Your Booth Game Tops Out at Twelve Plays an Hour Per Station. The Brief Assumes Thousands.

A booth game is capacity-limited hardware, not a media buy. One station clears roughly a dozen plays an hour, and your whole trade show brief follows from that.

The brief lands in your inbox and it reads like a campaign plan. Drive engagement across the stand. Build a leaderboard with a prize at the end of each day. Capture five hundred qualified leads. Somewhere in the deck there is a render of a single hero screen with a crowd around it, and the crowd is drawn at about forty people.

Nobody in that chain has multiplied anything. A booth game is not a media buy where impressions scale with spend. It is a queue served by a fixed number of physical stations over a fixed number of open hours, and the throughput of that queue is set by furniture and human hands, not by how good the game is. Work the arithmetic before you sign and half the brief rewrites itself.

🎪 The Demand Side Is Real. That Is Not the Problem.

Start with the part the research supports, because it is the reason exhibitors keep commissioning these things. Freeman's August 2026 trends report, summarised by Exhibit City News, surveyed 3,319 people — 2,508 attendees and 811 exhibitors, with a stated margin of error of ±2.1% and ±3.4% respectively. Tradeshow attendees said they want 64% of their learning time to happen outside formal sessions. 66% named learning from exhibitors as something they value, effectively level with peer conversation at 67% and ahead of keynotes and expo theatre at 60%.

Then the gap. 85% of exhibitors in the same study rated access to their experts as important, but only 61% said they prioritise hands-on demonstrations. Attendees want to touch something; a meaningful share of stands are still built around a wall of text and a bowl of sweets.

So the instinct to put something playable on the stand is well founded. The failure happens one step later, when the exhibitor who acts on that instinct sizes the deployment as if plays were impressions.

⏱️ The Cycle Time Is the Whole Ballgame

One station, one player. Break the cycle into the four things that actually consume the clock:

• Approach and greeting — 15 to 30 seconds before a hand touches the screen
• Explaining the rules — 20 to 45 seconds if a staffer has to say them out loud, close to zero if the attract loop teaches the mechanic
• The play itself — 45 to 90 seconds in a well-designed booth build
• Score, capture, hand over — 30 to 60 seconds, and this is the step every exhibitor underestimates

Add the ranges. A tight cycle is about 110 seconds. A loose one is closer to 225. That is 16 to 33 plays per hour per station, and only if the next player is already standing there when the last one steps away.

They are not. Show floor traffic is not uniform: it clusters after the doors open, thins during keynote slots, and collapses in the final hours of the last day. Assume a station is genuinely occupied 40% to 60% of open hours and the realised figure lands somewhere between seven and twenty plays an hour. Twelve is a defensible planning midpoint. Use your own show's traffic curve if you have last year's badge scans; use twelve if you do not.

The number that should worry you

There is a widely repeated industry figure putting average dwell time at a visited booth at 4.2 minutes. Treat it carefully — it circulates in 2026 exhibition-industry compilations with two incompatible attributions, one calling it CEIR's 2025 attendee study and another calling it a 2024 CEIR finding specifically about interactive stands versus static ones. None of the compilations link to the underlying report. Do not build a business case on it.

Do notice what it implies if it is even roughly right. A 110-to-225-second game cycle is not a slice of the average booth visit. It is most of it. The game does not sit alongside the conversation your sales team came to have — it replaces it, unless you design for both to happen at once.

📉 Multiply by Open Hours and the Brief Falls Over

Now apply the ceiling to a real show length. A conventional three-day B2B exhibition runs roughly seven open floor hours a day, so about 21 hours of capacity. Larger consumer-facing shows run longer — CES 2026 ran four exhibit days, and its own published attendee itinerary put the show floor opening at 10 AM, which pushes that window closer to 32 hours.

At twelve plays an hour per station:

• One station, 21 hours: roughly 250 plays for the entire show
• Two stations: roughly 500
• Four stations: roughly 1,000
• One station at a four-day, 32-hour show: roughly 380

Read the first line again against a brief asking for five hundred captured leads. A single hero screen cannot produce that number even if every player consents to capture and nobody walks away, which they will. The brief was not ambitious. It was unbuilt.

The pressure gets worse when you account for how attendees move. One 2026 exhibition statistics compilation, published by Pure Exhibits, cites Freeman research putting the average number of booths an attendee visits at 26 per show day. It gives no year for that figure, so hold it loosely, but the direction is not controversial to anyone who has worked a floor. Your visitor is on a route. You are one stop of dozens. A game that needs four minutes of their attention is competing with twenty-five other stands for a budget they are actively rationing.

💸 The Denominator Decides What You Can Afford

Once you have a plausible play count, the cost conversation becomes tractable. Divide the fully loaded cost of the interactive — build, hardware, floor space it occupies, the staffer who runs it — by the plays the arithmetic says you will get. That is your cost per play. Compare it against what the show costs you per lead by other means.

Here the published benchmarks disagree, and the disagreement is instructive. The Pure Exhibits compilation puts average cost per trade show lead at $142, attributed to CEIR with no year given. Wave Connect's 2026 statistics roundup puts it at $112, attributed to Exhibit Surveys for 2025, against $259 for a field sales call. A third 2026 compilation quotes a B2B range of $112 to $186. So the honest planning number is a band — somewhere between roughly $110 and $190 per lead — and anyone quoting you a single figure to two significant digits is quoting a compilation, not a study.

A band is enough. If your interactive costs $18,000 all-in and the capacity math says 1,000 plays across four stations, you are at $18 a play before anyone converts. If a third of players consent to capture, you are at $54 a lead and comfortably inside the band. If you ordered one hero screen and got 250 plays, you are at $72 a play and $216 a lead, and you have lost the argument at your own budget review.

