Calculating an exhibitor compliance rate that survives an audit before move in

Two weeks before a show opens, somebody senior asks what the exhibitor compliance rate is. The answer comes back as 87 per cent. Nobody asks which date it was measured on, what counts as compliant, or which exhibitors are in the denominator, and the number goes into a risk report where it will sit for a year.

Then an insurer, an auditor or a venue asks the same question about last year's edition, and the 87 per cent cannot be rebuilt from the data. The documents are all still there. The date they were true as at is gone.

Every word in the definition does work

Compliance rate is compliant exhibitors divided by contracted exhibitors at a stated cutoff date. Four things in that sentence need pinning down, and each one is worth several points of headline percentage.

• Compliant means every published requirement met at the same moment, and the one requirement somebody happened to check does not stand in for the rest. A show asking for a certificate of insurance, a tax form and a stand design approval has an exhibitor who is compliant only when all three hold together.
• Exhibitors needs an entity rule. Count contracting companies. A parent with three brand stands under one contract is one exhibitor, and counting stands instead will move the rate by two or three points on any show with a serious corporate presence.
• Contracted sets the denominator, and it should be every company with a countersigned contract as at the cutoff, including the ones you already expect to fail. Excluding known problems is how a compliance rate becomes a negotiation.
• At a stated cutoff date is the part that gets dropped, and it is the part that decides the answer.

Which date should the rate be measured on?

Take a show with 366 contracted exhibitors and four honest measurement dates.

At the published document deadline, 214 exhibitors met every requirement. That is 58.5 per cent.

Seven days before the first move-in day, 318 did. That is 86.9 per cent.

On the morning of move-in, 349 did, which is 95.4 per cent.

By the time the doors opened to buyers, after eleven exceptions had been granted at the service desk, 358 did, which is 97.8 per cent.

Same 366 companies, same file, four rates spanning 39 percentage points. Anybody quoting one of them without the date attached is quoting a number that cannot be checked, and anybody comparing this show against another one is comparing measurement policies.

The shape between those points is worth plotting once. Compliance against days to move-in is an S curve on almost every show, flat for months, steep in the last fortnight, flattening again against a ceiling somewhere below 100 per cent. Measuring anywhere on the steep part gives you a number that moves several points a day, which is why two people pulling the same report 48 hours apart can produce an argument out of nothing.

My own preference is the seven day figure as the operating measure. It leaves a working week to do something about the gap, which the move-in morning figure does not, and it falls late enough that the remaining failures are real rather than administrative. Publish the deadline figure next to it as an early warning, and publish the final figure after the show as the closing position.

The deadline is a policy choice

Document deadlines sit wherever the show decided, and the distance between the deadline and the show varies enormously.

NAMM's published exhibitor insurance requirements for its 2027 show, issued in 2026, ask exhibitors to send their certificate by 13 November 2026 for a required coverage period running 22 January to 2 February 2027. The deadline sits roughly 70 days before the coverage period even begins, and about 76 days before the exhibit hall opens on 28 January.

The American Urological Association set 21 March 2025 as the certificate deadline for its 2025 annual meeting, with coverage required from the first day of the exhibitor move-in period to the last day of move-out.

Both are reasonable and they produce very different compliance curves. A show measuring at its own deadline is reporting the position 70 days out, when brokers are still working and nothing is at risk. A show measuring at move-in is reporting the position when the only remaining lever is refusing entry. Neither is more correct, and a portfolio comparing shows on compliance rate without normalising the measurement date is producing a chart about deadline policy.

Can you reproduce the number six months later?

This is the question that separates a compliance rate from a compliance number, and most systems fail it.

The failure is architectural. A document record that carries a status field and an updated timestamp tells you what is true now. It cannot tell you what was true on 4 March, because the row has been overwritten since. Ask that system for last year's compliance rate at the cutoff and it will either refuse or, worse, answer using today's statuses against last year's date, which produces a figure far higher than anything anyone saw at the time.

What an audit needs is a point in time reconstruction, and there are two ways to get one.

The cheap way is a snapshot. On the cutoff date, write the full compliance state of every exhibitor to an immutable table with the date stamped on it. It costs one scheduled job and it answers the audit question forever. It also lets you compare editions honestly, since each snapshot was taken on the same relative day.

The better way is an event log. Record each state change as its own row with a valid from timestamp, and any historical position becomes a query rather than a hope. That is more work upfront and it survives every question anybody asks later, including ones you have not thought of yet.

If you have neither today, take the snapshot. It is a small job and it stops the clock on the damage.

What counts as compliant needs a rule per requirement

A compliance rate is only as defensible as the checks underneath it, and this is where a rate quietly inflates.

A coordinator marking a certificate as received is doing something different from a coordinator confirming that the limits meet the requirement, the required parties appear, and the policy covers the full tenancy. The first takes ten seconds, the second takes a few minutes, and both produce the same green tick in most systems. A show whose insurance check is a received flag will report a higher compliance rate than a show doing the work, and will find out the difference during an incident.

So write the check as a list of testable conditions per document, and record which ones passed. The six fields on a certificate that decide access are the place to start for insurance, and the policy dates need testing against your own move-out day rather than your show open, which is its own piece of expiry tracking.

Recording the passing conditions has a second benefit. When the rate moves between editions, you can see which condition moved rather than guessing, and in my experience it is almost always one condition doing all the work.

Exceptions, and keeping them out of the numerator

Every show grants exceptions at the service desk. An exhibitor arrives with a certificate that names the venue and misses the city, the show is opening in four hours, and somebody decides to let them build.

Those decisions are legitimate and somebody has to make them at four in the afternoon with a truck waiting. An exception is a documented acceptance of a known gap, and it belongs in its own count with a named approver and a reason code, sitting outside the numerator.

The reason code is the part worth insisting on, because it turns a pile of individual judgement calls into a list you can act on before the next edition. Eleven exceptions with eleven reasons is bad luck. Eleven exceptions where nine of them say the certificate named the venue and missed the city is a requirement that your request wording has failed to communicate, and that is fixable in a morning.

On our 366 exhibitor show, the closing figure of 358 included eleven exceptions. The honest closing report says 347 exhibitors met the requirement, 11 were admitted under a named exception, 5 withdrew before move-in and 3 were refused entry. Those four numbers are more useful than 97.8 per cent, because the exception count is the thing that grows quietly year on year when nobody has been asked to report it.

Where this stops

A compliance rate measures documents, and documents are a proxy for risk. An exhibitor with a perfect certificate can still rig a truss badly. A rate of 95 per cent tells you about paperwork discipline and says nothing about what happens on the floor, which is why safety teams keep their own records and rightly refuse to be measured on this number.

The second limit is the entity problem in the denominator. Co-exhibitors sharing a stand, national pavilion participants whose space is bought by a trade body, and sponsors with a small presence all sit awkwardly in a contracted exhibitor count, and different shows resolve them differently. Whatever you decide, apply it to both sides of the fraction and write it down, because this single choice moves the rate more than most process improvements will.

The third is that a high rate at a late cutoff can be produced by weak requirements. A show asking for less will always look more compliant than a show asking for more, and the two rates are not comparable in either direction. Read the rate next to how long exhibitors take to get there and next to the requirement itself.

Pick your cutoff this week and write it into the onboarding calendar as a dated job: on that morning, snapshot every exhibitor's compliance state to a table nobody edits. Do it once and next year's audit question has an answer that takes a minute instead of a fortnight.