Three sponsorship valuation methods and when each one gives a defensible number

A procurement lead at a sponsor asked our commercial director how the lanyard price of 24,000 had been arrived at. He said it reflected the value of the asset. She asked what that meant arithmetically. The meeting went quiet, and the renewal took four months longer than it should have.

Three sponsorship valuation methods are in general use, and they are worth naming because they produce different answers for the same object. Cost-plus prices what the thing costs you to make. Market comparable prices what somebody else charged for something like it. Impression-based prices the exposure at a rate borrowed from another medium. Run all three on one lanyard and the spread will surprise you.

What each of the three methods actually measures

Cost-plus takes the production and delivery cost of the asset, adds a margin, and stops. It answers the question of what you need to charge to be better off than if you had never sold it.

Market comparable takes prices observed elsewhere for similar assets, normalises them against something, usually audience size, and applies the normalised rate to your show. It answers the question of what the market has been paying.

Impression-based takes an estimated number of exposures and multiplies by a rate per thousand. It answers the question of what an advertiser would have paid to reach that many people somewhere else.

Those are three different questions. There is no arithmetic that makes them agree, and the habit of picking whichever one produces the most comfortable number is how rate cards end up undefendable.

One lanyard, three numbers

Take a show with 6,000 verified attendees running over three days, and a lanyard sponsorship. Illustrative figures throughout.

Cost-plus first. Lanyard manufacture for 6,500 units comes to 8,400, artwork and proofing 900, distribution and on-site handling 2,700. Total production 12,000. Apply a 40 per cent margin and you get 12,000 times 1.4, which is 16,800.

Market comparable second. You find three shows in adjacent verticals whose lanyard prices you can actually see: 14,000, 18,000 and 26,000. The raw median is 18,000, which is where most people stop. Normalise instead. Those shows draw 4,800, 6,500 and 11,000 attendees, so the per-attendee rates are 2.92, 2.77 and 2.36. Take the median rate of 2.77 and apply it to your own 6,000 attendees, and you get 16,615.

Impression-based third. Assume 6,000 lanyards are worn, each is seen twenty times a day by somebody other than the wearer, across three days. That gives 6,000 times 20 times 3, or 360,000 exposures. At eight dollars per thousand, the value is 360 times 8, which is 2,880.

So the same lanyard is worth 16,800, or 16,615, or 2,880. The top of that range is 6.25 times the bottom. Two of the three methods happen to land within 200 of each other, which looks like corroboration and is coincidence.

It is worth pausing on how easily that coincidence could be presented as evidence. Two independent methods agreeing to within 1.1 per cent is the kind of thing that goes on a slide with the word triangulation next to it. Change the margin assumption from 40 per cent to 25 per cent and cost-plus drops to 15,000, which is 9.7 per cent below the comparable. Change the comparable set by dropping the 11,000 attendee show, which is arguably too large to compare against, and the midpoint of the two remaining per-attendee rates is 2.845, giving 17,070. Neither change is unreasonable and both were available to whoever built the slide.

Why do the three answers diverge by a factor of six?

Because the inputs have nothing to do with one another.

Cost-plus is bounded by what your lanyard supplier quoted, which is a fact about manufacturing in a particular year. Move production to a cheaper supplier and the method says the sponsorship is worth less, which is obviously wrong as a statement about value and perfectly correct as a statement about cost.

The comparable is bounded by decisions other organisers made, and those decisions may themselves have been cost-plus, or last year plus four per cent. Comparables inherit whatever logic produced them, including no logic at all.

The impression method is bounded by the rate per thousand you imported. That rate came from a medium that defines its unit of exposure carefully. Geopath's 2019 best practices document defines out of home impressions as "the number of eye contacts that people have with an OOH unit(s) in a week", which is a specific and formally defined quantity. A lanyard glance counted by assumption is a different animal wearing the same word, and multiplying it by an outdoor rate transfers precision that was never there.

