The event finishes on Friday. The team spends Monday clearing up, Tuesday on invoices, and by the time the feedback form goes out on Thursday the attendees have had a full working week between the event and the question. What comes back is short, general and kind: it was a good day, the venue was fine, thanks for organising it.
Nothing in that is useful. Not because attendees were being polite, but because the specific memories that make event feedback worth collecting have already gone. Which session ran over. Which room was impossible to find. The moment in the afternoon when half the audience checked out. Those details are available for hours, not days, and the form that catches them has to be ready before the event ends rather than after.
The window decides what the form can collect
| When it is sent | What it reliably gets | What it cannot get |
|---|---|---|
| In the room, before people leave | Specific sessions, specific problems, the largest number of responses | Whether anything was actually used afterwards |
| The same evening | Most of the detail, plus longer written answers | Responses from anyone who has switched off for the day |
| The next morning | An overall impression, one or two specifics | Which session, in most cases |
| A week later | A score, a short comment, a low response rate | Almost everything else |
That table is the whole argument for building the event feedback form before the event rather than after. The version sent in the room is not a rushed compromise. It is the only version that can answer the questions organisers actually have, and the later versions are what remains when the window was missed.
There is one thing a same-day form genuinely cannot do, which is tell anyone whether the event changed anything. That requires a second, much shorter contact two or three weeks later, and it is a different question rather than a repeat of the first one.
So the working pattern for most events is two stages. Three questions while people are still in the building, and one question a fortnight later asking what has actually been used since. Two small forms beat one large one sent at the wrong time, and the second one takes a minute to build if the tool can duplicate a form.
Ask fewer questions than feels responsible
An attendee answering in the room is standing up, holding a bag, and deciding whether to bother. The form has to win that decision in the first second, which means it has to be visibly short.
| Question as usually written | What comes back | Question that produces an action |
|---|---|---|
| How satisfied were you with the event overall | A 4, from everyone, every year | Which session was the most useful, and what made it useful |
| Was the venue suitable | Yes, from people who could not find room 3 | Was anything hard to find, get into, or hear |
| How did you hear about the event | Guesses, mostly wrong | Left off the form, and taken from the registration data instead |
| Any other comments | Blank | What would have to change for you to come back next year |
Three rules come out of that.
Take from the registration data what the registration data already knows
Job title, organisation, how somebody found the event, which ticket type they bought: all of it was collected at registration and none of it belongs on the feedback form. Asking again costs completions and produces worse data, because attendees answer registration questions carefully and feedback questions in a hurry. A tool that keys responses to the same person by email address puts the registration record and the feedback answer side by side without either form asking twice.
Keep exactly one scored question, and never change its wording
A score is only useful as a trend across events, and a trend requires identical wording and an identical scale. One consistent question, such as how likely somebody is to recommend the event, is worth more than six that get rewritten every year by whoever builds the form. Everything else can be free text.
Ask one forward-looking question
What to change is a better question than what was wrong, and what would bring somebody back is better than both, because it produces the material the next event is planned from. It also gets answered by people who enjoyed the day and have nothing to complain about, which is most of the room.
Two answers pay for the whole exercise
Most responses to an event feedback form are noise in the statistical sense: they confirm what the organisers already believed. Two kinds are worth the entire effort, and both need handling rather than counting.
The first is the quotable one. Somebody writes three sentences about exactly why the day was worth their time, in language no marketing team would produce. That sentence is the best material the next event will have, and using it requires permission, which means the form needs a checkbox asking whether the comment may be quoted and whether a name and organisation may be attached. Adding that checkbox is the single highest return change available to an event feedback form, and it costs one line.
The second is the complaint that needs a reply today. An accessibility failure, a charge that was taken twice, a session that did not happen, a member of staff who was rude. These arrive mixed into a spreadsheet with two hundred rows of scores, and they have a response time measured in hours. A form where each response carries an owner and a status makes them visible and assignable; a downloaded spreadsheet makes them somebody's job to notice. Event use cases usually turn on that distinction rather than on the questions.
Both of those also explain why an anonymous-only form is a poor default for events. Anonymity raises candour, and it makes it impossible to reply to the complaint or to ask permission for the quote. An optional contact field, with plain wording about who sees it, keeps both doors open.
The people who did not turn up
Every event has a gap between registrations and attendance, and the reasons for it are worth more than most of the feedback from the people who came. They are also easy to collect, because a single question sent to the no-show list gets a decent response rate from people who feel mildly guilty.
One question is enough: what stopped you coming. The answers cluster quickly, and they are usually about things that can be fixed. The time of day. The travel. A registration confirmation that never arrived, or arrived in a spam folder. A calendar invitation that was never sent. That last cluster is not an attendance problem at all, and it will keep happening at every event until somebody checks whether confirmation emails are actually reaching inboxes rather than being recorded as sent.
