Viewability: Break Through the Noise

Recently, a client remarked “Well, there certainly is a ton of noise in the industry around Viewability right now.” For media buyers, it is nearly impossible to determine up from down when discussing this topic with measurement vendors - nested cross domain iframe this, javascript that, pixel over here, guarantee over there. The problem is that this is a fairly new metric for digital video and unfortunately, some are trying to take advantage of the situation with carefully worded marketing and technology reliant on smoke and mirrors. 


So, what questions should you ask any vendor claiming to measure or guarantee against Viewability and what are acceptable answers to those questions? Below you’ll find a list of key questions and expected answers which should help us all break through the noise.


Are you accredited by the MRC for Video Viewability measurement?
Right now, there are only three vendors accredited by the MRC for Video Viewability measurement – Videology, Moat and Tremor*. Accreditation is a critical achievement because it connotes adherence to the established standard and submission to independent audits. 


What Measurable Rates are you seeing? What Viewable Rates are you seeing?
First some definitions which every vendor should be following (but some are not):

  • Measurable Rate: (Viewable Impressions + not Viewable Impressions) / Total Impressions
  • Viewable Rate: Viewable Impressions / (Viewable Impressions + not Viewable Impressions)
    • Note that this is different from Viewable % (Viewable Impressions / Total Impressions)

Based on our analysis of over 3 billion impressions this year, a Measurable Rate of 45% or more is above average and a Viewable Rate of 50% or more is above average.


Are you able to determine whether or not the ad ran in the active browser tab?
This can actually impact Viewable Rates more than you might think. This is because some publishers will begin new content immediately after the requested video. This means additional ads run, and if the user has tabbed away, none of those subsequent ads are viewable.


Are you able to measure Viewability when the ad runs within an iframe?
The unfortunate fact is there is not technology that can accurately measure Viewability within a cross-domain iframe. Many purport to do so, but none are able to do it accurately. How do they claim to do it, and why doesn’t it work? I’m glad you asked:

  • Pixel – A pixel is inherently ‘dumb’. It cannot measure frame rate. Let’s take for example, an 800x400 video – that’s 320,000 pixels. How could a single 1x1 pixel be used to determine whether all of the other 319,999 pixels were in view? It can’t!
  • Companion Banner – To assume that because a companion banner was in view means that the entire ad was in view is fraught with peril. In addition, running ads alongside companion banners creates an artificial supply constraint.
  • Flash – Flash is unable to break out of an iframe - it suffers the same fate as javascript.

We are actively testing proprietary technology which will accurately measure within an iframe (and we expect others are as well), but so far nothing meets our stringent standard for accuracy. 


Do you offer guarantees against Viewable Rates? Do you offer pricing based on Viewability (e.g. Cost per 1000 Viewable Ads)?
Wouldn’t this be an effective way to transact on digital video? The truth is that the industry is just not ready for either of these methods yet.  The key reason is that if you are only buying Viewable ads, you are only buying measureable ads. This means you are cutting out about 40% of inventory off the top due to the fact that they are on VAST and not VPAID. Many VAST publishers include FEPs, broadcasters, and other Tier 1 inventory. They haven’t yet had motivation to move to VPAID because they get big CPM’s and have plenty of demand. This will likely change over time, but we aren’t there yet. 


At Videology, we offer the use of MediaLists which contain inventory with historically high Viewability performance. We also offer the ability to set Viewability as a campaign level KPI where our optimization engine will make real time adjustments to route more impressions to high performing sites within your sitelist. We offer either option at no additional cost, though CPMs may be slightly higher due to the natural supply constraint. Using these options has been generating Measurable Rates above 80% and Viewable Rates above 65% for our clients. 


For independent third parties – are they truly independent?
Unfortunately many of these companies are actually owned by publishers or have tight partnerships with networks or DSPs. The key question should be accreditation, not perceived independence. 


The questions above are merely the tip of the iceberg. Within the next few weeks, we’ll be publishing around the related topic of Fraud Risk and Prevention. The hope is that in the near future additional vendors will achieve accreditation because a rising tide lifts all boats. Carefully worded marketing and dishonesty about capabilities only furthers advertisers’ existing caution around shifting TV spend to digital. Once we’re all honest with each other about the state of Viewability measurement, the industry can easily make giant leaps in rates. The focus should be on education and transparency - only then can we break through the noise.


* Source: http://www.mediaratingcouncil.org/Accredited%20Services.htm

Keep Reading...

Click to get Videology insights delivered to your inbox
© Videology, Inc. All rights reserved Member of or accredited by: