Every organiser of free events knows the maths: the room is set for 200, catering is paid for 200, and 120 walk in. It feels like bad luck. It isn't. It's one of the most predictable numbers in the industry, and one of the most fixable.
The benchmark data is remarkably consistent in direction, if not in exact figures. Check-in data analysed from more than a thousand live events found a median no-show rate of 28% for free events against 17% for paid ones, and whether the event charged admission was the single strongest predictor in the entire dataset. Other benchmark compilations put the gap even wider, with free events routinely losing 40 to 60% of registrants while paid in-person events see 70 to 90% of ticket holders turn up.
Why the spread between datasets? Definitions. Some count only people who registered and silently vanished; others include late cancellations. Some measure invite-only corporate events, which perform better (75 to 85% attendance is a reasonable planning figure), while open free registrations perform worse. But wherever you measure, the direction never changes: remove the payment, and a large slice of your audience treats registration as a bookmark rather than a promise.
The psychology is not mysterious. A free registration costs nothing to make and nothing to break. There is no loss to avert, no sunk cost, no transaction to honour. "I'll grab a spot just in case" is a rational decision for the registrant, and a logistical problem for you.
The experiment that proved the point
In 2026, software company 37signals ran a small but telling test. Its previous free event attracted 84 registrations, of which only 38 people showed up - a show rate of about 45%. For the next edition, the company charged $100, fully refunded at the door to everyone who attended. Registrations dropped to 55. Attendance rose to 50.
Read that again: a third fewer registrations, a third more people in the room. Co-founder Jason Fried attributed the result to simple loss aversion and to the fact that paying for something makes you value it. The money was never the point; the decision to part with it was.
The refundable deposit deserves to be far more widely used than it is, because it works at almost any price point. A study found that adding a $3 refundable deposit to a previously free event cut the no-show rate from 38% to 14%. Three dollars. The mechanism isn't the amount. It's the moment of commitment.
What works, ranked
If your event must stay free - internal events, community sessions, sponsor-funded formats - you still have levers. Here they are, roughly in order of impact.
1. A refundable deposit. The single most effective intervention short of actually charging. Even a token amount, returned at check-in or converted into a drinks voucher, changes the psychology of the registration. Be transparent about the refund mechanism, and make refunding effortless. The goodwill cost of a clunky refund outweighs the attendance gain.
2. Reconfirmation, not just reminders. There's a difference between telling people the event is coming and asking them whether they're still coming. A reconfirmation email a few days out with a visible, guilt-free cancel button has been associated with no-show reductions of around 30%. Counterintuitively, making it easy to cancel is the point: a cancellation you know about is a seat you can reallocate. A no-show is a seat you've wasted.
3. A waitlist that people can see. Scarcity works both ways. If registrants know others are waiting for their spot, cancelling becomes a social good rather than an admission of flakiness. Cap your event registration visibly, run an active waitlist, and promote from it as cancellations come in. This also converts your no-show problem into a demand signal for the next edition.
4. Deliberate overbooking. Airlines have done this for decades; organisers are oddly squeamish about it. If your historical show rate for this event type is 60%, registering 130% of capacity is not reckless. It's arithmetic. The discipline lies in using your own historical data for each event type, not an industry average, and in having a graceful plan for the rare full house: standing room, an overflow screen, and a livestream link at the door.
5. Reminder timing and channel. One email the day before is not a strategy. A useful baseline sequence: confirmation at registration; a value-focused touch a week out (programme, speakers, who else is coming); reconfirmation two to three days out; and a logistics message the morning of - with parking, entrance, and timing - so attending feels frictionless. Where you have consent, a same-day SMS or WhatsApp message outperforms email in open rates by a wide margin.
6. Make the registration slightly harder, but only if it stays free. Data show that events with near-frictionless registration flows (90 to 95% completion) had the lowest attendance, while higher-friction registration flows correlated with more committed attendees. But the same analysis found the effect disappears once payment is involved: among paid events, no-show rates were flat regardless of registration friction. Translation: don't add friction to a paid checkout: you'll lose sales, not flakes. For a free event, a brief motivational question or a required calendar step can serve as a mild commitment filter.
What doesn't work
Guilt-based messaging ("we've reserved catering for you!") reads as manipulative and doesn't move the numbers. Punishing no-shows by banning them from future events punishes your most engaged segment for a behaviour your format invited. And chasing higher registration totals as a success metric is the root cause of error: registrations are a vanity metric for free events. The only number that pays sponsors, fills rooms and justifies budgets is the one at the check-in desk.
The planning takeaway
Whatever mix you choose, the operational rule is the same: forecast attendance from your own historical show rate for that event type, order catering and set the room to the forecast - not the registration count - and treat every intervention above as a way to narrow the gap between the two numbers. A free event will never behave like a paid one. But the difference between a 50% show rate and an 80% one is not luck. It's design.
Source: Photo: iStockPhoto 1153829110
