• by Joe Marchese , May 15, 2007

The Internet is one of the great expressions of the free market. For any clearly defined demand, you will see an overwhelming rush to supply. Due to the Internet's ability to spawn innovative solutions at breakneck speeds, the rush to create supply commonly produces a massive surplus. Due to the reactive nature of the Internet economy to continuously and immediately oversupply any clearly defined demand, advertisers and advertising technologies have played, and continue to play, a defining role in crippling the Internet by creating demand for impressions.

It's generally agreed that the impression is dead -- yet I haven't seen it really going anywhere recently. Even when talking to some of the most progressive thinkers in the space, any conversation around Internet advertising quickly reaches the question "So what's my eCPM?" While anyone with common sense can tell that the impression is a virtually meaningless metric, there is nothing else for comparison across media. The problem is that the market for impressions in digital media has caused a fundamental misalignment of goals between advertisers and publishers.

The ever-increasing association between advertisers and the content they advertise on introduces significant challenges to advertisers in the digital generation. One of the greatest of those challenges is the need to completely reevaluate the quantitative method for determining the role the surrounding content plays in determining the value of delivering an advertising message.

For most forms of "traditional" media, the best way to generate impressions is to create high-quality, engaging content appropriate for mass audiences (which commonly means brand-safe content). Unfortunately, there are many ways to create impressions that have nothing to do with quality content and engaging experiences. As a matter of fact, rather than spend the time and effort to create quality content, it is commonly cheaper to engineer pages, perform click arbitrage, SEO Web sites, spam emails and generally trick people to generate more impressions, clicks, leads etc. This results in a significant portion of the Internet consisting of content unfit for brand advertising and providing little or no benefit to users.

Even when a brand manager can buy the influence he needs through a quality medium, the eCPM will appear horrid compared to the volume of impressions many of the ad networks can provide. The marketplace for impressions places quality content producers in the unenviable position of competing with click farms and low-quality impression inventory, incenting even quality content producers to play the games that will increase impression volume at the expense of the user's experience.

Even when comparing opportunities in quality digital content, the impression is practically meaningless, because even two pieces of quality content don't necessarily offer the same brand lift to any given advertiser. This is because the advertisements are in the content, rather than between the content. The need for content relevance demands that the advertiser, at least in part, leverage the context of the content to build a brand message. The result of this is that an impression within content X may be worth one amount to Advertiser 1, and a very different amount to Advertiser 2. The impression is woefully inadequate for accounting for this.

Impressions have their role in the next generation of advertising metrics, but as a component. I would argue that any meaningful metric will account for volume of impressions, quality of content and alignment of content with brand advertiser goals. Combining these variables will create the demand market for influence needed to correctly align publishers' and advertisers' goals. We must unlock digital media's potential as a brand advertising medium in order to support and sustain the development and distribution of media in a digital world, but first we must determine how we will measure what we are unlocking.

What elements do you feel need to be included in digital media's new measurement metrics?