Contact us

Is viewability still a valid metric in 2019?5 min read

5 min read

The viewability debate has provoked a high level of ad nauseam in the display world for years now, with the argument coming to the fore in 2013 when Comscore released comprehensive analysis suggesting a staggering 54% of online display ads were not viewable.

Yet despite this, the viewability debate rages on today. The floors of agencies and trading desks continue to be chock-full of marketers arguing the value and validity of this widely-contested metric, with some practitioners championing it, and others shunning it to the bottom of their KPI list. The truth is, it’s almost impossible to verify the accuracy of viewability metrics in any given media environment, be it online or offline, but does that mean it’s no longer a valid metric?

In the following article, Croud’s programmatic team assesses both sides of the divide, weighing up the pros and cons of viewability as a valid metric in 2019.

Here to stay

Connie del Bono, Programmatic Associate Director

In a modern world where it’s possible to spend our clients’ budgets with the flip of a switch, it’s fundamental that we approach this with extreme caution and implement all facets that we have the potential to control. Viewability is one of these metrics and, in my opinion, should always be implemented. This is not solely for our clients’ marketing budgets, but also for each individual users’ experience of brands that we are representing. The industry is constantly pivoting and although viewability in a results-driven industry can seem redundant at times, we have to believe that these viewable placements are worth more to both the user and the brands.

Rebecca Millward, Programmatic Media Executive

Naturally, every ad we serve we want to serve in a ‘viewable’ place. I believe that efficiency is greater than an ad which is situated in a less viewable place. If the ad cannot be seen then the ad does not impact or influence that user, and if we didn’t optimise towards viewability we wouldn’t be able to ensure our clients’ ad is being seen. Therefore, this ad would be a waste. Finding the ideal balance is essential to get the maximum value from your budget, achieving the maximum reach at the lowest CPV impression.

A re-think is needed

Henry Mansell, Programmatic Media Manager

The debate needs to pivot. We need to move away from the wider question of whether or not viewability is a valid metric, and towards whether or not the current and widely-adopted definition of what constitutes a viewable impression is still fit for purpose. The Media Rating Council, building on existing guidelines developed by the IAB, released their ‘Viewable Ad Impression Measurement Guidelines’ in 2014. The report states that for a display ad impression to be deemed viewable, two key criteria must be met (in the following order):

  1. Pixel requirement: Greater than or equal to 50% of the pixels in the advertisement were on an in-focus browser tab on the viewable space of the browser page.
  2. Time requirement: The time the pixel requirement is met was greater than or equal to one continuous second, post ad render.

If the above criteria were to be applied to any other form of media, be it online or offline, it’s highly unlikely that seeing 50% of an ad for one second would be enough to drive any real influence on a prospective customer. The industry needs to reassess both the notion of viewability, and the prerequisites needed for an ad to be deemed viewable.

A viable alternative is Avocet’s view-duration metric. The platform measures the time, in seconds, that 50% of a display ad spends viewable to a user online. Avocet allows advertisers to set predicted view-duration thresholds within their media buys, a more useful lever that brands can employ to ensure awareness and exposure.

Omri Kedem, Senior Programmatic Media Manager

When it comes to evaluating the pertinence of viewability, there exists a vast number of elements and factors to consider, but ultimately: KPIs come first. In a performance agency, viewability as a metric will by no means define the success of your campaign. Fundamentally, programmatic trading desks are tasked with reaching the primary KPI. In Croud’s case, this is predominantly direct-response-focused in the form of ROI (return-on-investment), CPA (cost-per-acquisition), or the volume of any given form of acquisition. Which probes the question, should we let viewability inhibit our potential to maximise returns for clients? I don’t believe we should.

If we are going to evaluate viewability, we should really be prioritising vCPM (viewable cost-per-mille). vCPM is a far more valid and representative metric to both measure and optimise towards. This is because rather than just looking at the percentage of viewable impressions served, vCPM ties this back to the empirical volume of viewable impressions that were delivered. Thereby resulting in optimisations which are far more effective in delivering results, primarily due to consumers being more likely to convert if they are served a viewable impression.

However, it’s important to mention that when it comes to viewability and vCPM, marketers need to be smart with their optimisation strategies. If you want to ensure a strong viewability threshold throughout a campaign, you should not be using DSP (demand-side-platform)’s or third-party’s pre-bid viewability filter as the dominant lever. It is far more effective, from both a scalability and performance perspective, to utilise traditional optimisation methods including App/URL bespoke white/black-lists, browser and custom exchange exclusions; then combining these signals with strategy-specific bid multipliers for a more dynamic approach.

Whilst the debate on the validity of the viewability metric continues to spark conversions within programmatic teams across the globe, one thing that remains clear is, in the coming years, advertisers will be focusing more on understanding how to optimise exposure to drive the best results.

To find out more about our programmatic services, or about Croud, get in touch.