GSA Instagram Facebook Twitter LinkedIn

INtelligent Direct: Was it Your Content or the Type of Impression?

A common metric for measuring the success of display advertising, social advertising and sponsored social posts is impressions. Impressions are a measurement of responses from a web server to a page request from the user’s browser. Counting impressions is how most advertising campaigns are paid for.

When you have high impressions but little to no click activity, most marketers first question if the content is to blame. It’s also very possible that the ad serving technology plays some cards in this game.

The current standard for impressions is a “served impression,” which means that as long as the ad was displayed somewhere, it counts as an impression. This means the ad was displayed on a page you were looking at, but you never actually saw it because it was very much buried in content and you didn’t scroll down or over far enough to see it. Worse yet, since an estimated 50+% of web traffic is made up of bots, those impressions served to bot traffic aren’t getting you engagement on your ad dollar either.

“56% of digital ads served are never seen by users” – Google 2014

If the current standard of ad impressions means the ad can be served in a way where real humans don’t see them, yet you still pay, then there is a LOT of room for improvement in the way digital advertising is being measured to make it fair for buyers.

“Viewable impressions” are ads at least 50% visible to the user for at least one second. This means you have a fighting chance of getting some engagement. It would be ideal if 100% of the ad is visible, or close to, to be considered a viewable impression, but it’s still an improvement from where served impressions are today. I think you would agree.

I recently promoted a webinar that we, Infinity Direct, produced. Our target audience was in LinkedIn, so we sponsored a post to reach non-followers, which netted us 812 impressions. This is small potatoes. We were testing the waters. What inspired me to write this is that even with 812 impressions of the webinar registration invitation, there wasn’t one click to the registration link. Not one. When I simply post the same content to our usual followers on LinkedIn, we get an average of one click per 34 impressions for the exact same webinar. I understand that our followers are more engaged with us, so I don’t expect apples-to-apples engagement with an audience that doesn’t know us, but going from 2.94% response for organic reach to 0% response for paid reach doesn’t make me want to open my wallet next time. It’s not a good Return on Ad Spend (ROAS).

This had me looking for answers to learn the difference between served impressions and viewed impressions, what type of impressions LinkedIn uses to measure impressions, and if any real humans were served my sponsored post. Unfortunately, I still don’t know the answers as LinkedIn uses its own method to determine whether an impression meets the viewability criteria. It’s also not possible to achieve 100% viewability in campaigns yet due to technical limitations with measurement.

What can be done?

Cue the 3MS (Making Measurement Make Sense) initiative to change the way digital media is measured, planned and transacted across the advertising industry to make it better for everyone involved.

The 3MS initiative was founded by the American Association of Advertising Agencies (4A’s), the Association of National Advertisers (ANA) and the Interactive Advertising Bureau (IAB). The Media Rating Council (MRC) is responsible for setting and implementing measurement standards.

These organizations are helping lead the change as the technology to measure impressions evolves and improves how many ads can be measured as viewable.

What We Use and Why

At Infinity, we use El Toro for our clients’ campaigns if digital advertising is a part of them. Their ad serving technology drastically reduces ad serving to fraudulent traffic by forgoing targeting by cookies. They also only serve ads to verified IP addresses of wireless networks that tie to a physical address with a statistical confidence level of at least 95%.

In hindsight, I would have had more traction with serving an ad for the webinar using a mapped polygon of business parks as my target audience and serving ads to them using El Toro versus using LinkedIn. Live and learn.

Jenny Lassi • September 19, 2018

Previous Post

Next Post