Send that question to the no-show list only, not to everybody. A form asking why somebody did not attend, delivered to somebody who did, reads as carelessness and undermines everything else in the message.
Getting the link in front of people
The form can be perfect and still collect thirty responses if nobody sees it at the right moment. Four things move the number.
Put a short link and a code on screen during the closing session, and leave it up. The last two minutes of an event are the highest-attention moment the form will ever get.
Say how long it takes. Three questions is a fact worth stating out loud and in the email subject, because attendees estimate the cost of a form before opening it and their estimate is always a long survey.
Send the same-day email within a few hours, from an address that belongs to the organisation rather than a generic one, and check that it arrives. An event feedback form that lands in spam produces the same numbers as an event feedback form that was never sent, and the two are indistinguishable unless the tool records whether the message was opened.
Send one reminder, two days later, to the people who have not answered. Not four, and not to everybody.
Make sure the form works on a phone with one question per screen. Attendees answer on a train or in a taxi, and a long scrolling page loses them halfway down.
Sponsors, exhibitors and speakers need their own forms
An event has at least four audiences and most organisers build a form for one of them. Attendees answer about the programme. Speakers answer about the room, the timings and the technical set up, which is where the detail about what went wrong backstage lives. Exhibitors and sponsors answer about footfall, position and whether the day was worth the fee, and that form decides next year's budget more directly than anything an attendee writes. Volunteers and staff answer about the parts of the day the public never saw, such as the registration desk at nine in the morning.
These forms share almost no questions, so combining them produces a long questionnaire full of items that half the recipients skip. Four short forms are easier to write and far easier to read. What matters is that the four sets of answers can be looked at together afterwards, because the interesting findings are the contradictions: attendees rating a session highly that the speaker thought was a disaster, or an exhibitor reporting empty aisles during the slot the programme was proudest of.
Two practical notes. Sponsor and exhibitor forms should not be anonymous, since each response is the start of a commercial conversation and needs a named owner to follow it up. And speaker feedback should be collected before the closing party rather than in the following week, for the same reason attendee feedback should: the specifics evaporate.
What happens to the responses after the event
This is where most event feedback dies, and the failure is always the same shape: everything arrives, somebody reads it once, a summary slide gets made, and the following year the same problems appear in the same form.
Four things stop that. Each response that needs action gets a named owner, so that the complaint about the room is somebody's job rather than everybody's. Each one gets a status, so that a month later it is clear which were answered and which were noted and closed. Replies are sent from the same screen the response lives on, so the reply is part of the record instead of sitting in one person's sent folder. And the decisions the team recorded, such as what changed for next year, export alongside the answers rather than living in a separate spreadsheet that leaves when its author does.
Those four are the working difference between a form builder and a tool with response management, and they are worth testing on one real submission rather than reading about. Submit an answer to a three question version of the real form, assign it to a colleague, mark it handled, write a reply from inside the tool, and then find that reply a month later.
What to change first
Build the three question form before the event and put its link on the closing slide, rather than writing it the following week. Add a checkbox asking permission to quote, and an optional contact field so the complaints can be answered. Then make sure every response that needs action has an owner and a reply attached to it, with Halict or any tool built the same way.
Q1. When should an event feedback form be sent?
While attendees are still in the room, with the link on screen during the closing session, and by email the same day for anyone who left early. A week later the specific memories are gone and what comes back is a general impression with a low response rate. If the question is whether the event changed anything, that needs a separate one question form two or three weeks afterwards.
Q2. How many questions should an event feedback form have?
Three to five for the in-room version. Which session was most useful, what would have to change for somebody to come back, and one scored question kept identical from year to year will carry most events. Anything already collected at registration should not be asked again.
Q3. Should an event feedback form be anonymous?
An optional contact field is usually better than full anonymity. Anonymity makes it impossible to reply to a complaint that needs answering the same day, and impossible to ask permission before quoting a good comment in next year's material. Make it clear next to the field who will see the address.
Q4. How can the response rate be improved?
Show the link during the closing session, state that the form is three questions long, send the same-day email from an address belonging to the organisation, and send exactly one reminder to the people who have not answered. Check that the email is being opened rather than assuming it arrived, since a form in a spam folder looks identical to a form nobody wanted to fill in.
Q5. Is it worth asking the people who registered but did not attend?
Yes, and one question is enough. The answers cluster into fixable causes such as the time of day, the travel, or a confirmation email that never arrived. Send that question only to the no-show list, because asking somebody who attended why they did not undoes the goodwill the event just earned.