Media/Technology Executive

I had an interesting conversation with an ad reseller we do business with this past week. They called to check in and see how our partnership was going. Specifically, they asked why our sales seemed to be lagging year over year. When taking a closer look at our revenue, it’s apparent we are mostly selling run-of-site ads with geo targeting. Our behavioral targeting ads (which allow advertisers to focus on a niche audience based on previous website activities within a section or topic) have greatly diminished.

A common objection I’ve heard to selling behavioral targeting is that “it’s expensive and results aren’t much better than run of site or our own inventory.” When asking what metric is used to determine success for banner ads, the overwhelming response is click-through rates.

Several interesting articles have been written lately on click-through rates and how they’re given too much relevance in measuring the overall success of an online ad campaign (Click Through Should Not Matter).

The problem runs deeper than just ad click throughs.  Today most content companies measure success of the individual journalists based on uniques and page views.  Both criteria driven by ad sales.  In reality most of a content site’s traffic is driven by a small audience.  The old 80/20 rule.  Building new and expanding an audience are more critical than driving page views.  Here is a paragraph out of Columbia School of Journalism I think is really relevant.

“This shows how misguided current measurement systems are. Today’s obsession with the “Unique Visitor” metric drives the advertising market —and competition among news sites. Such fixation encourages an arms race in which, by all means necessary (games, fake URLs), news sites will shoot for an increase in their numbers of UVs and for the resulting ranking improvement. This is short-sighted: loyal readers—roughly the top 10% that will generate 80% of the page views —should be the measure of choice.” (Full Article)

Some facts about click-throughs: a small percentage of people actually click on ads, and in many cases they are clicked on by mistake. Even in cases where click-throughs are expected, the ads often do not have a clear call to action to encourage such activity. When click-throughs do occur, true success should be based on actions taken, such as time spent on referred site or a purchase on a deals page – and not the click-through itself.

In the case of our partner, we are primarily using click-throughs to determine the success of behavioral targeting instead of stressing the value in delivering quality brand impressions to an audience our advertiser seeks – even if people do not click through to their website or make a buying decision right at that moment.

If a person is going to Web pages that relate to buying a car, for example, then we know behavioral targeting ads about cars will be relevant to that person. And we know these people will be more valuable to advertisers.

Getting the right people to see your ad is critical, which means we can’t give up on behavioral targeting. In other words, it’s the overall quality of audience that matters most – and not quantity of clicks. Advertisers are already shifting their ad dollars. For now, ROS display ads still rule, but according to Borrell Associates, by 2015 the lion share of local online ad spending will be going towards targeted display advertising.

It comes down to the ability of my company and the entire ad industry to change the metrics used to measure success. It won’t be easy, but it’s necessary. So how can we address the problem?  We need to identify a core group of advertisers that are forward thinkers and see the world as something more than traditional page views and uniques.  We then focus on engagement factors with an audience they seek.  We create metrics around time spent with content and driving interaction.  There will need to be specific actions we seek that relate to an ultimate purchase or interacting with brand.  Ultimately we will need to show the advertisers insight about the audiences they are seeking.  As long as we show a level of engagement and specific actions were driven, we should not get into a page view and uniques discussion.

Specific to our own efforts with an iPad app we are seeing high engagement, especially with our feature stories, which are unique for the iPad.  We consistently see users spend more than 2 minutes per feature.  There are several studies that suggest people spend a lot more time with content on the iPad, here is one of them as reported by PaidContent.com.  As it relates to ad engagement on the iPad, the story is the same.  Here is a case study about a successful iPad campaign for Land Rover as reported by MobileMarketer.com.

I know it sounds foreign, but if we don’t move in this direction then we will forever be selling at a lower cpm ad unit.

(Photo courtesy of Davide Cassanello)