Measuring the ROI of a show feature area without double counting revenue

The startup zone brought in 561,000 of space revenue against 96,000 of build, staffing and promotion. Somebody divides one by the other and puts 484 per cent on a slide, and the room agrees to build two more zones next year.

The 484 per cent is wrong, and the reason it is wrong is not an error in the arithmetic. Of the 42 exhibitors in the zone, 24 had exhibited at the show before. Most of them would have bought space this year regardless, somewhere on the floor, at a broadly similar rate. Counting their money as a return on the zone credits a new structure with revenue that was already in the building.

Feature areas are the single easiest place in a show P and L to manufacture a return, because they sit on revenue that would otherwise have landed elsewhere and nobody is required to prove it would not have. Measuring the ROI of a show feature area honestly means starting from the exhibitor list instead of the revenue line.

Count the accounts, then count the money

The first move is to stop looking at revenue and look at the exhibitor list.

Of the 42 zone exhibitors: 18 had never bought space at this show in any prior edition, and 24 had. Those two groups need entirely different treatment and lumping them together is the whole mistake.

For the 24 existing accounts, the question is not whether they spent money in the zone. It is whether they spent more money than they would have spent anyway. An exhibitor who took 18 square metres in hall 3 last year and 18 square metres in the zone this year contributed nothing incremental, they moved. An exhibitor who kept their main stand and added a 9 square metre zone presence contributed the 9 square metres.

For the 18 new accounts, the question is whether the zone is why they came. That is genuinely harder and it is the subject of the next section.

Start with the cleanest possible version. Incremental revenue is the space revenue from the 18 new accounts only, which came to 214,000, averaging 11,889 each. Against 96,000 of cost, made of 62,000 build, 21,000 staffing and 13,000 promotion, the net is 118,000 and the return is 118,000 over 96,000, or 123 per cent.

That is a good result. It is also a quarter of the 484 per cent, and the difference is entirely the 24 accounts who were coming anyway.

Set the 123 per cent against the show's own margin before celebrating. CEIR's Performance Benchmark Playbook, whose second edition was released in July 2026, puts the average net profit margin at 55 per cent for exhibitions of 200,000 net square feet and above. A feature area returning 123 per cent on direct cost is comfortably above the rate the rest of the show turns, which is the comparison that decides whether the capital should have gone here or somewhere else. Building the show-level figure it is being compared against is the event ROI calculation in D17.

Would those exhibitors have come anyway?

Even the 214,000 is generous, because it assumes every one of the 18 new accounts arrived because of the zone.

Shows acquire new exhibitors every year without building anything. If your show has historically added around 34 new accounts a year and added 51 this year, then roughly 17 of the 18 are plausibly zone-driven and the figure holds. If it added 39, then only about 5 are, and the zone is underwater.

You cannot resolve that with the data you have, so state it as a range instead of picking a point.

At 18 attributable accounts, incremental revenue is 214,000, net is 118,000, and the return is 123 per cent. At 11 attributable, incremental revenue is 130,800, net is 34,800, and the return is 36 per cent. At 7 attributable, incremental revenue is 83,200, net is minus 12,800, and the return is minus 13 per cent.

Publishing those three lines is more honest and more useful than publishing any one of them. It converts the decision from do we believe the number into how many of these accounts do we think the zone brought, which is a question the sales team can actually answer from memory and from the enquiry log. The general habit of publishing a band instead of a point, and choosing which assumptions to flex, is sensitivity analysis on event ROI in D21.

There is a cheap way to do better next time, and it costs one field. Ask every new exhibitor at contract stage what brought them, with the feature area as an explicit option, and record it on the account rather than in an email. After two editions you have a base rate rather than an argument. It is self-reported and imperfect, and it is enormously better than nothing.

Displaced space is a cost and it is usually missing

The zone occupied 2,400 net square feet that came out of general inventory, and whether that matters depends on one thing.

If the hall did not sell out, the space had no alternative buyer and the displacement cost is zero. If the hall sold out, that 2,400 square feet would have sold at the standard rate, which at 265 per square foot is 636,000 of revenue the show did not earn. Put that in the cost side and the zone is deeply negative at every attribution level above.

