The PR multiplier for sponsorship has no evidence behind the number three

The agency recap for a 60,000 dollar sponsorship arrives with a line reading total PR value, 540,000. Underneath it, in eight point type, sits the working: media value 180,000, PR multiplier 3x.

Ask where the three came from and you get one of two answers. It is what we always use, or it accounts for the fact that earned coverage is more credible than advertising. The first is not a reason. The second is a hypothesis that somebody could have tested at some point in the last sixty years, and the reason a PR multiplier for sponsorship keeps appearing in decks is that almost nobody has.

Where did the multiplier come from?

It has a traceable origin, which makes the absence of evidence behind it more awkward rather than less.

Watson (2013), in a history of advertising value equivalence published in Public Relations Review, traces multipliers to the late 1960s. Ruff (1968) was among the first to claim that non-advertising publicity could deliver more value than advertising, from a comparative study of product enquiries generated by publicity against print advertisements for the same new product. Macnamara (2008), quoted by Watson, records what Ruff actually found: publicity outperformed advertising for that product by seven to one, but for some publications the ratio was only 2.5 to one, and for others advertising outperformed publicity by 2.5 to one.

Read that carefully. One product, one study, and the effect reverses direction depending on the publication. A finding that runs from 7.0 in favour of publicity to 0.4 against it, in the same experiment, is the origin of a coefficient now applied as a constant.

What happened next is that the claim got repeated in trade press through the 1980s until it stopped needing a source. Watson (2013) records that the value is often boosted by "multipliers which can range from 2.5 to 8.0", citing Weiner and Bartholomew (2006). Those multipliers are, in Noble's phrase quoted by Watson, "justified on the basis of the received wisdom that editorial coverage is more credible than advertising", although Weiner and Bartholomew concluded there is "no known objective research to support this claim".

AMEC published a list of 22 reasons why AVEs are invalid in 2017 as part of its campaign against the metric, and the multiplier compounds every one of them, because it takes an unstable base figure and scales it.

Working the sensitivity, because the range is the finding

The most useful thing you can do with a multiplied figure is not to argue about it. It is to recompute it across the published range and put the results next to each other.

Start with the 180,000 media value. At the low end of the range Weiner and Bartholomew reported, 2.5, the multiplied figure is 450,000. At three it is 540,000. At eight it is 1,440,000. The same asset delivery, reported under three coefficients all of which are in active use, spans 990,000 dollars.

Now add the instability in the base. A media value is itself built by picking a comparable advertising rate, and reasonable people pick different ones. Suppose the base could defensibly have been computed as anything from 120,000 to 240,000, which is a factor of two and not an unusual spread. Combine the two ranges and the reported figure spans 120,000 times 2.5, which is 300,000, up to 240,000 times 8, which is 1,920,000.

That is a 6.4 to 1 span on a single package, produced entirely by choices nobody has to disclose. Against a 60,000 rights fee, the implied return runs from five times to thirty two times depending on which end of each range the author preferred.

No sponsor finance team can act on a number with that property, and most of them can smell it without doing the arithmetic. Doing the arithmetic in front of them is a better move than defending the middle of the range.

What the number three would have to be measuring

Suppose you wanted to defend a multiplier honestly. What would you have to demonstrate?

You would need a controlled comparison between the same message delivered through a paid channel and through an earned one, to comparable audiences, measured against a behavioural outcome rather than a recall question, with enough replications across categories and audience types to say the ratio is stable. Then you would need to show the ratio transfers from editorial media to sponsorship assets, which is a much larger jump than it looks.

That jump is where the events version falls apart on its own terms. The credibility argument rests on a third party choosing to write about you. A journalist read the release, judged it worth publishing, and their judgement is what the reader trusts. Now apply that reasoning to a lanyard. Nobody at the show made an editorial judgement about the lanyard. The sponsor bought it, your operations team printed it, and an attendee put it round their neck because it holds their badge. There is no third party endorsement in the chain, so the coefficient that exists to price third party endorsement has nothing to attach to.

The same holds for hall banners, escalator wraps, aisle signs, charging stations and app splash screens. Every one of them is paid placement in a controlled environment, which makes them closer to advertising than editorial coverage. If you were going to apply a coefficient to those on credibility grounds, the honest direction of travel would be downward.

What does a multiplied figure do to next year's negotiation?

This is the part that gets skipped, and it is the part with money attached.

Reporting 540,000 of value against a 60,000 fee tells the sponsor their package returned nine times its cost. Only two things can follow, and both of them are bad for you.

If they believe it, the obvious next question is why the package costs 60,000. A sponsor running a procurement model that targets a four to one return on marketing spend will look at a nine to one claim and conclude the asset is underpriced by a factor of 2.25, which puts a fair fee somewhere around 135,000. That is your own document making the case for a price you then have to substantiate when their category buyer asks for the underlying counts. You will not have them, because the 540,000 was never built from counts.

If they do not believe it, and most experienced buyers do not, the damage spreads. A finance reviewer who spots one unsupportable figure applies a discount to the rest of the document, including the parts you measured properly. Your badge scan count at the sponsored lounge, which came off a real reader and is defensible to the individual scan, now sits in the same file as a coefficient somebody invented. The good number pays for the bad one.

There is a third possibility that is worse than either. They believe it this year, budget against it, get an internal review two years later, and the whole relationship is repriced downward at once.

What to send the sponsor instead

The replacement has to survive the same test you just applied to the multiplier, which means both numbers in the ratio have to exist somewhere auditable.

Cost per outcome does. The rights fee comes off an invoice. The outcome count comes off a scan log, an entry counter or a meeting record. Dividing one by the other produces a figure with two checkable halves, and the full method for building it belongs with what to report instead of AVE.

The exposure count belongs on the same page, unmultiplied, with the method that produced it. Weiner and Bartholomew (2006) recommended the plain version for media impressions: for print, impressions equal circulation, for broadcast they equal audience, and online they equal visitors. The sponsorship equivalent is that passes equal passes and scans equal scans, with any adjustment named as an adjustment.

Where you genuinely have no measurement for an asset, say so in the row rather than filling it with a coefficient, which is the practice covered in reporting sponsorship results honestly. A blank with a reason survives a sponsor's procurement review. A multiplied estimate does not.

Where this stops

Removing the multiplier will sometimes make your report look worse than a competitor's, and it is worth being honest that this is a real commercial cost rather than a purely intellectual victory.

A sponsor comparing two shows will see 540,000 from one and an exposure table from the other, and a junior analyst under time pressure may well take the larger figure at face value. That happens. The defence is that the exposure table is the one that survives contact with a procurement audit, and the sponsor who checks is usually the sponsor with the larger budget.

There is a second limit, which is that dropping the multiplier does nothing about the base. An unmultiplied advertising value equivalency is still a rate borrowed from a market you do not sell in, and the argument for retiring the base figure entirely is the subject of the AVE itself. Halving the error is progress. It is not the same as fixing it.

The last limit is the one nobody enjoys. Some sponsors will ask for the multiplier explicitly, because their own internal reporting has used one for a decade and changing it means somebody senior admitting a decade of numbers were wrong. You will not win that in a renewal meeting. Give them the exposure counts and the cost per outcome as well, in the same document, and let the two versions sit side by side for a cycle.

The first step takes twenty minutes. Open the last sponsorship recap you sent, find the multiplier, and recompute the headline figure at 2.5 and at 8. Put the three totals in a row and look at the spread. That row is the argument, and you can show it to your own commercial director before you ever show it to a sponsor.

The wider discipline of tying every reported figure back to a countable source is the same one that governs exhibitor analytics on the booth side of the show.