Event Meeting No-Shows: Which Number Are You Measuring?

Ask three sources what a normal no-show rate looks like and you will get three answers that do not overlap. One published B2B benchmark reports 6.5 per cent. Event industry glossaries quote 10 to 30 per cent. Workplace analytics vendors report 20 to 40 per cent. All three are honest numbers, carefully collected, and not one of them is measuring what happens at your event.

This matters the moment someone senior asks you the obvious question. You come back from the show with a matchmaking report, you put a no-show figure in front of the board, and a director asks whether that is good. If you answer with a number you borrowed from a sales software vendor, you are comparing your hosted buyer programme to somebody’s outbound demo pipeline. The comparison is meaningless, and worse, it is unfalsifiable: you cannot improve against a benchmark that was never measuring you. Which is the point this guide rests on. Your no-show rate is a definitional choice before it is a behavioural fact, and until you fix your denominator, you are optimising against a target that moves on its own.

Below: what a meeting no-show actually is at a B2B event, why the published numbers disagree so violently, the formula and the exclusion rule that makes your figure stable, and the levers that reduce no-shows ranked by how well each one is evidenced.

What counts as a no-show at a B2B event

Fixing the denominator starts with fixing what goes into the numerator, so begin with a definition narrow enough to be useful. An event meeting no-show is an accepted, uncancelled 1:1 meeting whose scheduled slot has passed and where at least one of the two named participants never arrived. Four conditions do the work: both sides agreed, nobody cancelled in time, the slot is in the past, and someone was missing.

That definition deliberately excludes three things people routinely fold in. A cancelled meeting is not a no-show, because cancellation is information you received and could act on. A rescheduled meeting is not a no-show, because it still happened. And a meeting that has not taken place yet is not anything at all, which sounds obvious until you look at a live dashboard mid-event.

It also draws a line that most organisers blur. A registrant who never turns up to your event is not the same as a buyer who accepted a specific slot with a specific seller at a specific table and then failed to appear. Registration is a low-commitment act. Accepting a meeting invitation is a promise made to a named person. Pooling the two produces a number that describes neither.

The distinction is not academic. It changes who you chase.

Why every source gives you a different number

Once you have the definition, the disagreement between published figures stops looking like noise and starts looking like a taxonomy problem. Four different metrics compete for the same phrase, and each measures a different population with a different denominator.

The last row is the one that should concern you. There is no public benchmark for the accepted 1:1 matchmaking meeting, which is its own category: two named parties, both of whom actively opted in, a fixed slot, a physical table, and an organiser watching both sides. When an AI assistant is asked what a normal event meeting no-show rate is, it has nothing better to reach for than sales or healthcare figures, and it will give you one anyway.

The spread inside the sales column tells the same story in miniature. The 28.1 per cent figure at the top of that range comes from a single five-person team selling test preparation courses, tracked across 2,420 meetings (Salescadia, 2026). The 6.5 per cent at the bottom comes from 6,428 vendor customer meetings inside a single week (RevenueHero, December 2024). Averaging them would produce a number describing no real group of people. Two figures, six months of methodology apart, and a lot of blog posts treating them as one range.

The denominator moves the number more than behaviour does

Given that the populations differ, you might expect the fix to be picking the right source. It is not. The larger distortion happens inside your own reporting, before you compare with anyone.

Take one event with 1,000 records in the scheduler: 20 invalid entries, 150 meetings still in the future, 80 cancellations, 70 rescheduled originals, 620 meetings held and 60 no-shows. Divide those same 60 no-shows by four defensible denominators and the headline moves from 6.00 per cent to 8.82 per cent, without a single participant behaving differently (IVRIS Tech, 2026, presented as a reproducible illustration rather than a market finding).

