Three reports predicting online political ad revenues in 2008 have been published since December, and none agrees on how much will be spent or on what. Each uses a different methodology. And each factors in a different set of criteria — anything from potential expenditures on e-mail lists and staffing resources to non-candidate related public policy advertising.
A recently published Lehman Brothers report takes the most bullish approach, and includes forecasted expenditures on all forms of online advertising by candidate and public policy related campaigns. The analyst firm prognosticates political advertisers could spend over $110 million on Web advertising this year, 3.6 percent of the $3 billion in total political ad spending it predicts will be spent across all media.
Meanwhile, a December report from PQ Media expects far more to be spent on political advertising across media, $4.5 billion. Though it casts a wider net, PQ predicts just $73 million will go towards the Web, compared with Lehman’s $110 million forecast.
Enter Borrell Associates with the most conservative estimation of online political ad dollars. Though the research firm anticipates $4.8 billion will be spent in total on political advertising in ’08, the highest total revenue forecast of the reports, it believes just $20 million in political ad money will go to the Web.
“We were all surprised that it was as low as it was,” said Borrell Associates Director of Research Larry Shaw. However, once he recognized that campaigns are taking advantage of as much free online media as possible, such as social network profiles and search engine optimization, it made more sense.
Borrell configured its estimate based on ad spending data submitted by political operations to the IRS, along with several additional data points such as the number of employees dedicated to digital media efforts, according to Shaw.
The research firms also seem to disagree on what segments of online advertising will reap the rewards from political advertisers. Borrell’s numbers show almost half of online political ad budgets will be spent on paid search ads, 25 percent will go towards standard ad formats, about 24 percent will be allocated towards streaming audio and video, while just 2 percent will be spent on e-mail.
PQ Media, however, pegged e-mail spending at about 75 percent of all online political ad dollars. “Spending is probably about a three-to-one, with more of it going towards the e-mail marketing and fundraising aspect of it,” PQ Media VP head of research Dr. Leo Kivijarv told ClickZ News in December.
Shaw disagreed, contending, “Most of the money is being spent on acquiring those e-mail addresses” through online display and search ads.
Exactly what expenditures PQ included in the e-mail category is unclear; the firm was not available to comment for this story. However, if more than $50 million (75 percent of the $73 million PQ predicts will go online) is being spent by political advertisers in a campaign cycle of about 2 years, chances are the estimate is based on more than just the cost of delivering those messages. PQ may have considered things like e-mail writing staff or e-mail list purchases.
“From what I’ve been able to glean, [PQ Media’s] definition is extraordinarily broad,” said Michael Bassik, VP Internet advertising at political consulting firm MSHC Partners. But what holds true is if you take out the e-mail number from PQ [and compare it to Borrell’s $20 million prediction], there is some consistency.”
As for Borrell’s estimate that around 50 percent, or $10 million, will go towards paid search ads in ’08, Bassik thinks search advertising could garner closer to 80 percent.
“While we recognize that [search advertising] is still new for presidential campaigns, we’re more optimistic in our projections,” for 2008, said Google’s political ad sales manager, Peter Greenberger. In addition to geo-targeted search ads from the presidential candidate campaigns, Greenberger noted advocacy groups and unions have bought Google ads targeted to early primary states. The company also expects money to trickle in from congressional, statewide and local candidates this year.
It’s even less clear what the Lehman Brothers’ $110 million number is based on. “Importantly, we note that our estimates are directly associated with the advertising budgets of each candidate,” states the report, which suggests that spending on online initiatives beyond strict paid ad placements, such as resources dedicated to social media efforts, were included in its estimate for 2008 online spending. The company did not make anyone available for an interview for this story.
To provide an additional online ad spending gauge, ClickZ News estimated the amount presidential campaigns spent on online display advertising in 2007. According to Nielsen Online AdRelevance, the candidate campaigns ran around 285 million display ad impressions last year, the majority of which appear to have been served in remnant inventory through ad networks.
Though much of it was probably sold on a cost-per-click basis, applying the average maximum remnant inventory CPM rate of $0.50 can be used to approximate the cost of those ads. If purchased at a $0.50 CPM, the presidential campaigns would have spent just around $142,000, a rounding error when it comes to television ad spending.
“How these [research] organizations are defining advertising is not consistent,” said Bassik. “Candidates are spending tremendous amounts of resources, time and energy to using the Internet, but in terms of paid political advertising, the lower estimates are more likely.”
Time is running out to feature your company in our inaugural Mobile Vendor Reader Survey.
Marketers create personas to better understand their target audience and what it looks like. If marketers can understand potential buyer behaviors, and where they spend their time online, then content can be targeted more effectively.
What’s behind a successful data-driven marketing strategy?
Audience targeting can be challenging in social media, especially when brands make quick assumptions about their target users. How can you avoid generalisation and what are the real benefits of it?