An RSVP Is a Sentence. It Was Never a Signature

Nobody lies when they RSVP. At the moment they click yes, they mean it. The problem is that a yes given three weeks out is a statement about intention, and the event happens to a person with a Tuesday - and between those two people sits every no-show you have ever had.

Quick answer

Eventbrite's own data puts free-event no-show rates as high as 50 percent, against as low as 10 percent for paid events. The variable is not enthusiasm. It is whether saying yes cost anything.

An organiser who ran a tech meetup for five years measured roughly a 30 percent drop-off between RSVPs and attendance, and simply built it into his catering: order for 75 to 80 percent of the confirmed list.

The two fixes that actually move the number are adding a small cost to the yes - even a token one - and running a reminder sequence rather than a reminder. Everything else is a variation on those.

The two different people who make up one RSVP

Three weeks before an event, a person reads a description and imagines a version of themselves who is free, rested, and interested. That person clicks yes, honestly.

On the night, a different person exists. That one has had a day. The commute is worse than expected, it is raining, and there is a real, immediate cost to leaving the house against an abstract benefit that was much more vivid three weeks ago.

No-shows are the accounting difference between those two people. Treating them as a character flaw in your attendees is both unkind and useless, because it points at the one variable you cannot change. The useful framing is that a yes was cheap to give and free to break, and both of those are things an organiser controls.

You are not competing with disinterest. You are competing with a sofa, on a Tuesday, in the rain.

How large the gap actually is

It is worth having real numbers rather than a vague sense that some people do not turn up, because the size of the gap determines whether it is a planning problem or a rounding error.

Eventbrite's published analysis puts free events at no-show rates that can reach 50 percent, while paid events run as low as 10 percent. That is not a small spread. At the top of that range, half of your catering, half of your seating, and half of your speaker's expected audience are theoretical.

Dave Kiss, who ran a tech meetup for five years, reported a consistent figure of around 30 percent drop-off - squarely between those two poles, which is roughly where a free-but-well-run community event tends to land. His response is the pragmatic one: he orders food for 75 to 80 percent of the RSVP count and budgets around $250 to $300 per event accordingly.

Event type - No-show rate - Plan catering for

Free, no reminders - Up to 50% - ~50–60% of RSVPs

Free, well-run community event - ~30% - 75–80% of RSVPs

Paid ticket - As low as 10% - ~90% of RSVPs

Notice that the practical advice falls out of the number immediately. An organiser who knows their real rate does not have a no-show problem; they have a multiplier. The organiser in trouble is the one planning for 100 percent and being surprised every single time.

Why a token price does so much work

The most counterintuitive finding in the ticketing data is that the effect of charging is not proportional to the amount charged. Most of the benefit arrives in the gap between zero and any number at all.

Eventbrite's analysis notes that even a charge well under a dollar measurably raises the perceived value of a ticket that would otherwise be free. The money is irrelevant to the organiser's budget at that level. What changes is the nature of the act: clicking yes on a free thing is a reflex, while entering a payment method is a decision, and decisions are remembered in a way reflexes are not.

There is a second effect that matters more for planning. A paid ticket creates a small, concrete loss for not attending. Behavioural economics has a long-standing name for why that matters disproportionately, but you do not need the theory - the 50-percent-versus-10-percent spread is the finding.

This is not an argument that every event should charge. It is an argument that if you need a reliable headcount and you are currently getting a coin flip, the smallest available intervention is also the most effective one.

The jump from free to $1 does more for attendance than the jump from $1 to $20.

A reminder is not a reminder sequence

The single most common structural mistake is sending one reminder, usually the day before, and concluding that reminders do not work.

Kiss's five-year meetup ran five touches, and the spacing is the point: the initial announcement, then two weeks out, one week out, three days out, and the platform's own automatic reminder. Each one catches a different failure. The two-week touch catches people who need to arrange something. The one-week touch catches the calendar. The three-day touch catches the actual decision to go.

Timing - Catches - What it should contain

Announcement - Interest - The reason to come at all

2 weeks out - Logistics - Anything requiring arrangement - childcare, travel, a colleague

1 week out - The calendar - A confirmation that it is real and unchanged

3 days out - The decision - Practical detail: address, timing, what to expect

Day-of automatic - Memory - The bare facts, nothing else

The reason this outperforms a single reminder is that no-shows do not have a single cause. Some people forget, some never arranged the thing that would have made attending possible, and some simply lose the thread. One message can only address one of those.

The reminder that makes it worse

There is a failure mode worth naming, because organisers stumble into it while trying to do the right thing.

A reminder sent inside a group chat costs every recipient a notification, which means an organiser who is being considerate learns to send fewer of them. That is exactly backwards: the plan needs more touches and the medium punishes them. This is the same structural problem that makes a chat a poor place to hold any plan - every update costs an interruption rather than an edit.

The fix is not fewer reminders. It is reminders that point at a stable address rather than restating facts, so that the message can be short and the details can be current without the two being the same thing.

Make the cancel button easy to find

This is the advice that feels wrong and consistently works: make it trivially easy for someone to change their yes to a no.

The instinct is that friction on cancelling preserves attendance. It does not. It preserves the RSVP, which is a different and much less useful thing. A person who has decided not to come is not going to come; the only question is whether you find out on Thursday or at the door.

An easy cancellation converts a silent no-show into a released seat, an accurate catering number, and a slot for someone on the waitlist. It also removes the social awkwardness that causes people to simply go quiet, which is the actual mechanism behind most unexplained absences.

A no on Thursday is worth more than a yes at the door. Price your friction accordingly.

Overbook, deliberately and quietly

If your real attendance rate is 70 percent and your room holds 40, confirming 40 people means running a 28-person event in a 40-person room, every time.

Airlines have priced this problem more carefully than anyone. The relevant caution from that research is not about whether to overbook but about what happens when you get it wrong: the cost of turning someone away rises steeply the later they find out, and the most expensive version by far is the person who discovers it after arriving.

So the rule is to overbook conservatively against a measured rate, not an optimistic one, and to have a real answer ready if everyone turns up. Standing room, a second overflow table, or an honest early message beats the alternative, which is a person who travelled across a city to be told at a door.

What to measure so this stops being guesswork

• Your RSVP-to-attendance ratio, per event, over time. One event tells you nothing. Six events tell you your multiplier, and the multiplier is the whole game.
• Whether the ratio moves with weather, day of week, or season. Most organisers find it does, substantially, and can plan around it once it is written down.
• Returning attendees versus first-timers. These almost always have different no-show rates, and knowing which group is drifting tells you whether the problem is your reminders or your event.
• Cancellations as a separate number from no-shows. A rising cancellation rate alongside a falling no-show rate is a success, not a problem, and a combined metric hides it.

Six events of honest record-keeping converts this entire topic from anxiety into arithmetic. That is the real goal - not eliminating no-shows, which is impossible, but knowing your number well enough that they stop mattering.

Three questions before your next event

• Do you know your actual attendance rate, or are you guessing? If it is a guess, the fix is a spreadsheet, not a strategy.
• Is your yes free to give and free to break? If both, you have chosen a coin flip. Changing either one moves the number.
• How many times do you contact a confirmed attendee before the event? If the answer is one, there is a large, cheap improvement available.

What to take from this

No-shows are not a people problem. They are the predictable output of a system where saying yes is a reflex, breaking a yes is free and silent, and the only reminder arrives after the decision has already been made.

Change the cost of the yes, even by a dollar. Contact people more than once, on a schedule that matches the different reasons they drop out. Make cancelling easy so that absence turns into information. And write down your real ratio, because an organiser who knows their number stops having a no-show problem and starts having a multiplier.

Frequently asked questions

What is a normal no-show rate for a free event?

Eventbrite's data puts free-event no-show rates as high as 50 percent, while a well-run community event with reminders typically lands closer to 30 percent. Paid events can run as low as 10 percent.

Does charging a small fee really reduce no-shows?

Yes, and most of the effect comes from the gap between free and any price at all. Eventbrite notes that even a charge under a dollar raises the perceived value of a ticket, because paying converts a reflex into a decision.

How many reminders should I send before an event?

One organiser with five years of experience used five touches: the announcement, two weeks out, one week out, three days out, and an automatic day-of reminder. Each catches a different reason people drop out.

How much food should I order relative to RSVPs?

A common practice among experienced organisers is 75 to 80 percent of the confirmed count, which matches an observed drop-off of roughly 30 percent at free community events.

Should I make it hard to cancel an RSVP?

No. Friction on cancelling preserves the RSVP but not the attendance, and it turns a useful early no into a silent no-show. An easy cancellation gives you back a seat and an accurate headcount.

Is it a good idea to deliberately overbook an event?

It can be, provided you overbook against a measured attendance rate rather than an optimistic one and have a real plan if everyone arrives. Research on airline bumping shows the cost of turning someone away rises sharply the later they find out.

Why do reminders sent in a group chat work poorly?

Because every reminder costs every member a notification, so considerate organisers send fewer of them - exactly the opposite of what attendance needs. Reminders work better when they point at a stable address rather than restating the details.

What should I actually track to improve attendance?

Your RSVP-to-attendance ratio across at least six events, split between returning attendees and first-timers, with cancellations counted separately from no-shows so an improvement doesn't hide inside a combined number.

Ontaym Editorial Team

Ontaym builds tools for organising real-world gatherings, so the team spends its days on the coordination problems this article describes. Articles are researched against primary sources, reviewed before publication, and revised when the underlying facts change rather than on a schedule.

Give your next event one address people can actually check.