What the research covers, and what it leaves alone

If you go looking for a standard method to point at, you will find the literature is pointed somewhere else. Cornwell and Kwon reviewed sponsorship-linked marketing research from 1996 to 2017 for the Journal of the Academy of Marketing Science, published online in 2019 and in print in volume 48, and their overarching conclusion is that there is a surplus of research examining audience responses to sponsorship and a shortage of research examining the marketing management of the sponsorship process.

Pricing sits squarely in the shortage. Two decades of work on how audiences respond to sponsors, comparatively little on how the rights holder should set the number in the first place. That is worth knowing before you go looking for an authority to cite in a negotiation, because there is not much of one.

It is also a reason to be modest about the output. The defensible number is the one whose inputs you can show, which is a lower bar than the number that is correct.

When cost-plus is the defensible answer

Cost-plus holds up in one specific situation: you carry real, variable production cost, and the sponsor could plausibly have built the thing themselves.

A branded charging lounge, a shuttle bus wrap, printed delegate bags. In each case the sponsor's alternative is to source it privately, so your cost base is genuinely close to their reference price, and the margin is the fee for access and for not having to run the logistics. A 40 per cent margin on 12,000 is a claim anybody can check.

Cost-plus fails badly on assets with no marginal cost. Category exclusivity costs you nothing to produce and can be the most valuable line on the card. Applying cost-plus there gives you a price near zero, which is the reason nobody applies it there and the reason a multiplier gets invented instead.

The margin figure is where cost-plus quietly becomes an opinion. Forty per cent is a common choice and it has no derivation behind it. If you are going to use the method, tie the margin to something you can name: the blended margin on the rest of the sponsorship book, or the margin the finance team already applies to exhibitor services. A margin picked because it produced an acceptable answer is cost-plus in appearance and guesswork underneath, and a procurement lead will find that in one question.

When a comparable beats a model

A comparable is worth more than any model when you have at least three genuinely similar assets at genuinely similar shows, and you normalise before you compare.

The normalisation matters more than the comparison. In the worked example above, the raw median of 18,000 and the size-adjusted figure of 16,615 differ by 8.3 per cent, and the raw median is the one that flatters. On a card of thirty assets that gap compounds into real money and into a set of prices your own sales team quietly stops believing.

Two traps. The first is that the comparable set is often your own portfolio, which makes the exercise circular: you are comparing your price to your price. The second is that published rate cards are asking prices, and the gap between the card and the invoice can be a third of the number. If you can see somebody's card but not their contracts, you are benchmarking against fiction.

When does the impression method earn its place?

When the exposure count comes out of a log.

A digital screen with a play log knows how many times the creative ran. An app banner knows how many times it rendered. A sponsored email knows how many addresses it reached. For those assets the denominator is a record, and a rate per thousand applied to a record is a real calculation.

For a lanyard, an aisle sign or a hall banner, the count is built from a traffic estimate multiplied by an assumed pass frequency. That is a legitimate thing to do, and the arithmetic for one signage asset is worth working through carefully, but the output is an assumption with a decimal point. Selling against it is fine. Believing it is the problem.

Where this stops

All three methods price the asset. None of them price the buyer, and the buyer is where most of the variance actually lives.

The same hall banner is worth more to a company launching a product at your show than to a company defending an installed base, and no cost, no comparable and no impression count knows the difference. That is the honest limit of the whole exercise. What you have at the end of it is a number you can explain, which is enough to open a negotiation and never enough to close one.

The second limit is that a first-year asset has no cost history, no comparable and no impression record, so all three methods return nothing useful. Pricing an asset that has never been sold needs a different construction, and so does the exclusivity premium, which usually gets handled by a multiplier applied on top when a calculation was available all along.

Take the five highest-priced lines on your current rate card and write next to each one which of the three methods produced the number. Where the honest answer is last year plus four per cent, you have found a fourth method, and it belongs in the same review as the rest of your pricing before the card goes out again.