Show week renewal signals you can only collect before teardown

By eleven on the Sunday the aisles are carpet tape and forklifts, and the only record that stand 4104 was staffed by two people who arrived at half past ten and left before the afternoon peak is in the memory of whichever floor manager happened to walk past.

That observation was worth more than most things in your CRM. It is gone. Not archived somewhere awkward, gone, because nobody wrote it down and the show does not happen again for a year.

Show week renewal signals are a category of evidence with no equivalent at any other point in the cycle. They are generated by the exhibitor's own behaviour, under conditions you control, on premises you own, and almost all of them are discarded within seventy-two hours of the halls closing.

Which signals only exist while the hall is open?

The ones worth building are the ones already being generated by systems you are running anyway.

Booth staff hours on site. Exhibitor badges scan at hall entry. Every scan is a timestamp attached to a person attached to an account. Aggregate to stand-hours per account per day and you have a measure of how much of their own payroll an exhibitor is spending on your show, which is a considerably better proxy for commitment than anything they say in a survey.

Whether the decision maker came. Your exhibitor contract has a signatory. Your account record has a primary commercial contact. Did either of them scan into the building at any point during the open days. It is a binary and it is the most quietly informative field on this list.

Stand visitor scans. Exhibitors running lead retrieval generate a scan count. Normalised by space and by open hours, it measures the outcome the exhibitor is buying. An exhibitor holding 54 square metres and taking 31 scans over three days is having a bad show whatever they say at the closing drinks.

Complaints and escalations logged with an account ID attached. Most operations teams log issues by stand number or by hall, which makes them unjoinable to the commercial record afterwards. Attaching the account ID at the point of logging costs nothing and turns a week of operational noise into a renewal input. What to do with the resulting counts is a separate problem, because raw volume punishes your biggest exhibitors, and normalising service tickets has its own treatment (G16).

Booth staff hours, worked

Take an exhibitor holding 54 square metres at a three-day show with seven open hours a day.

Last edition they registered eight exhibitor badges. Entry scans put their team in the building from roughly half nine to roughly half four across all three days, so call it seven hours each. Eight staff multiplied by three days multiplied by seven hours is 168 stand-hours. Divided by their 54 square metres, that is 3.11 stand-hours per square metre.

This edition they registered three badges. The entry and exit scans show arrival around half ten and departure around half three, five hours a day. Three staff multiplied by three days multiplied by five hours is 45 stand-hours, which is 0.83 per square metre.

Their commitment has fallen to 27 per cent of what it was, on the same footprint, paying the same money. Nobody has said anything to anybody. The stand looks the same in the photographs.

The headcount alone would have told you something: three registered against eight is visible on the Tuesday morning of a Wednesday-open show, because badge collection happens during build. The hours multiply the problem by a factor you cannot see from the headcount, and the two together give you a number that fits in one line of a rep's briefing.

The comparison that matters is against the account's own history. Measured against the show median, a big spender that halved its team still sits comfortably above the middle of the floor and never surfaces. A consultancy that always sends two people is fine at two people. An account that has sent eight for four editions and sends three this time has changed, and change against self is the reading with the least noise in it.

Whether the decision maker came at all

This is the signal I would build first if I could only build one, because it is nearly free and it is very hard to explain away.

Exhibiting is expensive and senior people attend the shows they intend to keep buying. When the person who signs the contract stops appearing on the floor of a show their company is exhibiting at, the account has been delegated, and delegation is the step before cancellation more often than it is the step before growth.

The build is a join. Take the signatory and primary contact from the account record, match them against the badge scan file for the open days, and produce a flag. The awkward part is the match, because the contract signatory is frequently registered under a different email from the one on the contract, and the same person may appear in your registration file three times.

Cross-referencing those records properly is its own discipline and there is a whole method behind it. For this signal you can get away with a crude version: surname plus company, checked by hand for the top two hundred accounts by value. Two hundred manual checks is a day of somebody's time and it produces a field that will still be true in eleven months.

How do you capture this without adding work on site?

Nobody on site has spare capacity, so any capture design that needs a human to do something new will fail.

Everything above comes from three files that already exist: the exhibitor badge registration export, the hall entry scan log, and the operations issue log. The work is joining them to the account record and writing them somewhere durable, and it is a data engineering task that can be done entirely after the show closes provided the raw files are preserved.

That is the actual failure mode. Access control logs get rotated. The scanner vendor's portal expires. The operations log lives in a spreadsheet on somebody's laptop and gets superseded. Preserving the raw files is a fifteen-minute job on the Sunday and it is the difference between having this data and not. These signals also sit at the short end of the lead time ranking that G14 builds, roughly thirty days of warning, so they belong in the renewal campaign and feed the wider renewal intelligence record rather than the annual planning session.

Explori and UFI, working with SISO, ran the Global Recovery Insights study in 2020, gathering more than 9,000 responses from visitors and exhibitors across 30 countries in two parts published in October and December of that year. What they led on was a strong exhibitor preference for live events in every area, networking most of all, alongside a view that digital events were not at that point delivering a good return. The operational reading is that the show floor remains the place where exhibitor intent gets expressed, so the show floor is where the measurement should happen. The earlier UFI and Explori Global Exhibitor Insights report, published in November 2017 and built from post-show research at 1,040 trade shows across more than 40 countries, found exhibitor advocacy low everywhere, with only 25 per cent of shows carrying a positive net promoter score. A base that lukewarm is one where a fortnight of behavioural evidence is worth collecting properly.

Acting on it during the show, which is the point

Collecting these for the post-show model is worth doing. Acting on them on the Tuesday is worth considerably more, because the exhibitor is standing in your building and their team is bored.

The workable version is a short list produced each morning from the previous day's scans. Accounts whose staff-hours are down more than half against their own prior edition, accounts where the signatory has not scanned in by day two, and accounts whose visitor scans per square metre sit in the bottom decile. Twelve to twenty names on a good-sized show, which is a morning's work for the sales team on the floor. This list is not the rebooking desk queue: what an onsite rebooking rate measures and hides is G1's subject, and the two get conflated on most floors.

The conversation is easier during the show than it will ever be again. You are asking how it is going, in person, with the answer visible around you, before anyone has written a report internally that hardens into a position.

Where this stops

Every signal here depends on scan coverage, and scan coverage is not uniform across your exhibitor base. Exhibitors with their own contractor entrances, exhibitors on stands adjacent to a fire door that gets propped, and large exhibitors who negotiate loose badge terms will all under-scan. Their measured stand-hours will look low for reasons that have nothing to do with commitment.

The correction is to compare each account only against itself, which handles a constant bias but does nothing about a bias that changed this year because you moved the entrance. If you changed the hall layout, changed access control vendors, or changed the badge policy, last year's figures are not comparable to this year's and you should say so in the field rather than let a rep discover it in front of a customer.

Lead retrieval scans have a worse version of the same problem, because exhibitors choose whether to use the system at all and a scan count of zero often means they used their own app. Treat missing as missing and not as zero.

Before your next show closes, ask whoever owns access control to export the raw entry scan log with timestamps and badge IDs, and put a copy somewhere permanent alongside the exhibitor badge registration file. That single act preserves the option to build all of this later, and it takes one email sent this week.