Originally posted in IAB SEA+India | November 28, 2018
Viewability is the measurement of whether a user viewed an ad served to them. There are many reasons that a user might not see an ad — the ad could have been served at the bottom of the page, or maybe the user left the site before the page fully loaded — understanding the viewability of impressions will help advertisers understand the media they are buying, and publishers the value of a particular placement.
The viewability rate for desktop and mobile display is currently defined by the Media Rating Council (MRC) as follows: 50% of ad pixels must stay in the viewport (visible area of a web page on a device) for 1 continuous second. For video, a minimum of 50% of the ad must stay in view for a minimum of two continuous seconds.
Based on this definition, IAB SEA+India’s Uncovering Viewability in Southeast Asia report: Display Advertising released in 2017 found that the average viewability rate in Southeast Asia was 53%. This came close to the US-defined standards for desktop display advertising.
Some advertisers also choose to look at stricter definitions to suit their brand objectives and needs – such as a display ad impression being valid and fully on-screen (100% pixel in view) for a longer length of time, or the video impression being valid, audible and viewable on completion. This article will not address whether the current MRC definition should be challenged, but rather the state of measurability of inventory in Southeast Asia.
Whilst it can be important for an advertiser to achieve a good viewability rate, it is also imperative that a high proportion of impressions served is actually able to be measured, and the sampling rate of those measurable impressions is understood.
TECHNIQUES TO MEASURE VIEWABILITY
There are multiple technology vendors providing radically different technology to collect viewability data but each of them has limitations.
Individual telco providers do also have the capability to profile their subscribers’ location patterns over time using rich, always-on methods. The location behaviour can then be made available to advertisers in anonymised and privacy-compliant methods.
This is the most commonly employed technique for measuring viewability. It uses geometric information gathered from the page, including the size and relative placement of the page, browser viewport and ad to determine viewability. Though this is a really simple approach, it suffers from many technical limitations, particularly if the impression is delivered within nested cross-domain iframes. Using cross-domain iframes are the common approach used by publishers to prevent ads and other external content from intentionally or inadvertently interfering with their websites.
This technology infers viewability for impressions by detecting user actions on the ad. For example, by tracking mouse movement one can infer that any ad where the mouse passed over must have been in view during that moment. However, it introduces biases related to the specific placement of the ad. For example, ads located in the upper left corner of a website may be inadvertently moused over far more frequently than ads placed in the upper right-hand corner.
This uses flash technology where you would drop a pixel into the center of the container and then look for the frame rate responses. The changes in frame rates when a user is in viewport/out of viewport is what is used to determine viewability. However, unlike the Geometric method, it is hard to ascertain exact viewability unless you launch multiple flash pixels across the container. Since almost all browsers stopped supporting flash, this technology is almost obsolete and rarely used.
This method works by placing a pixel within the publisher’s website that executes on every page view and makes available an API that can be accessed by other parties even within cross-domain iframes. This is also the basis of the IAB’s SafeFrame technology that could replace iframes, offering the same security but allowing access to the geometric data needed to compute viewability accurately. However, it requires significant work by publishers as well as ad servers to overhaul their existing system to support this new framework. So, unless there is a significant push from media buyers, it will be some time before this technology is widely adopted.
Most browsers can detect what content is currently in view and can optimize the content not in view to conserve device resources. These optimization techniques can be employed by publishers and ad servers in their code to accurately measure viewability of the Ads. This approach may be the most reliable, as it piggybacks off of the very accurate viewability data that the browser is already collecting. However, this method is very difficult to execute as different browsers will have different approaches to optimizing content rendering that will vary by circumstance, operating system and device. This means publishers and ad servers not only need to ensure compatibility with different browsers, they also need to keep updating their technology with browser version changes to ensure quality as well.