The event business data room checklist a disciplined buyer works through

The data room opens and there are 340 documents in it. Most are PDFs. There is a folder called Commercial containing eleven presentations, a folder called Financials containing the audited accounts and a management pack, and a folder called Operations containing the venue contract for the current year and nothing else.

An event business data room checklist that works is short and specific about form. Ask for artefacts you can recompute from, name the file type, name the grain, and name the number of editions. A summary of exhibitor retention is worth almost nothing. The booking file that summary was made from is worth several days of your analyst's time and usually a price adjustment.

What makes an artefact worth asking for?

One test. Could you, holding this file, produce the seller's number and also produce a different number under a different definition.

That test rules out almost everything in a typical event data room. A slide saying retention was 88 per cent fails it. A dashboard export of attendance by day fails it. The exhibitor booking file passes, because from it you can compute logo retention, revenue retention, square metre retention, concentration, price realisation and cohort growth, under whichever definitions you choose, consistently across five editions.

The second test is whether the artefact carries its own history. A current floorplan tells you how the hall looks. A floorplan with version history tells you which stands moved in the last fortnight before doors, which is how you find out whether the show was genuinely full.

The fifteen artefacts

Per show, per edition, five editions minimum. Ask for all fifteen in the first request rather than discovering them one at a time.

1. Exhibitor level booking file: company, parent company, contract, stand number, square metres, list rate, discount, net revenue, contract date, payment dates.
2. Registration and badge export at row level, with badge type, source, qualification answers and every entrance scan timestamp.
3. Floorplan files with version history and dates, including the version at which sales closed.
4. Rate card history, with published terms, early booking deadlines and any zone premiums.
5. Venue contracts, with attrition clauses, cancellation terms, rate escalators, option years and expiry dates.
6. Sponsorship and ancillary revenue schedule, contract by contract, with what was delivered against each.
7. Exhibitor contract templates and every non-standard side letter signed in the period.
8. Comp policy and the issuance log, including exhibitor guest pass allotments.
9. Marketing spend by channel and campaign, with dates, so registration pacing can be read against it.
10. Aged debtors and bad debt written off, per edition.
11. Deferred revenue schedule with the recognition policy actually applied.
12. Headcount by role with tenure, notice period and any retention arrangement already in place.
13. Vendor contracts for registration, lead retrieval, matchmaking, the general services contractor and the audience acquisition agency, with expiry and change of control terms.
14. Association endorsements, licences, joint venture and co-organiser agreements.
15. The edition calendar: dates actually run for every edition, every date move, every cancelled edition, with the reason recorded.

Fifteen artefacts across three shows is 45 requests, each covering five years, so 225 artefact years in total. That number matters, because you are going to score against it.

Scoring what arrives

Run the request as a tracked list and record three states per cell: supplied and usable, supplied and unusable, absent.

On a three show portfolio I would expect something like this. Nine of the fifteen categories arrive complete for all three shows and all five years, which is 135 artefact years. Three categories arrive for only the two most recent editions and only for two of the three shows, which adds 12. Three categories do not arrive at all. Total 147 of 225, or 65.3 per cent complete.

Unusable is its own category and it is common. A booking file supplied as a PDF of a report is unusable. An exhibitor list without revenue is unusable for everything except logo counts. A badge export with the qualification answers stripped out for data protection reasons is genuinely unusable for qualification analysis, and that constraint is legitimate, which is why the state is separate from absent.

Publish the completeness table in the main body of the report, alongside the financial analysis, where the investment committee will read it. An appendix is where findings go to be ignored. A buyer who has recomputed 65 per cent of the story has a defensible view on 65 per cent of the story.

What does a 65 per cent completeness score buy you?

It buys you a specific negotiating position instead of a vague uneasiness.

Look at where the 78 missing artefact years sit. If the gaps are spread evenly across categories and years, you are looking at a business that never built the systems, and the finding is an integration cost: somebody has to rebuild five years of history after completion, and that work has a price and a duration.