The same compilation citing Exhibit Surveys for 2025 also reports that 80% of trade show leads are never followed up on. If that is anywhere close to true, arguing about lead volume is arguing about the wrong end of the pipe. Capacity planning is not about maximising a number your sales team will not call.

🏆 Your Leaderboard Breaks at 250 Plays

This is the failure mode nobody predicts, and it kills more booth games than connectivity does.

A leaderboard is a competitive instrument. It works when the field is deep enough that the top of it keeps moving. With 250 plays spread across three days, the top three scores are set on the first morning by whichever four people happened to be good at it, and then the board is decorative for the remaining nineteen hours. Every player after lunch on day one looks at a number they cannot reach, shrugs, and does not play.

Fix it in the design, not with more marketing:

• Reset hourly or per session block, not per day. A board that clears at the top of every hour is always winnable.
• Rank by percentile, not position. "Top 15% of players today" is achievable for many; "3rd place" is achievable for two.
• Prize on a threshold, not a rank. Beat the target, take the item. This decouples the reward from a volume you do not have.
• Show the last ten scores, not the top ten. Recency makes a thin field look busy and gives the person who just played something to point at.

🔢 Stations Are the Only Lever You Actually Control

Cycle time has a floor: you cannot make a greeting shorter than a greeting. Show hours are fixed by the organiser. That leaves parallelism as the one variable that moves total capacity, and it is usually the cheapest thing in the budget to change.

Four tablets on arms cost a fraction of one large-format touchscreen, occupy less floor, and quadruple throughput. The instinct to buy one impressive screen is an instinct about how the stand photographs, not about how it performs. If the render matters, buy the big screen for the attract loop and put the playable stations around it.

Two constraints on adding stations:

• Staffing ratio. An unattended station produces plays and no leads. One staffer can realistically float across two or three stations if the capture step is on-screen; if a human has to scan a badge for every player, it is one-to-one and your station count is capped by your headcount, not your hardware budget.
• Head-to-head halves you. A two-player versus mechanic is excellent for conversation and terrible for throughput — two people consume one cycle. That can be the right trade, but make it deliberately, and double your station count if lead volume still matters.

⏳ Design Rules That Fall Straight Out of the Ceiling

Once you accept that plays are scarce, a set of build requirements stops being preference and becomes specification:

• Hard time cap, 45 to 90 seconds. No endless runner, no "play until you lose". One skilled player can otherwise occupy a station for six minutes and cost you three prospects.
• No tutorial screen. The attract loop demonstrates the mechanic on repeat. If the game needs instructions read aloud, you have added 30 seconds to every single cycle.
• One input. Tap, or drag, or steer. Not all three. Every additional control adds explanation time and failure at the start of the round.
• Capture at the end, never at the gate. A badge scan before play is friction at the widest point of the funnel; the same scan after a score people are pleased with converts better and costs you nothing in throughput.
• Instant reset. From game over to a fresh attract loop in under three seconds, with no staffer intervention. A game that needs someone to tap through a results screen will sit on that screen through the busiest ten minutes of the show.
• Run it offline. Not a capacity point directly, but a station waiting on hall Wi-Fi is a station producing zero plays an hour.

🚫 Five Ways Exhibitors Destroy Their Own Throughput

1. Ordering one screen because it looks better in the render. The single most common and most expensive capacity mistake on any stand.
2. Gating play behind data capture. Every form field before the fun costs you players, and the players it costs are disproportionately the curious walk-ups you wanted.
3. Uncapped session length. A leaderboard plus an endless mechanic is a design that actively rewards one person for blocking your station.
4. One staffer who knows how to reset it. When they go to lunch, the station goes with them. Everyone working the stand needs the reset gesture, printed and taped inside the counter.
5. Measuring plays instead of cycles. Instrument time-per-cycle and idle time per station, not just a play counter. Idle time tells you whether to add stations next show; play count alone tells you nothing you can act on.

🎮 Where a Licensed Catalogue Changes the Sum

The capacity ceiling is indifferent to whether the game is custom-built or licensed — 250 plays is 250 plays. What changes is the numerator. A bespoke build carries its full cost against one show's play count unless you run it across a circuit; a licensed title from an existing catalogue moves the cost per play down by an order of magnitude, and lets you put four different games on four stations so the person who bounced off the first one plays the second.

That is the practical split. Commission custom when the mechanic itself has to carry a product message. Licence from a catalogue when the job is dwell time, attract value and giving a staffer a reason to start talking — which is most booth games, honestly. Forestry Games licenses HTML5 titles that run offline on booth hardware for exactly this use; the trade show games and games for events collections are built around that, and the wider catalogue is where you would look if you want several stations running different titles at the same stand. For campaign work that extends past the show floor, the marketing games side is the same catalogue pointed at a different brief.

🧭 Do This Before You Approve the Next Brief

Open the brief and find the number it promises — plays, scans, leads, whatever it is. Then work backwards in four lines:

1. Look up your show's published floor hours and multiply by days. That is your capacity window.
2. Time the intended cycle end to end with a stopwatch and a colleague who has never seen the game. Not the play length — the whole cycle, greeting to handover.
3. Divide 3,600 by that cycle in seconds, multiply by 0.5 for realistic occupancy, and multiply by your capacity window. That is your per-station total.
4. Divide the brief's promised number by that total. The result is how many stations you actually need to order.

If step four returns a number larger than your floor plan can hold, you have found the real decision — and you have found it now, while the fix still costs three more tablets instead of a post-show apology. Most exhibitors find this out on day two, standing beside a queue they cannot serve, watching the people they rented the floor to meet walk past toward the next stand on their list.