Rebooking cancellation rate is the number your gross figure leaves out

The Friday board pack said 402 exhibitors rebooked on site. In March the floorplan shows 371 of them holding space. Somewhere in between, 31 companies left, and when the show director asks when that happened and which accounts they were, the answer takes a day of work because nobody kept a record of the exits. The contract table was overwritten. The 402 became 371 by subtraction, quietly, on a status field with no history behind it.

A rebooking cancellation rate fixes this, and it is one new number in the reporting pack.

Signature to cash, and the exits in between

There are three moments where a signed rebooking can disappear before it becomes revenue, and they have different causes and different cures.

A withdrawal inside a stated cooling-off period, where the contract terms give the exhibitor a defined window to cancel without penalty. A cancellation after the window but before the first payment falls due, usually accompanied by a request to waive the deposit. A non-payment, where nobody cancels anything and the invoice simply goes unpaid until you write the space back onto the floorplan.

The third is the one that corrupts your data, because there is no cancellation event to record. The account remains marked as signed in the CRM until somebody notices in February and changes the status. By then the timestamp records when your operations team caught it. The exhibitor decided some weeks earlier, and nothing in your data marks that moment.

Fix that first. Any account whose first payment is more than 30 days past due should generate a dated at-risk record automatically, and if it later comes off the floorplan the cancellation date is the payment default date. Without this, every cancellation figure you compute is a measurement of your own admin cycle. Ageing debt is a churn signal in its own right, well beyond this one window.

Gross and net on one cohort

Take the 402 onsite signings from our industrial show, tracked to the first payment deadline. 31 are gone by then: 14 that never cleared a deposit, which is where the onsite funnel stops looking, and 17 that paid one and withdrew afterwards. Net rebooking from show week is 371.

The give-back rate is 31 divided by 402, which is 7.7 per cent. Against the 640 exhibiting companies present, the gross onsite rebooking rate was 62.8 per cent and net onsite rebooking is 371 over 640, or 58.0 per cent. Four and eight tenths of a point separate the number that went in the board pack from the number that describes the floorplan.

Weight it by money and the picture shifts again. At an illustrative average contract value of 6,500, the 402 signings represent 2,613,000. The 31 cancellations averaged 5,600 each, which is 173,600, so the give-back is 6.6 per cent by value against 7.7 per cent by count. Smaller accounts left slightly more often than larger ones, which is the common pattern and worth confirming on your own file rather than assuming, because the reverse pattern is a much more serious problem and looks almost identical in a count-based report.

Report both figures every cycle, gross and net, with the give-back rate between them. Reporting only gross overstates your book. Reporting only net delays the number by a quarter, since you cannot compute it until the payment deadline has passed, and the sales team needs something on the Friday.

Where do cancellations cluster?

An aggregate give-back rate of 7.7 per cent is a starting point. The useful version splits it.

Split our 402 by how the signature was obtained. 96 signed under a deadline offer that expired at teardown, and 306 signed on standard terms. Of the 96 deadline signings, 19 cancelled, a give-back of 19.8 per cent. Of the 306 standard signings, 12 cancelled, a give-back of 3.9 per cent. The deadline offer produced 96 signatures and 77 survivors.

That is a five-fold difference in cancellation rate between two groups signing the same contract in the same hall, and it changes how you read the headline. A show that leans hard on a teardown deadline will post a strong gross figure and hand a chunk of it back before Christmas, and its rebooking pace will look better than a comparable show that does not, right up until the payment deadline.

I would not conclude from this that deadline offers are wasteful. The right comparison is what those 96 accounts would have done with no offer, and a cancellation rate cannot answer that. What it does establish is that a gross figure containing a large deadline cohort is a lower-quality number than the same figure containing few, and that the two should never be compared without the split visible.

Also split by tenure and by whether the signing decision maker was the person on the stand. Cancellations concentrate where a stand manager signed something a marketing director had not budgeted, and that pattern is visible in the data if you record who signed.

