CPM pricing for event signage borrows a method built for billboards

The gantry banner spanning the crossing between halls 2 and 3 is priced at 15,000. A sponsor's media buyer asked what the CPM was, and the sales team went away, did the division, and came back with 417 dollars. The buyer's next question was what an impression meant, and that one took three weeks.

CPM pricing for event signage is arithmetic anybody can do in a spreadsheet: price divided by impressions, times a thousand. The number that comes out is only as meaningful as the definition of the unit in the denominator, and exhibition halls have no agreed definition at all.

The arithmetic, worked once

Take the gantry banner. Illustrative figures for a three day show.

Verified unique attendance is 9,000. The banner sits over the crossing aisle every attendee uses to move between the two halls, and from a traffic study you estimate each attendee passes under it an average of four times across the show. That gives 9,000 times 4, or 36,000 passes.

Call each pass an impression and the CPM is 15,000 divided by 36, which is 416.67. Round it and you have the 417 that went back to the buyer.

Three inputs, one of which is a price you set, one of which is an attendance figure somebody else audits, and one of which is an assumption a colleague made about how people move. Two thirds of the calculation is not measurement, and the output carries two decimal places.

What the other media mean by an impression

The reason the buyer's question was hard is that she came from media where the answer is written down.

The Media Rating Council published its Viewable Ad Impression Measurement Guidelines, version 1.0, on 30 June 2014, prepared in collaboration with the IAB Emerging Innovations Task Force. For a display advertisement the document requires that "Greater than or equal to 50% of the pixels in the advertisement were on an in-focus browser tab on the viewable space of the browser page", and that "The time the pixel requirement is met was greater than or equal to one continuous second, post ad render." Video needs two continuous seconds. Where a measurer uses a 30 per cent pixel threshold for very large display units, the guidelines say the practice "should be fully disclosed to data users."

Out of home did the same work in its own terms. Geopath's 2019 best practices, standards and protocols document, produced by the Geopath Futures Council, defines impressions as "the number of eye contacts that people have with an OOH unit(s) in a week", and the same document sets out the components that feed the count, including traffic, vehicle occupancy, illuminated circulation and a visibility adjustment.

Neither definition is obviously correct. One continuous second at half the pixels is a committee's choice. What both have that an exhibition hall lacks is a published threshold, so that two people using the word impression are describing the same event.

Why does an event CPM come out so far above a billboard rate?

Because the denominator is thousands where outdoor's is millions, and the arithmetic does the rest.

This is the point where somebody usually produces an outdoor CPM from a blog post and declares that event signage is forty times overpriced. I would leave that comparison alone. The published outdoor figures that circulate are rarely traceable to a primary source with a stated method, and comparing your carefully constructed 417 to an untraceable 8 is a false precision on both sides of the ratio.

The defensible version of the comparison is qualitative and still useful. A high CPM against a general-population medium is expected when the audience is filtered. Every one of your 9,000 attendees registered for a trade show in the sponsor's sector, gave a job title, and travelled to be there. A billboard's audience includes everybody who drove past. If a sponsor is paying a large multiple for a much smaller and much better qualified audience, that can be a good trade, and it is a trade they should be allowed to evaluate with your denominator on the table.

The size of the number was never the problem. What sinks a signage CPM in a negotiation is a denominator built from an assumption nobody wrote down, held up against denominators that were.

There is a version of the comparison worth making, and it goes the other way. Ask the buyer what they pay per qualified lead through their other channels, and work backwards. If their blended cost per marketing qualified lead is 340, then a 15,000 banner has to produce about 44 of them to break even on their own arithmetic. That reframes the discussion onto ground where both sides have real figures, and it does not require either of you to accept an impression definition you cannot verify.

What should an impression mean at your show?

Write the rule before you publish the rate, and keep it simple enough that operations can apply it to thirty assets without a meeting each time.

A workable definition for physical signage has three parts. A viewing zone, described geometrically: the area of floor from which the face is unobstructed and subtends a reasonable viewing angle. A time window: the show's open hours, excluding build and breakdown. And a counting rule: one impression per entry into the viewing zone by a badged attendee, with re-entries counted separately after some interval.

Then say what you have excluded. Exhibitor staff, contractors, and anybody in the hall outside open hours. Those exclusions matter more than they sound. On a show with 9,000 attendees and 2,400 exhibitor staff, including staff in the count inflates every physical signage impression by 26.7 per cent, and the resulting CPM falls from 416.67 to 328.95 without a single extra person seeing anything.

The definition also has to say whether a pass is an impression or an opportunity for one, because those are different quantities and the gap between them is large. That distinction is the whole subject of the visibility factor, and the honest thing is to name which of the two your published CPM is built on.

What moves the number, and by how much

Run the sensitivity once and put it in the same document as the rate, because the buyer will run it anyway.

Hold the price at 15,000 and vary the assumption. At two passes per attendee the count is 18,000 and the CPM is 833.33. At four it is 416.67. At six it is 277.78. Applying a visibility factor of 0.30 to the four-pass figure gives 10,800 adjusted impressions and a CPM of 1,388.89.

So a set of choices that all sound reasonable produce CPMs between 278 and 1,389, a factor of five, on a banner whose price never changed. The pass frequency assumption alone swings the answer by three times. Anyone quoting a signage CPM to two decimal places without that range beside it is reporting a preference.

The choice of attendance base does the same thing again, and that one has enough in it to be its own decision with its own consequences once you start putting several assets on the same axis.

What a CPM cannot price

A CPM is a cost per unit of exposure, so it prices exposure and nothing else.

It has no view on whether the banner reached the right people, which is the main thing the sponsor bought your show for. Two assets at the same CPM can differ completely in the seniority and buying authority of the people passing them, and the cheaper one on a CPM basis is frequently the worse buy.

It also has no view on sequence or context. A banner seen on the way into a keynote does different work from the same banner seen at 17:40 on day three, and no impression count distinguishes them. Where a sponsor's objective is a specific action, the CPM belongs as a secondary figure on the report with the action count above it.

The last thing it cannot do is price a creative execution. Two sponsors take the same gantry for consecutive editions. One puts a logo and a stand number on it, the other puts a single question and a QR code. Same asset, same price, same impressions, and one of them will produce measurable traffic while the other produces goodwill. The CPM is identical in both cases, which tells you it has measured the inventory and has nothing to say about the campaign. That is fine as long as everybody understands it, and it becomes a problem the moment a renewal conversation treats the CPM as a performance figure.

Where this stops

The method's honest ceiling is that it prices an estimate against a price, and only one of those is a fact.

The estimate can be improved. A gate count gives you the attendance base, a sample of movement gives you a pass frequency with an error band, and building the count properly from its three inputs is a day of work that most shows have never done. What no amount of improvement gives you is a measurement of attention, and a CPM presented as though it were one will eventually meet a sponsor who tests it.

Take the three most expensive signage assets on your card and compute the CPM for each under two pass-frequency assumptions, one conservative and one generous. Put both columns in the same table and take it to your commercial lead before the next card goes to print, along with the impression definition you would be prepared to publish in the sponsor's own reporting view.