If the gaps cluster, the finding is different. Missing floorplan history for exactly the two editions where the seller claims the show was sold out. Missing rate card detail for the year realised price fell. Missing side letters for the largest three accounts. Clustering is not proof of anything, and it is a reason to convert the affected claims into warranties with specific indemnities rather than accepting them as diligenced.

Either way, the arithmetic of the ask is what makes the conversation possible. Fifteen artefacts, three shows, five years, 147 delivered. Nobody argues with that framing, because it was agreed at the start.

The lemons problem, and why silence is information

Akerlof (Quarterly Journal of Economics, 1970) set out what happens in a market where sellers know the quality of what they are selling and buyers do not: buyers discount everything to the average, good sellers withdraw, and the average falls again. The mechanism is the reason a well-run event business should want to hand over the booking file, and the reason a buyer should read a refusal as a signal about which half of the market this seller is in.

Bergh, Ketchen, Orlandi, Heugens and Boyd (Journal of Management, 2019) reviewed 223 articles on information asymmetry across management research and organised what the field knows about how it is created, signalled and reduced. The part worth carrying into a data room is the distinction between reducing asymmetry and signalling around it. A seller who supplies raw files reduces it. A seller who supplies an accountant's report on the raw files signals about it, which is worth something and is not the same thing.

In practice, most gaps in an event data room are capability rather than concealment. Organisers change registration platforms and lose the history. Booking data lives in a system the sales team abandoned four years ago. The floorplan is a CAD file that gets overwritten every edition. None of that is sinister and all of it is expensive, because the acquirer inherits a business that cannot answer questions about its own past.

The four artefacts sellers resist hardest

Worth knowing in advance so you can ask early and escalate once instead of three times.

Side letters. Discounts and commitments that live outside the standard contract, usually with the largest exhibitors, and often granted by someone who has since left. These matter because they set the base the buyer inherits, and they explain price realisation gaps that otherwise look like sales indiscipline.

Venue contracts in full. Sellers often supply a summary because the contract has a change of control clause, an escalator, or an attrition liability nobody wants to lead with. Read the actual document, and diarise every option date in it.

The comp issuance log. It is the single fastest route to understanding the real audience mix, and it is embarrassing when it turns out the number of comps was set by whoever asked. Rebuilding the audience figure from those rows is the attendance data due diligence exercise.

The edition calendar with reasons. A show that moved from March to June in 2023 and back in 2025 has two comparability breaks in a five year series, and everybody's growth rates are wrong until that is disclosed.

Where this stops

A completeness score measures what you received. It cannot measure whether what you received is accurate, and the two failures look completely different in a report.

A booking file can be complete, well formatted, five years deep and still wrong, because the discount field was only populated from 2024 onward, or because a migration in 2023 mapped stand sizes in feet into a metres column for one show. Those defects are found by reconciling the file against something independent: total booking file revenue against the audited revenue line, total square metres against the floorplan, exhibitor count against the printed show guide. Run those three reconciliations for every year before you trust any of the analysis built on top, and expect at least one of them to fail. The revenue reconciliation in particular is where the quality of earnings for events work starts, because the difference between what the booking file says and what the accounts recognised is usually deferred revenue, barter or an edition that moved.

The other limit is timing. A checklist assumes an orderly process with time to iterate. Competitive auctions do not give you that, and a buyer with four weeks has to choose. If I had one week and one show, I would take the exhibitor booking file and the edition calendar and give up the other thirteen, because between them they let you rebuild retention, concentration, price realisation and comparability, which is most of the commercial story. The questions those two files raise then need answers, and those belong in the questions to ask a show seller session.

Send the fifteen line request today, before the model is built, with the file format and the number of editions named against each line. Then track the responses in a table with three states, and bring that table to the first synergy and integration review. What is absent will shape the deal at least as much as what arrives.