Tracking campaign ads in the 2012 elections is no easy feat. Between the flurry of spots from the Obama and Romney campaigns as the presidential race enters its home stretch, and the massive expenditures by outside groups such as super PACs, it is hard to get a handle on who is spending what to influence the nation’s biggest political decision.

The confusion is being aggravated by another basic obstacle: that each major source of data on campaign ad spending provides widely different figures.

Last week, we reported that two of the key sources of hard numbers on ad spending, the Federal Election Commission and the private research group Kantar Media, provided vastly divergent numbers and had broad differences in their methodologies. Today, we look at another leading source of campaign spending data—Smart Media Group, an ad-buying firm whose statistics on presidential ad expenditures are cited by NBC News and The Wall Street Journal—and find that their statistics deviate drastically from both Kantar and the FEC.

The graphic below (click to enlarge) shows the three sources’ respective totals for several outside organizations buying broadcast ads in the presidential campaign. (We examined buys between March 19 and September 9 in order to cover the same time period for all of the groups.) The discrepancies are staggering: for example, Smart Media Group’s ad tracking service SMG Delta finds that the pro-Obama super PAC Priorities USA spent $48.1 million over this period while Kantar found that they spent only $8.1 million, less than a fifth of SMG Delta’s total.




Sources: Federal Elections Commission, data accessed via the Sunlight Foundation, includes ad buys from March 19-Sept.9. Broken down by “Purpose” to isolate TV and radio ad spending. Kantar Media, accessed via the Washington Post’s Mad Money campaign ad tracker, includes ad buys from March 19 - Sept. 9. SMG Delta, accessed via NBC News “First Read,” stories on Sept. 10 and Sept. 12.

The result is that candidates and deep-pocketed influence groups continue to run ads amid a haze of uncertainty—and avoid the scrutiny that comes with a clear accounting of their activities.

As we wrote last week, there are a number of good explanations for the discrepancies between the FEC and Kantar. To name a few, the FEC tracks expenditures that include future ads while Kantar uses technology to scour the airwaves and capture spots as they run; the FEC includes all ad buys while Kantar does not include local cable stations; the FEC measures exact spending while Kantar employs estimates based on market rates. But why does Kantar vary so widely from SMG Delta? And why does SMG Delta in turn vary from the FEC?

Paul Winn, the political director of Smart Media Group, said its approach includes local cable stations and reflects the increase in ad rates that occurs shortly before an election. But he declined to offer further explanation of how SMG Delta gets its statistics. “We don’t really discuss our methodology, other than that we stand by the numbers we produce,” Winn said.

According to Elizabeth Wilner, Vice President of Kantar’s Campaign Media Analysis Group, SMG Delta obtains its statistics by leveraging its position as an ad-buying firm for Republican candidates. Broadcasters offer regular updates to ad buyers on the activities of other ad placement firms, which cumulatively offer a national picture of political spending.

“It’s a longstanding courtesy that stations provide to media buyers,” Wilner said. “He [Smart Media Group director Kyle Roberts] is turning around and selling that information to NBC News and anyone else who is paying him for it.”

Wilner said that SMG Delta’s approach would not capture whether an ad had actually aired. Some spots are purchased as cheaper “preemptible” advertising, whose cost is later refunded to buyers if the station exercises its right not to air them. Campaigns and outside groups also pull ads off the air at the last minute, in response to polling data or other indicators of the relative competitiveness of races.

“He’s basically tracking reservations, and we’re tracking what someone actually ate,” Wilner said of the difference between her group and SMG Delta. She added that many news organizations prefer not use data from a partisan source such as SMG Delta, which describes itself on its website as “the official media buying agency for the McCain-Palin campaign.”

There is another notable difference as well: according to an NBC News article, SMG Delta’s data includes radio ads in addition to television ads. Winn, Smart Media Group’s political director, did not return follow-up phone calls or an email seeking to confirm these differences in the two groups’ methodologies.

The vastly differing statistics from each of the leading sources on spending in the ad wars raises a basic question: What should reporters do to provide the best information to their audience?

One answer is to be precise and accurate in describing their sources of data. For example, Kantar statistics are estimates rather than exact totals, a difference that should be reflected in descriptions of them. There is also an important distinction between ad reservations and ads that have already aired, and this should be noted rather than simply referring to “TV ad spending.”

A second takeaway is that ad spending takes place in a volatile marketplace, in which rates rapidly change, many ads can be preempted and then must be refunded, and candidates and outside groups pull ads on short notice as they reallocate resources. Robin Kolodny, a political science professor at Temple University who studies campaign advertising, said a precise picture of TV ad spending will not emerge until early December when broadcasters disclose invoices that, unlike order forms, provide the costs and details of which political ads actually aired. “It’s much easier to do it after the fact then it is in real time,” Kolodny said. “That’s the real lesson.”

Finally, there is far more to be learned about campaign ad spending than simply overall expenditure totals. Unlike SMG Delta or the FEC, Kantar provides spot counts, which The New York Times yesterday described as a better measure of the ad wars than spending because it reflects an organization’s reach rather than the disparate prices that campaigns and outside groups pay for airtime. Kantar also offers a breakdown of the content of ads, which provides insight into campaign strategy and messaging. “Advertising isn’t just about the money,” said Wilner of Kantar’s campaign analysis team.

If you'd like to get email from CJR writers and editors, add your email address to our newsletter roll and we'll be in touch.

Sasha Chavkin covers political money and influence for CJR's United States Project, our politics and policy desk. He has written for ProPublica, the Center for Public Integrity, and The New York World. Follow him on Twitter @sashachavkin.