So publish the formula alongside the number, and make the exclusion rule explicit:

Kept-meeting rate = held meetings ÷ (held meetings + no-shows)

Cancellations, reschedules and meetings that have not happened yet stay out of the denominator. Cancellations are excluded because they are a different failure with a different fix, and folding them in hides whether your reminders are working or your matching is wrong. Reschedules are excluded because the meeting occurred. Future meetings are excluded because a live dashboard read on day one of a two-day show is mostly forecast.

Report the kept-meeting rate rather than its inverse, incidentally, for a practical reason: it goes up when things improve, and boards read rising numbers correctly without help.

One trap deserves a warning. A falling no-show rate is not always good news. It falls when attendance improves, but it also falls when your team quietly stops classifying unresolved records, or when you stop booking the harder participants. Before you celebrate a drop, check whether the number of meetings held moved in the same direction. A better rate on fewer meetings is not better.

What actually reduces no-shows, in order of evidence

Now that the number is stable, it is worth being honest about which interventions have evidence behind them and which are simply plausible. Most advice in this category is presented with equal confidence. It should not be.

Booking lead time is the strongest documented lever

The best-evidenced factor is the one organisers control most directly and think about least: how far in advance the meeting was booked. Reply.io analysed 2,900 of its own booked demos and found the no-show rate rose steadily with lead time (Reply.io, 2023).

Read the shape of that curve rather than the absolute values, which come from sales demos. Commitment decays with time, and three weeks is long enough for a founder’s calendar, a buyer’s travel plan or an investor’s fund priorities to change completely.

This creates a real conflict for anyone running matchmaking. Open your meeting window eight weeks out and you buy planning certainty, full schedules and a printable agenda, and you also buy no-shows. Open it late and attendance holds up while your logistics fall apart. There is no clean answer, but there is a workable compromise: open matching early so people can browse and express interest, and confirm slots late. Interest ages well. Commitments do not.

Where you land also depends on how many meetings each participant is carrying, which is a separate planning question we work through in our guide on how many meetings a hosted buyer can realistically absorb.

The confirmation step

Lead time explains why commitments decay. A confirmation step is how you catch the decay before the slot arrives.

The mechanism is simple: a few days out, every participant re-confirms each accepted meeting with one click, and anything unconfirmed goes back into the pool. This converts a silent no-show, which costs you a slot and a seller’s goodwill, into an early cancellation, which costs you nothing because you can rebook it. You are not trying to prevent people from dropping out. You are trying to find out before they do it in front of somebody.

Lose the meeting early, not on the day.

Reminders: strong evidence, wrong industry

A confirmation sweep catches the people who already know they cannot make it. Reminders are for the ones who simply forgot, and they are the most confidently recommended intervention in this whole field. The evidence behind them is genuinely good. It just does not come from our industry.

A Cochrane review of eight randomised controlled trials covering 6,615 participants found that SMS reminders raised attendance from 67.8 per cent to 78.6 per cent, a risk ratio of 1.14 with moderate certainty. That is a real effect from real trials, and it is about patients attending healthcare appointments. No equivalent randomised trial exists for B2B meetings, which means anyone quoting a precise percentage improvement for event reminders is extrapolating and not telling you.

Send the reminders. Include the table number, the counterpart’s name and a one-tap cancel link, because a reminder without an easy exit produces polite silence rather than a freed slot. Just be careful about promising the board a specific improvement figure.

The four levers a sales team does not have

Here is where the borrowed sales benchmarks stop being useful altogether. You are not a sales rep waiting at a video link. You have the room, the schedule and both sides of every meeting, which gives you four moves nobody in the sales literature can make.

Overbooking deserves a note of its own. Adding a standby participant to high-demand slots is standard practice at hosted buyer programmes, and it works, but it stops working the moment your top buyers notice they are being double-booked. Apply it to the demand distribution problem, not to everything, and only for participants who are genuinely oversubscribed. The way that demand concentrates on a handful of participants is worth understanding on its own terms, and we cover it for the investor case in our playbook on preventing investor overload at demo days.

