open answers | Read them before the chart. Tag each to the rating item it is about, then count the tags. | Not a vote. Ten people on the coffee queue is one problem, not the biggest. |
What a response rate is really telling you. Mostly, when you asked. In one hospital study about half of the people approached in person answered, against about three in a hundred for the same questions posted afterwards.[3] The moment moves your rate further than the wording ever will. Anything every survey needs, one link, a stated length, a closing date, is in the survey research guide.
The recommend question at the end. One number that survives a change of programme, which is why it closes the survey instead of feeding the event rating. Read it as a share of high scores against a share of low ones, or use the NPS survey template to trend it across a series.
Event feedback survey FAQ
How many questions should an event feedback survey have?
Twelve here: eight ratings, one about timing, two open answers and one recommend question. Where they sit matters more than how many: the rating nobody will skip goes first, and the open answers go before the end, while people still have a reason to type.
Should an event feedback survey be anonymous?
This template asks for no name and no email, so an attendee cannot be picked out of the answers alone. The organiser who set it up is the person who reads them, and that is what to say before people answer. Promise no more: at a small event, one added question about which team somebody came from is enough to name them.
What is the difference between an event feedback survey and an event evaluation form?
Only the noun. An evaluation form usually means a paper sheet handed round the room and a survey usually means a link; the questions are the same either way. This one is a link that works on a phone while people are still sitting down, the job the paper sheet used to do.
Methods and sources
What this template is based on
Four measurements carry the timing argument: an arrival curve after email invitations,[1] a lag study of satisfaction ratings,[2] a comparison of asking in person against posting,[3] and the peak-end pattern in retrospective ratings.[4]
Question licensing
Event feedback has no standard instrument, so every question here is ours to copy, edit and translate. The closing question uses plain recommend wording, free to anyone. Net Promoter(R) and NPS(R) are registered trademarks, and Net Promoter Score(SM) a service mark, of Bain and Company, Inc., NICE Systems, Inc., and Fred Reichheld.[5]
Limits and disclosure
No finding here was measured on event attendees. The arrival curve is one platform, email only, 2009 to 2010; the lag study is 98 students rating ordinary days; the in-person comparison is an emergency department; peak-end comes from pain and anxiety research. Each is cited for direction, not as a rate.
SuperSurvey makes the editor this template opens in. Everything here is free to use with or without it.
References
- Zheng, J. "How many days does it take for respondents to respond to your survey?" SurveyMonkey Curiosity. surveymonkey.com
- Anic, P. and Toncic, M. "Affect and satisfaction time dependency: An experience sampling study." PsyCh Journal 13, no. 3 (2024). pmc.ncbi.nlm.nih.gov
- Willner, K. "Real-time Patient Experience Surveys Lead to Better Scores." Western Journal of Emergency Medicine, 2025. pmc.ncbi.nlm.nih.gov
- Muller, U.W.D., Witteman, C.L.M., Spijker, J. and Alpers, G.W. "All's Bad That Ends Bad: There Is a Peak-End Memory Bias in Anxiety." Frontiers in Psychology 10 (2019). pmc.ncbi.nlm.nih.gov
- Bain and Company, NICE Systems and Fred Reichheld. Net Promoter trademarks and licensing. netpromotersystem.com
Have the link ready before the closing slide
Twelve questions, about three minutes. Open it in the editor, change any wording, and put the link on the last slide.
Use this template