One sentence exec summary: New measurement of advertising impressions are helping drive performance for marketers.
When we purchase media, one of the major components of the buy is around how many impressions we will achieve with a campaign. The common thought here is that the optimal amount of impressions against a target audience will yield positive advertising results. This is fairly common thinking around Madison Avenue and has led us to using the Nielsen metrics as currency for TV, circulation metrics for print, and unique visitors as being important for digital media. The higher the rating, the larger the circulation and the more unique visitors a media property has, the higher the amount of impressions, theoretically* and thus, the more media spend that they can handle.
I personally and professionally believe that it’s not about the impressions, but about the engagements. Engagements can be quantified many different ways (thru sales to interactions) but just impressions alone is not a basis for marketing working or not.
In digital, there is a never ending supply of advertising inventory especially when publishers like this put stories across 5 or 30 pages. Creating more ad impressions is not the problem with digital. 3rd party ad serving systems such as DART, Atlas, MediaMind, MediaPlex, AdZerk, adk2, OpenX, and others were created to help manage advertising inventory for pubs and advertisers. Not only do ad servers help us serve media but they also help us track the performance which allows us to understand which media partners are doing well.
Historically, this was always about the amount of impressions served and the respective clicks or engagements on a unit. This post is not about clicks or click thru rates, as that’s an entire topic in itself, and you can read about it here. What we will discuss through the rest of this post is how we as marketers should be tracking an impression.
As you have viewed on many websites during your normal course of web surfing, banners and ad units are all over a respective page and sometimes, you cannot even see the units as they might be below the fold. Generally** as long as the web page and assets load (tags fire), then the advertising units on the page log impressions, regardless if the user scrolls over them or engages.
Sounds weird, right? An advertiser pays for impressions without confirmation that the user has at least had the ability to see the advertisement.
Since 2008, we’ve seen startups like DoubleVerify, AdXpose, and AdSafe (amongst others) help marketers understand what how many of their impressions are in-view along with many other non-standard ad server metrics. The in-view metric is important because that’s the number that should count for impressions (in my theoretical perfect world). Lately, ad serving solutions such as MediaMind have incorporated this metric into reporting but it’s not historically standard across 3rd party ad servers (which is shocking).
Today marks a big day for the digital media community because comScore released the vCE, which is what they are touting as “campaign essentials” and it’s built on a product they created and merged with AdXpose called the vGRP to validate the impressions served and displayed by publishers. The vGRP is very interesting because it goes beyond traditional measurement of 3rd party ad serving solutions to include things like time on screen and in-view/out of view. While admittedly I’ve not run a campaign yet with the vCE/vGRP, I hypothesize based on historical campaign performance we’ve run, that a paid media campaign will yield better performance when a higher composition of impressions are in-view and for no less than a minimum amount of time. Not rocket science.
Additionally, other companies are building advanced measurement solutions to help understand display impressions better. The Goodharts over at MOAT are building a brand analytics dashboard which help marketers understand how many people are hovering over their display advertisements. The theory here is that the more that a user hovers over an impression, the higher the engagement, and thus, the better a campaign will perform.
Evolution is occuring in the digital space. From the days of offline media based on circulation and rating points to a now “validated” actualized number of impressions being seen digitally. Over the next decade or so, we’ll see our online measurement systems (tools/technology/process) brought to traditionally offline channels. Then, we’ll be able to really understand our media performance.
* Why I said theoretically is because advertising dollars don’t always follow the amount of impressions. Some impressions are worth more, and potentially, you can spend more for fewer qualified impressions. Think webmd or other niche publishers.
** I say generally here because it’s more often than not
*** I lifted the above image from the website of MOAT.