The cost pressure behind the second thoughts

Some of the give-back is buyer's remorse about a number that keeps rising.

The Exhibitor Advocate, in its 2025 annual survey of exhibition rates, audited independently by EVOLIO Marketing, analysed 224 publicly available exhibitor manuals and rate forms across 23 major US cities and found four years of compounding increases: material handling base rates up 21.3 per cent since 2022, secondary material handling rates up 26.4 per cent, electrical overtime labour up 41.2 per cent, and installation labour up between 12.7 and 16.7 per cent depending on the shift. Companion research from the same two organisations in 2025 found 80 per cent of exhibitors citing cost management as their top challenge and 55 per cent saying increased costs outweigh the value of the investment at some events.

None of those increases sit in your space rate. All of them sit in the total an exhibitor has to sign off internally, and the gap between signing a space contract in June and receiving the services estimate in October is exactly where a rebooking cancellation is born.

There is a matching signal in the industry's own index. The CEIR Index reported through IAEE in 2025 put second-quarter exhibitor counts 8.8 per cent below their 2019 level and real revenues 15.6 per cent below, so revenue has fallen further than participation across the tracked event population. A show whose give-back rate is rising should read those two together before concluding the problem is its own sales process.

Making the number reproducible

A cancellation rate needs three fixed decisions or it will not survive contact with a second analyst.

The measurement date. I would use the first payment deadline stated in the contract, because it is a term both sides agreed and it exists in the document. A calendar date such as 31 January works too, and is easier to run across a portfolio with different contract terms, but it will cut some shows before their deadline and some after.

The cancellation event definition. Withdrawal, cancellation and payment default all count, and each needs its own reason code so you can report the split. A downsize is not a cancellation and belongs in the space measurement, which is its own topic.

The treatment of replacements. If a cancelled 36 square metre stand is resold to a different exhibitor in April, the cancellation still happened and the give-back rate still counts it. Netting resales against cancellations produces a number that measures your sales team's recovery rather than your exhibitors' behaviour, and you want both, separately.

What should the give-back rate change?

A give-back rate is useless if it only appears in a report, and it is one of the few renewal intelligence measures that names the fortnight in which to act.

At 7.7 per cent overall and 19.8 per cent in the deadline cohort, the operational response is a structured contact with every deadline-cohort account in the fortnight after the show, before the cooling-off window closes, with the specific purpose of confirming the internal budget holder knows about the commitment. That is roughly 96 calls, which is under two weeks for one person.

The second response is a forecasting one. Finance should be forecasting from net, and the give-back rate is what converts one to the other. If your last three editions gave back 7.4, 8.1 and 7.7 per cent, then applying about 7.7 per cent to this year's gross gives finance a defensible planning number in November instead of a surprise in March.

Where this stops

The give-back rate measures accounts that left visibly, and it will miss the ones that stay on the floorplan and shrink. An exhibitor who signs for 72 square metres in the hall and renegotiates to 36 in January never appears in your cancellation count, and the revenue loss can easily exceed several outright cancellations.

The rate is also unstable at small counts. 31 cancellations from 402 gives a rate whose confidence interval spans roughly two points either side, so a move from 7.7 to 9.1 per cent between editions is not evidence of anything on its own. Pool editions, or compare cohorts within one edition where the split is large enough to carry the comparison.

The deeper limit is that a cancellation is the end of a decision that was made weeks earlier, somewhere you were not present. By the time it lands in your data the exhibitor has already had the internal conversation, priced the services, and lost the argument. The rate tells you how often that happens and which cohorts it happens to, which is genuinely useful for allocating attention. It will never tell you what was said in the room.

Take your last edition's onsite signings, join them to the payment ledger, and count how many were cancelled or unpaid at the first payment deadline. Then split that count by whether the signature came under a deadline offer. If the two rates differ by more than a couple of points, your gross rebooking figure and your net one are measuring different sales processes and both belong on the slide.