This is the single largest omission in feature area reporting and it is almost never included, partly because it is uncomfortable and partly because sold out is a slippery condition that means several different things. A hall with a waiting list is sold out. A hall that reached 96 per cent with three awkward corners left is not, and the 2,400 square feet the zone took were probably some of those corners. What counts as sold out exhibition space is F26's subject and it determines whether this line is 636,000 or nothing.

The workable rule is to charge displaced space at the rate the marginal booth actually sold for, not the rate card, and only in a year where the show turned business away. In most years for most shows the honest answer is zero, and writing zero with the reason next to it is very different from omitting the line.

Count exhibitors the way the auditing standard counts them

The account counts above only work if everybody agrees what one exhibitor is, and feature areas are precisely where that breaks.

UFI's Calculation Standards and Definitions, which is Annex 2 of its auditing rules for UFI Approved Events, was written for exactly this argument. Only direct exhibitors are counted, and that covers main exhibitors, meaning the bodies contracting directly with the organiser, plus co-exhibitors, meaning organisations present on a main exhibitor's stand with their own staff and their own products or services, clearly identified on the main exhibitor's application form or in the catalogue.

Collective participation then gets its own rule, and it is the rule a startup zone or a country pavilion lives under. The space is rented and paid for by the exhibitor organising the collective participation, and the participants sharing it count as co-exhibitors only if they occupy their own area, appear under their own name, and present their own products or services with their own staff. Where any of those conditions fails they are represented companies, indirect exhibitors, and the standard says they may not be counted in the exhibitor tally at all.

So a startup zone sold as pods under one incubator's contract is either one exhibitor or thirteen, and which it is turns on whether each pod had its own name above it and its own people standing in it. The standard also requires you to state which category of exhibitor you counted, which is the sentence everybody skips.

Both counts are legitimate and they answer different questions. The revenue question follows the contract, because that is who paid. The market position question follows the exhibiting entity, because that is who occupied the category. Report both, and be consistent between editions, since a change in how pods are counted will produce a growth story that never happened. The CEIR Index, released by IAEE on 4 May 2026, tracks number of exhibiting companies as one of its four headline metrics, and a show feeding that series on an inconsistent basis is corrupting its own history.

What is a feature area genuinely good at?

A demo theatre that recruited no new exhibitors at all can still be the best thing on the floor, because feature areas do work that does not show up as space revenue.

They move traffic, which is a question for attendee analytics rather than for the contract file. A well-placed zone pulls attendees into a dead quadrant, and the exhibitors who benefit are the ones nearby who paid full rate for a location that was previously hard to sell. That effect is real, it shows up in the following year's renewal rate for those specific booths, and feature area placement on the floorplan belongs to F20 rather than here.

They also generate content, media coverage and a reason for a lapsed attendee to return, none of which is capturable in a one-year incremental revenue figure. The right response to that is not to invent a value for it. It is to write the non-financial objectives down before the show and score them separately, which is what D18's structure is for, and to keep the money calculation narrow and clean.

Where this stops

Incremental contribution is the right concept and there is no way to measure it exactly, because the counterfactual never happened.

Everything above is an estimate dressed in arithmetic. The 18 new accounts are a count, which is solid. Whether the zone caused them is a judgement, and the range from 123 per cent to minus 13 per cent is how much that judgement is worth. Presenting the top of that range as the answer is how feature areas multiply until a show is 30 per cent features and the general floor stops working.

The second limit is the time horizon. A zone that recruits 18 small new accounts this year is buying a pipeline, and small exhibitors either grow into standard stands or churn at a high rate. Neither outcome is visible for two more editions. Judging a launch zone on one year's incremental revenue is judging an acquisition programme on its first month, and the fix is to keep the cohort tagged and rerun the number annually rather than to argue about the first one.

The third is that a proper answer would require an experiment nobody can run. You cannot hold the zone out at half your show and compare, and there is no second identical edition. This is a class of question where the honest position is that the estimate has a range, the range is wide, and the decision has to be made inside it.

Take your most recent feature area, list its exhibitors, and mark each one as new to the show or not. If more than half were existing accounts, whatever return figure is currently in circulation for that zone is at least double what the zone earned.