Most post-event surveys die the same death: a 20-question form emailed three days later, opened by nobody, and quietly forgotten by the marketing team that sent it. The event happened, the feedback opportunity happened, and neither one talked to the other.
Done right, a post-event survey is one of the highest-signal data sources you have, it tells you who's coming back next year before they've decided themselves. Done badly, it's a compliance checkbox nobody reads.
Timing beats question design
Before you write a single question, fix the timing, because timing has a bigger effect on response rate than almost anything else you'll tweak.
Event surveys sent by SMS get 32% more completions when sent within two hours of the event ending, compared to waiting even a single day. In-person collection at the event itself, a tablet at the exit or a quick QR code on the closing slide, can reach 85-95% completion rates, far above anything you'll get from a follow-up email.
Capture feedback in three windows if you can: a one-question pulse check at the exit, a full survey within two hours by SMS or app push, and a deeper follow-up email at 24-48 hours for the people who want to write more.
By comparison, email NPS surveys alone typically land 15-25% response rates, and SMS surveys can reach 45-60%. If your only channel is a follow-up email sent days later, you're leaving most of your possible responses on the table before you've written a single question.
The four questions that actually predict next year
Long surveys kill response rates and mostly get skipped past question three. Cut to what actually predicts behavior.
- The NPS question β "How likely are you to recommend this event to a colleague, 0-10?" A strong NPS response rate in 2025 runs 30-40%, treat that as your benchmark, not 100%.
- "Will you attend next year?" β a direct intent question, more predictive of actual return attendance than the NPS score itself.
- "What was the single most valuable session or moment?" β open text, tells you what to keep and double down on.
- "What almost made you not attend, or almost made you leave early?" β this is the question most surveys skip, and it's often the most actionable one you'll get.
NPS (Net Promoter Score) is a 0-10 loyalty metric, scores of 9-10 are Promoters, 7-8 are Passives, 0-6 are Detractors, and the score itself is percent Promoters minus percent Detractors.
Four questions, answerable in under ninety seconds, will outperform a fifteen-question form almost every time. If you want deeper qualitative detail, offer it as an optional second page, not a mandatory gate before submission.
Segment the score, don't just report the average
A single blended NPS number hides almost everything useful. The real insight is in the cuts.
Split your NPS by attendee type at minimum: first-timers versus repeat attendees, sponsors versus general admission, and VIP tier versus standard badge. A blended score of 42 can be hiding a first-timer score of 15 and a repeat-attendee score of 65, two completely different problems that need completely different fixes.
Cross-reference the "will you attend next year" answer against the NPS score too. A Promoter who says "no" to next year is a different, more urgent problem than a Passive who says "no", the Promoter liked the event but something else (budget, timing, a competing event) is pulling them away, and that's worth a direct follow-up call, not just a data point in a spreadsheet.
Closing the loop, the step almost everyone skips
Collecting feedback and doing nothing visible with it trains attendees to stop bothering with your survey next year.
Two things close the loop cheaply. First, a short "you said, we're doing" recap email or landing page a few weeks after the event, listing two or three specific changes based on direct feedback themes. Second, a personal follow-up to every Detractor score, not a template, an actual person reaching out to understand what went wrong, because a recovered Detractor often becomes your most vocal advocate the following year.
Neither of these costs much, but both directly affect whether your response rate goes up or down next time. Attendees can tell the difference between a survey that feeds a slide deck nobody sees and one that visibly changed something, and that difference is exactly what turns a 20% response rate into a 40% one over a few event cycles.