• by Tig Tillinghast , May 2, 2002

Jupiter Media Metrix has “discovered” that fewer, larger media buys in a campaign are less efficient than many, smaller buys. My first reaction to the study’s press release was, “No duh!” But upon further reflection, this study-of-the-obvious might actually have been quite necessary.

The matter goes back to the early days of online advertising, when we first discovered that larger buys gave us worse performance. But, as time progressed, this inefficiency worsened rather than improved. The underlying causes multiplied over time.

The Over-Delivery

In the mid-1990’s, banner server control systems were so rudimentary, that each individual buy would be subject to a 20-40% over-delivery, just to make sure the right amount of media was delivered. If sites didn’t do this, too many of their buys wouldn’t make their guaranteed impression rates for the time period. A site that received 1,000,000 impressions a day would put an ad up for at least 12 days to deliver 10,000,000 impressions, lest they discover at the end of the month that they owed a make-good with a penalty. The lower the amount of media bought, generally, the higher the over-delivery percentage.

One enterprising media supervisor who worked for me back then took this to an extreme, parsing out a major campaign for Microsoft over many, many sites with tiny media buys. He did succeed in multiplying the over-delivery by sites, but karma came around and bit him on the behind by also multiplying the effort needed to husband that campaign to completion. Worse, the client then got spoiled by the results.

It became a rule of thumb that the lower the average buy size, the better the price and greater the efficiency.

Small Sites have Superior Targeting

While we once believed that the larger sites would be the ones to first implement sophisticated targeting systems, the industry has instead reverted to the same type of content and context-related targeting that we’ve always done in print. Because of this relatively primitive targeting remains the state-of-the-art in online media, the smaller sites have an advantage in targeting efficiency. They are precisely the niche sites that allow for finer audience resolution.

More Congenial Supply/Demand Ratios

Smaller sites have always struggled to make money, and their relative desperation has made their prices lower. As ad agencies tend to be lazy and purchase larger amounts of media from larger sites, the supply/demand curve for smaller site media has become distorted toward more favorable prices for buyers.

I naively believed this was a temporary trend, as smart buyers would see the opportunity and bid up the efficient media. I hadn’t counted on the next few factors, however.

Large Media Company Hubris

The large online media companies like Yahoo!, AOL and Microsoft, began to aggressively market themselves to large brand advertisers in the late 1990’s. Rather than sell their online media based on complicated response metrics or targeting options, they tried to sell enormous and expensive media packages stressing their brand names rather than their media efficiencies.

In an irrational environment like that of the late 90’s, these media packages actually sold. It became a trend that fed itself. When a big media company like Yahoo! sold a 5-year contract for $100 million, the stock prices of both Yahoo! and the buyer went up. Just the fact that these large brand companies were seen as starting to use the Internet was enough to jump their stock. Lost in this fervor was the fact that they were paying through the nose for untargeted media.

This set a precedent of very high prices for the top site brands. It concentrated the brunt of the industry’s media dollars into a few of the largest hands, and stifled a great deal of media diversity.

When the bubble broke, and people had to start worrying about making money again, I expected a movement back toward the smaller sites. What I didn’t count on was that the large brand advertising companies were exactly the organizations that still don’t have to be accountable for their results. Since when was branding accountable? Rather than admit they were wrong to spend most of their money on three sites in year one, many just renewed the contracts.

The direct response business and other, smaller companies more directly and visibly dependent on media efficiency results have moved toward smaller sites, but they represent a smaller and smaller proportion of the gross ad dollars over time, as the large brand budgets begin their overdue media shift toward online.

Glamour

Finally, of all the media companies in contact with buyers, these same large media companies are the glamorous ones. They have the skyboxes at the big games. Having good contacts in those organizations will help a buyer’s career more than making hundreds of other reps at small sites happy.

The lunches are great, the reps wear very, very expensive clothes. They’re like the clique in high school you told everyone you despised, yet felt flattered anytime one of them spoke to you. When buyers have media dollars to drop, they can purchase a little bit of that “in crowd” feeling every time they purchase a high CPM deal with a top ten site.

So, looking back, maybe it’s an important thing that JMM’s study underlines this silliness. Maybe now that we can quantify it, buyers may be embarrassed into being a little more rational.