Our Worst Decisions

Working in interactive advertising provides plenty of opportunity to make mistakes and to have those mistakes look embarrassingly obvious in retrospect. It just goes with the territory. You make a brilliant prediction, it comes true, and people mistakenly believe that everyone knew that was going to happen. Your prediction doesn’t come true, and everyone seems to think that only you were clueless.

In fact, the “common sense” opinion in the industry, while given great deference by businesspeople and columnists alike, is usually wrong. Remember the imminent dominance of “push” technology? Or the “death of the banner”? Perhaps you can recall everyone predicting with confidence 3 years ago that it would take online advertising 10 years to surpass the billings of outdoor and radio. (It happened in 1999 and 2000.)

Oddly, rather than put a great deal of scrutiny on industry consensus, we cite it in our PowerPoint presentations as the rationale behind our media strategies. This can be dangerous.

Personally, I’d prefer media planners write that they believe something because of a hunch and a couple of corroborating facts rather than have them merely cite a Forrester study from a few months back (nothing against Forrester). But I know I’m asking for too much. The media culture relies too much on the perception that conclusions are more fact-based and less opinion-based than they actually are.

This perception has led to some very bad decisions in our industry. Taking this “common knowledge” approach frequently burns us, particularly on the buy side of the industry. Here are some examples:

• Employing banner servers as counting mechanisms.
Since sites use different definitions and different metrics to count impressions and other media data, we need some sort of consistent reporting mechanism. Early on, it became tempting to use banner servers for this purpose, and now we’re pretty much stuck with them. Rather than forcing sites to adopt a common definition of performance and to be able to automatically get that data into an agency database, we punted. This not only adds another 10 percent to the cost of the media buy for our clients — we agencies don’t pay for this — but also limits our data.
• Allowing vendors to become infrastructure partners. This would never be allowed to happen in traditional media, but somehow the youngsters got away with it in online advertising. Many of the technological infrastructure providers to agencies (i.e., the companies that develop the internal trafficking and management systems) have been bought up by web sites and web site networks. Whereas a few years ago, agencies would have dropped those media vendors because of conflict of interest, for some reason today the consensus holds that this straddling of both sides of the fence provides a vendor with greater stability. Yes, it does, because the vendor is taking advantage of you. It would be the equivalent of an accounting package in traditional media — like Donovan — selling its own media.

• Buying larger sites and networks rather than the little guys. Since the very beginning of the online media industry, we’ve always had difficulty purchasing interactive media with any decent economy of scale. As a result, it’s been tempting to keep to the larger sites and networks, making fewer purchases for more dollars. Even then, we have to put more labor into an online buy than into a traditional media buy. But the larger media properties have taken advantage of this, raising their effective rates and soaking up any benefit the larger scale would have provided.

  Here’s the rub: When a media property raises its prices, the client pays that difference. When a transaction is less efficient to conduct, the agency pays that cost. As a result, agencies are making clients pay higher prices to give themselves better margins. What would happen if we looked merely at the efficiency of the media? We’d be buying little sites almost exclusively. But, somehow, the persistent trend toward the larger sites has made this behavior the accepted norm for the industry. It would look strange, if not downright nonsensical, to eschew the large networks and sites. That would be against the industry consensus.

If I were to make a generalization about some of these types of poor decisions, I’d have to say that the buy-side crowd, and particularly the agencies, tend to move in packs. We’re not comfortable doing something others aren’t doing, so we frequently run to join the lemming herd and then scurry along in the same direction.

Another trend, perhaps the causal factor here, is that clients tend not to reward the agencies that do things differently. I think this may be so because many midlevel managers who handle the agency relationship aren’t sophisticated enough in interactive media to understand when a policy harms their interests. They, too, follow the pack.

The good news is that there are agencies out there that do things a bit differently. And there are clients who demand this higher level of service and independence. The trick for most agency buyers out there is to gauge both their clients’ increasing level of savvy and the flexibility of their agency policies. Don’t be afraid to adapt as your sophistication grows and as your clients’ sophistication demands.