Upfront Preview: Behind the Bust

It seemed like a good idea at the time back in 2005—scrapping the upfront marketplace and instead making national TV time available on an electronic auction platform for advertisers to bid on. At least it did to Julie Roehm, who was then director of marketing communications for the Jeep, Dodge and Chrysler division of DaimlerChrysler, and a handful of other big national marketers.

Roehm commanded a lot of attention when she presented her radical proposal for rearranging the TV buying process at the Association of National Advertisers TV Forum five years ago—as would any executive with roughly $2 billion in budgetary clout. But it became clear quickly that the major broadcast networks weren’t interested for fear of devaluing their inventory. Randy Falco, a top NBC executive at the time, called the suggested switch “ridiculous.” Other nets made it clear they felt the same.

Ultimately, most of the major cable networks also refused to go along, citing similar concerns. The tests that followed in 2007 with eBay, which was chosen over Google after an RFP process to build an auction-based national TV trading platform, followed a game plan with drastically reduced goals. Replacing the upfront was by then no longer in the cards, and marketers tried to position the auction option as an alternative to the scatter market.

But even with those scaled-back ambitions, eBay and a core group of marketers joining Chrysler that included Home Depot, HP, Intel, Toyota and Brown-Forman couldn’t generate much excitement from the rest of the industry for the auction proposition.

In June 2008, after little more than a year of trials, eBay called it quits with a terse statement that shed little light on what went wrong: “We have been refocusing our resources on our core marketplace business and have determined that this initiative did not warrant continued time and attention.”

Around the time that eBay rolled out its system—spring 2007—Google TV launched its own TV auction platform using Dish TV inventory and, importantly, access to Dish set-top box viewing data. Google TV has since added a limited menu of inventory from some NBC Universal cable properties and a handful of other cable nets such as Game Show Network, Hallmark and CBS College Sports. That system is still up and running, but it’s widely perceived as a market for smaller advertisers with shorter-term planning strategies and direct response ads.

Google declined to disclose what percentage of its ad sales are for DRTV, but that’s precisely the market that some agency executives believe the auction model could successfully serve in the national TV space.

“DRTV might be the best application,” says Steve Farella, CEO of New York-based TargetCast tcm. “The way it is sold and bought now is in fact a manual exchange.” Farella estimates that perhaps 5 percent of national cable inventory is slotted for DRTV. Since “it’s hard to call everyone,” he adds, an automated auction exchange would benefit both buyers and sellers.

DRTV market aside, why is “auction” such a dirty word in the mainstream network TV space? Lots of reasons, buyers say. “There’s a place for auctions but not for offline,” says Rino Scanzoni, chief investment officer, WPP’s GroupM. “Auctions work when you have extremely large supplier and inventory bases. You also need limited parameters; you can’t weigh 30 different factors.”

In the online world where there are 20,000-plus commercial Web sites, Scanzoni explains, “That creates an abundance of supply that can’t be absorbed by the advertising demand for those sites, so the opportunities for exchanges become more significant.”

In the national TV space, where the supply of inventory is much more limited, an auction system would have the effect of raising prices—clearly not in the interest of clients. Scanzoni also notes that network TV clients each have separate pricing bases, which have been established over time and are tied to factors such as the number of years a client has been doing business with a network, the size of inventory packages and to what extent they are spread over different properties within a media owner’s portfolio.

Roehm, now a strategic marketing consultant, has no regrets about her upfront pot stirring. Quite the contrary: “It’s one of the things I’m most proud of doing,” she boasts. She acknowledges coming out “guns ablazing” at the process when she proposed a new TV ad time   buying model based on a Nasdaq auction-type process—an approach that she admits now was ambitious.

But the bigger point, notes Roehm, was to get a dialogue going to address problems that many marketers had (and continue to have) with the upfront process, including the lack of flexibility, arbitrary price increases and the fact that advertisers are often forced to buy time in lower-rated and unwanted programs in order to obtain access to premium content. “How it all shook out was less important,” she says. “For me it was more important to get like-minded people to change things up a bit, or at least give it a try.”

Roehm came to realize that some big long-time advertisers who had relatively low network prices would not accept the auction process. Procter & Gamble, she notes, wasn’t interested: “They felt they were going to get a better deal by not having to play with everybody else.”

And though she now believes that auctions are not essential, automated exchanges of some sort could help wring waste and inefficiency out of the buying process. “The process hasn’t evolved to the point I personally would have hoped for,” Roehm adds. “But the medium is changing. Look at the money being spent in digital now. That’s struck a balance that didn’t exist five years ago.”

Others agree, including a number of agency executives who see a role for automation moving forward. “We are on the eve of a significant reinvention” of the TV buying process, explains Tracey Scheppach, svp and video innovation director at Publicis Groupe’s Starcom. It’s a reinvention, she adds, that is based on data, addressability and automation.

And that’s why Scheppach says “there are pieces of the Google TV system that we love.” It’s just not the auction part, which is a non-starter for big long-time TV advertisers whose pricing leverage accumulated over time “goes right out the window.”

That said, Google’s data “is fabulous” adds Scheppach, “and they have outstanding analytics. But the flexibility to move off the auction model to a more advantageous business model just isn’t there.”

Google has no plans to move away from the auction model anytime soon, says Steve Stukenborg, senior project manager, Google TV Ads. “It’s really the one way you can buy Google TV Ads, and it’s been really successful. We really believe in it.”

It’s hard to tell how successful it’s been because Google TV does not release sales data for the service. Stukenborg says he understands that many existing advertisers have historical pricing relationships with networks. But part of Google’s mission is to bring new advertisers to the medium who don’t have those relationships. It says about 30 percent of its clients are first-time TV advertisers. “We borrowed the auction model from online and online technology,” he notes.

Some agency executives counter that that’s the problem. “The essence of TV is completely different from the online model where you can continue to add space and optimize page views,” and where demand doesn’t outstrip supply, says Craig Woerz, managing partner at planning and buying agency Media Storm.

Brand advertisers in traditional TV frequently require certain pod positions or particular programs or times coupled with a precise mix of networks. That’s an environment where auction models don’t compute, adds Woerz. “As an industry, we sometimes tend to gravitate to the new shiny thing until it’s not shiny anymore, and then we head to another one. That’s what the whole ad auction thing was all about,” he says.

Antony Young, CEO at Publicis Groupe’s Optimedia, agrees with Scheppach that the industry is ripe for reinvention, which he argues hurts the prospects for the widespread adoption of an auction model. “Auctions work when things are stable, but clients are questioning whether we’re doing it right,” he contends, noting issues such as shortfalls in the current  ratings system, increasing viewer fragmentation and buying ad packages across multiple video platforms. “The auction mechanism isn’t really going to address all those issues.”

But like Farella, Young believes there is a limited role, potentially, for an auction model in the TV space. He likens it to the different choices available to buy stocks, with some traders preferring full-service brokerage firms while others are satisfied with E*Trade. “If you’re a marketer and you know what you want to do and want to get there at the lowest possible cost,” says Young, an auction exchange could work.

Roehm wasn’t the first executive to propose an auction system for buying TV ads. Before it imploded a decade ago, energy firm Enron proposed its own auction setup, which never got off the ground. With little support from major broadcast and cable networks, the model would seem to have limited application for mainstream national TV.

Of course, with Google involved, one can’t easily dismiss the effort. After all, the company knows a thing or two about revolutionizing certain media sectors.