Include/Exclude Political Ad Spend: How Do You Vote?

A funny thing happened on the way to Zenith Media's new ad forecast this week. It decided to drop political ad spending from its estimates.

The move is noteworthy for several reasons.

First and foremost, it effectively makes all of its historical data -- 35 years of forecasts it began publishing in 1988 -- uncomparable with its current, and all future forecasts from the Publicis Media unit.

That's an interesting move for an agency that worked hard to establish itself as a player in ad industry economics.

It was the first to muscle in on Interpublic's long-standing role as the ad industry's -- and arguably Wall Street's -- default source for ad industry growth and volume.

It also begs the question: Why now?

I put that question to a Publicis Media spokesperson shortly after I covered Zenith's initial release, and missed the fine print noting its change in a footprint at the end of this year's forecast (see below).



I'm still waiting to hear Zenith's rationale -- other than what it says here, that the move is intended to "provide a more stable and representative projection of the advertising marketplace -- but it's something I've been thinking about more generally, so I figured I'd write a column about that now.

The truth is, I never really understood why two other big agency holding companies -- IPG Mediabrands' Magna and WPP's GroupM -- originally became the first to separate U.S. political ad spending from their U.S. and worldwide ad-spending projections.

So I followed up this week with the guy responsible for that -- former Magna and GroupM forecaster and current Madison and Wall publisher -- Brian Wieser.

Wieser made the move shortly after he succeeded long-time Interpublic forecaster Bob Coen, and his rationale then was the same as today: because the erratic nature of U.S. political ad spending causes distortions and anomalies in the underlying tracking of ad industry growth.

Wieser brought that approach to GroupM when he joined that agency, but unlike what Zenith is doing now, Magna and GroupM each continue to report their ad industry growth estimates for both political and without it.

The reason those agencies utilize "ex-political" as their default stat for "nominal" or "underlying" ad industry growth, Wieser told me, is because they didn't want to confuse readers, but it's there for anyone who wants to factor it.

MediaPost's readers probably already know that I've been using their totals -- including political -- to make industry comparisons and build composite charts tracking and trending what the Big 4 agency holding companies are estimating.

I've been doing that for several reasons, including the fact that two of them Zenith, as well as Dentsu, had been reporting totals.

But now that Zenith is going to report only ex-political, I'm honestly not sure what to do about that. Should I drop Zenith from our total ad-spending composite? Should I default to an ex-political view and drop Dentsu? Should I try to come up with some cockamamie eye chart mashing up -- and footnoting -- the two (which would definitely do what Wieser suggests: confuse readers)?

Well, that's my problem. But if you have any suggestions, I'd appreciate hearing them at

But before you do, consider the second part of my reasoning here, which is that while U.S. political ad spending is indeed erratic and full of peaks and valleys, it is becoming a bigger part of total U.S. ad spending, and the deltas between the biannual spike years and the "odd" years are growing smaller, albeit marginally.

To look at that, I calculated the share of U.S. ad spending that political has represented in each year since 2014, courtesy of GroupM's nifty spreadsheet (see below).

Political's share of total U.S. ad spending in odd-year elections has doubled from 0.5% to 1.0% this year.

And while that's not nearly as stable as the massive spikes in even-year elections -- especially presidential ones -- it's not nothing.

Share of market aside, the magnitude of U.S. political ad spending keeps growing and growing and growing, and it's not just from major national candidates and parties, but from increasingly down-ballot ones, as well as a burgeoning marketplace of issues-oriented political advertising.

I know this from looking at AdImpact's tracking in recent years, which is perhaps the best, most detailed and granular view of the category, and if you spend the time to thumb through it, you'll unveil oodles of insights.

I wrote a "Red, White & Blog" column about one of them recently while attending an AdImpact meeting with TV station reps at which Founder and CEO Kyle Roberts pointed out how the Supreme Court's Dobbs decision has effectively created a half billion ad spending category virtually overnight for local media in states that have or are expected to put reproductive rights on their ballots.

In these highly contentious times, Roberts believes sub-categories of political advertising like that one are more likely in the future than not, so it makes me wonder why a major agency holding company would choose now to exclude it unilaterally from its view of the marketplace.

I mean, it may not be something Publicis' agencies deal much with, but it sure it something their media supply chain -- especially local broadcasters and increasingly big digital media platforms -- consider to be a part of their margins and bottom lines.

1 comment about "Include/Exclude Political Ad Spend: How Do You Vote?".
Check to receive email when comments are posted.
  1. Ed Papazian from Media Dynamics Inc, December 9, 2023 at 5:42 p.m.

    Leaving political out as a way to provide more continuity for "regular" ad spend sounds nice in theory---but doesn't political advertising eat up some of the GRPs that would otherwise be available for non-political advertisers? In which case the idea of comparability  might be questioned. Or do the non-political advertisers merely spend what they would normally spend in major political campaigning years while the media happily raises its ad clutter ratios to provide added GRPs for the political folks to exploit?  It's also true that the CPMs for political ads are not necessarily the same as for other types of advertisers so ad spend tallies may not reflect the extent of the political ad campaigns' GRP tonnage relative to non-political spenders. .

    In view of all of the pros and cons, I would say it's best to show the ad spend estimates both ways.

Next story loading loading..