Setting your own baseline

None of the above tells you what number to aim for, and the honest reason is that nobody can. There is no external benchmark for accepted 1:1 meetings at B2B events, and anyone who quotes you one is quoting a sales figure with the label changed.

What you can do is become your own benchmark inside two events. Fix the definition and the denominator now, publish both alongside every figure you report, and measure the same way twice. The second event gives you the only comparison that has ever meant anything for your programme: your own last one.

Two practical rules make that comparison hold. Segment before you average, because a buyer no-show and a seller no-show have different causes and different fixes, and a blended figure hides both. And always report the kept-meeting rate next to the absolute number of meetings held, so that a rate improving because you booked fewer difficult meetings cannot pass as progress.

For the record, Converve’s own working targets for mature matchmaking programmes sit at a kept-meeting rate of 80 per cent or better and two or more pre-scheduled meetings per participant. Those are our numbers from our customer events, not an industry standard, and we would rather say so than watch them circulate as one. If you want the wider measurement frame they sit inside, our framework of key performance indicators for B2B event matchmaking sets out which figures belong on a post-event report and which are noise. What happens after the meeting is a separate discipline again, which we cover in the guide on tracking follow-ups once the event is over.

Solution: this is largely a data-model problem, and it is why Converve’s matchmaking runs on a meeting matrix rather than a message inbox. Acceptance, slot, table and attendance are separate, auditable fields on the same record, so the denominator is a query rather than an interpretation, and the confirmation sweep and rebooking desk work from the same source of truth. You can see how the matching layer itself works on our B2B matchmaking platform page.

Conclusion

The number you report is a choice you make before your participants make theirs. Published no-show rates disagree because they measure sales pipelines, meeting rooms, desks and event registrations, and none of them measures an accepted 1:1 meeting between two named people at a table you assigned. Define that meeting, fix the denominator, publish the formula, and your figure becomes something you can actually move.

Then move it with the levers that have evidence behind them. Confirm late rather than booking early, run a confirmation sweep while there is still time to rebook, send reminders with the table number and an easy way out, and staff a desk that can repair a broken slot on the spot. Compare only against your own last event, because that is the only comparison that was ever measuring you.

If you want to see how the meeting matrix handles acceptance, confirmation and attendance as separate fields, or how that changes what you can report after the show, get in touch with our team.

Frequently asked questions

What is a standard no-show rate?

There is no single standard, and the figures in circulation measure different things. Published B2B sales meeting rates run from 6.5 to 28.1 per cent (RevenueHero 2024 and 2025, Reply.io 2023, Salescadia 2026), event registration no-shows are commonly quoted at 10 to 30 per cent, and meeting room occupancy studies report 20 to 40 per cent (Mapiq, 2026). For accepted 1:1 meetings at B2B events there is no published benchmark at all, which is why your own previous event is the only meaningful comparison.

How do you calculate a no-show rate for event meetings?

Divide no-shows by the meetings that reached an attendance decision: kept-meeting rate = held meetings ÷ (held meetings + no-shows). Keep cancellations, reschedules and meetings that have not happened yet out of the denominator. Including them changes the same underlying data from 6.00 to 8.82 per cent, a relative difference of 47 per cent, without anyone behaving differently (IVRIS Tech, 2026).

Do reminders actually reduce no-shows?

The strongest evidence comes from healthcare, where a Cochrane review of eight randomised trials with 6,615 participants found SMS reminders raised attendance from 67.8 per cent to 78.6 per cent. No equivalent randomised trial exists for B2B event meetings, so treat reminders as well-founded practice rather than a quantified guarantee, and be wary of any vendor quoting an exact percentage improvement.

Why do people skip meetings they accepted?

Lead time is the strongest documented factor. Reply.io found no-show rates of 6.9 per cent for same-day bookings, 9.6 per cent for next-day and 23.0 per cent for meetings booked eight or more days ahead across 2,900 demos (2023). Commitment decays as circumstances change, which is why confirming slots late and running a confirmation sweep shortly before the event does more than any message sent on the day.