What belongs in a post show report and what gets cut

The pack that came round after one show closed ran to ninety-four slides. Slide six carried an attendance figure. Slide sixty-one carried a different one, about eleven hundred higher, under a chart about session rooms. Nobody had caught it, because nobody had read both slides in the same sitting, and the show director had already quoted the first number in an email to the board.

Post-show reports get long for a reason that has nothing to do with rigour. Every team that contributed data wants its contribution visible in the pack, and the person assembling it has no authority to refuse anybody. The report grows until it is a filing cabinet with a cover page, and the handful of numbers that decide whether the show is healthy sit somewhere in the middle of it.

The way out is a decision about order, taken before anyone opens a spreadsheet.

What questions should a post show report answer, and in what order?

A post-show report answers four questions, and the sequence is the whole design. Who came. Who exhibited. What sold. What people said.

That order is not arbitrary. Attendance is the input every other number depends on, so it goes first and it goes on its own. Exhibitor count comes second because it is the second population, and because a reader who has just absorbed the attendance figure will immediately want to know how many stands those people had to visit. Revenue comes third because it is the consequence. Feedback comes fourth because it is the only section that explains anything, and it reads as excuses if it arrives before the numbers it is explaining.

The CEIR Index, published by CEIR through IAEE, measures the United States business-to-business exhibition sector on exactly four things: net square feet of exhibit space sold, professional attendance, number of exhibiting companies, and gross revenue. The 2026 edition forecasts the Total Index to grow 2.1 per cent for the year. Whatever you think of the index as a benchmark, the choice of four components is a useful discipline, because it is the smallest set that describes a show as a business.

The evidence that all four are needed is that they move apart. CEIR's Q2 2025 Index put attendees 3.7 per cent below Q2 2019, net square feet 4.9 per cent below, exhibiting companies 8.8 per cent below, and real revenues 15.6 per cent below. Four metrics, four different answers to the same question about recovery, spanning nearly twelve percentage points. A report carrying only one of them describes a different show depending on which one you picked.

The front page is four numbers and their prior year

Here is the front page for a mid-sized show with 42,000 net square feet sold. Four rows, each with this year, last year, the absolute change, and the percentage.

Net square feet sold: 42,000 against 40,600, up 1,400, up 3.4 per cent. Professional attendance: 24,300 against 25,200, down 900, down 3.6 per cent. Exhibiting companies: 1,090 against 1,062, up 28, up 2.6 per cent. Gross revenue: 11,420,000 against 10,930,000, up 490,000, up 4.5 per cent.

Every number on that page is a count with a defined population behind it, and every percentage sits next to the two figures it was computed from, so a reader can check the division without leaving the page. That second property matters more than it sounds. People check arithmetic they can see and accept arithmetic they cannot.

One derived line earns its place underneath. Attendance per exhibiting company is 24,300 divided by 1,090, which is 22.3, against 25,200 divided by 1,062, which is 23.7. Down 1.4 visitors per stand, or 5.9 per cent. That is the number every exhibitor will compute for themselves within a week of getting their scan data, and it is better coming from you first. It is also the line that explains why a show can grow on three metrics and still feel worse on the floor. The one-page version a chief executive reads is a different artefact with its own ranking rule, and the post event report executive summary sets that out.

The footprint line stops there. Working out which accounts and which product categories drove that 1,400 square feet is a separate exercise with its own method, and tracking footprint change year over year covers it.

Section two exists to defend section one

Behind each front-page number sits a page that shows where it came from. Not a chart. A derivation. That page is the plainest piece of attendee analytics in the pack, and the one most often missing.

For attendance that means the population, the filters applied, the time window, and the source system, followed by the ledger that gets you from registrations to the reported figure. D9 sets out that ledger as a numbered reconciliation, and it belongs in section two of the report rather than in an appendix nobody opens, because the reconciliation is the answer to the only hard question anyone will ask.

For exhibiting companies it means saying which of three plausible counts you published, since contracted accounts, directory listings and unique parent companies are different populations and D8 works through the gap between them. For revenue it means the accounting basis and the cut-off date. For net square feet it means whether the figure is paid space, and whether feature areas are inside or outside it.

This section is boring by design and it is the part that survives contact with a sceptical reader. When somebody in the room says the attendance number looks high, the argument ends in about ninety seconds if the derivation is on the next page, and runs for two weeks if it is not.

What people said goes last and gets a real method

The feedback section is where post-show reports go slack. It usually contains a satisfaction percentage with no denominator, three quotes chosen because they were nice, and a word cloud.

The 2025 Channel Insights Report, produced by Explori with UFI and supported by SISO, draws on more than 3,000 events surveyed between 2017 and 2025 and tracks exhibitor sentiment on four measures: overall satisfaction, likelihood of return, net promoter score, and importance. Average exhibitor net promoter score across that set is up 27 points since 2019. The useful part for a report author is the shape of it. Four named measures, tracked consistently, with movement stated against a fixed baseline year.

Copy that structure. Pick your measures once, report them the same way every edition, and state the response count and the response rate next to the score. A satisfaction figure without a denominator is not a measurement, and the person who wrote it usually knows that.

What gets cut, and where does it go?

Three categories of material should leave the pack, and none of them should be deleted.

Anything with no owner. If no named person will defend the number in a meeting, it does not go in. Orphan slides are how a second attendance figure ends up on slide sixty-one.

Anything the reader cannot act on at the altitude they are reading. Session-level dwell time is real data and it belongs to the content team's own review, not the show report. Individual exhibitor scan counts belong in the exhibitor pack, which D35 covers.

Anything that repeats a number already on the front page in a different unit. A pie chart of attendance by region is fine. A second pie chart of the same attendance by day, in a different section, with a rounding difference, is how you get to ninety-four slides.

Cut material goes into a companion workbook with a stable filename, one tab per cut section. The teams that lost their slides can still point at the tab, which is what they actually wanted, and the report keeps its shape.

The report and the sales conversation are different documents

There is a version of this argument I hear often and disagree with, which is that the post-show report should be written to help sales rebook. It should not.

A report written to persuade drifts, slowly and without anybody intending it, towards the flattering definition. Attendance becomes the largest defensible number. The exhibitor count picks up co-exhibitors. Revenue picks up a sponsorship deal that was signed for next year. None of that is fraud, and all of it destroys the series, because next year's team will make different flattering choices and the two editions will not be comparable. Holding a series steady across editions is a separate discipline, and standardizing show reports is where it lives.

Write the report to be true, publish it internally, then let the sales team build their own material from it. The sales cut is a real artefact with a real audience, and the cuts each stakeholder group needs handles how it differs. It reads from the report. It does not shape it.

Where this stops

A four-question report assumes the four questions are the right ones for your show, and for some shows they are not.

A confex where two thirds of revenue comes from delegate fees and sponsorship is badly described by net square feet, and forcing the CEIR four onto it produces a front page that is technically correct and practically useless. The same is true of a hosted-buyer event, where the count that matters is qualified meetings held rather than bodies through a door. The structure to keep in those cases is the sequence, populations first and consequences second, with your own four metrics in the slots.

The second limit is harder. This design assumes your four numbers are stable by the time you publish, and in most organisations they are not. Registration data keeps settling for weeks, exhibitor contracts get amended, and the revenue figure moves when finance closes the period. A report that goes out on day five and a report that goes out on day thirty will disagree, and the difference is not an error. D32 makes the case for a data freeze date, which is the only thing I have seen that stops the numbers moving after publication.

Take last edition's pack and count how many distinct numbers in it are described as attendance. If the answer is more than one, find out which of them went to the board, and start there.