Mobile ad viewability: what is it and does it matter?

Brand advertisers and their agencies only want to pay for mobile ads that are seen by a person. They do not want to pay for ads that are not ‘viewable’ i.e. they don’t render or are displayed where they are invisible; or for insufficient time to register the message.

With brand advertising a mobile ad does not count as an impression and the advertiser thus should not be charged if:

  1. The ad is not seen by a real person;
  2. Because it fails to render or is displayed where it is unseen, e.g. below the fold, when the visitor doesn’t scroll; or
  3. If there is insufficient time for the visitor to digest its message.

• The first factor is an invalid impression, often caused by ad fraud. These were discussed in the previous two columns.
• Factors two and three are referred to as viewability, viewable ads or in-view ads and will be discussed in this column.
• Brand safety or brand adjacency – i.e. the placing of content next to appropriate content, will be discussed in a subsequent column.

This viewability standard is very new to mobile. And, in practice, ensuring ads qualify as a chargeable impressions, is even more of a headache on mobile than desktop, especially for native app advertising.

Viewability is an issue on many levels:

1. Digital ad viewability, generally, is a contentious political issue.

It has led to some high-level industry bickering between advertisers, agencies, intermediaries and publishers. For example:

2. Measuring viewability on mobile is even more difficult than on digital.

  • The industry had to wait until June 2016, before there was any standard for mobile viewability. See MRC Guidelines below.
  • These standards are for the US, and do not necessarily have international scope. The viewability standard for mobile in the UK, for example, is still a work in progress, headed by JICWEBS (Joint Industry Currency for Web Standards).
  • Quantifying viewability on mobile apps is even trickier than mobile web.

3. Viewability isn’t the same as ad quality or effectiveness. It does not guarantee results.

  • You can thrust the wrong ad creative, in the wrong format, for the wrong product, in the wrong person’s face all day and they still won’t buy it.
  • Striving for max viewability can drive advertisers / publishers to use more intrusive ad formats, such as interstitials, which can alienate visitors and encourage ad blocking.
  • Viewability is more of an issue for brand advertisers, who pay CPM (cost per thousand impressions). For performance advertisers, who pay by results e.g. acquisition of a new subscriber / installation of a mobile app, it is far less of a concern than ad fraud.
  • Viewability is pointless if there is no guarantee that it is a human rather than a robot viewing the ads.

Definition of mobile ad viewability

MRC Mobile Viewable Ad Impression Measurement Guidelines (PDF), published June 2016 offers the following definition.

Mobile viewable display ad impressions are counted when the following criteria are met:

  • Pixel requirement: Greater than or equal to 50% of the pixels (density – independent) in the advertisement were on an in-focus browser or a fully downloaded, opened, initialized application, on the viewable space of the device, and
  • Time requirement: The time the pixel requirement is met was greater than or equal to one continuous second, post ad render. This time requirement applies equally to news feed and non-news feed environments.

Video Time Requirement:
To qualify for counting as a mobile viewable video ad impression, it is required that 2 continuous seconds of the video advertisement is played, meeting the same pixel requirement necessary for a mobile viewable display ad.

Why this definition is a good thing:

Jason Cooper, general manager, mobile at Integral Ad Science.

“For years we’ve seen media owners avoid addressing mobile viewability or brand safety concerns, citing reasons such as: ‘the MRC has not made up their mind.’ But now the MRC released its mobile viewability guidelines. So this should now help to align all sides of the industry in how to move forward.”

But are the MRC guidelines too lenient?

Anyone familiar with desktop display advertising will recognize that the mobile guidelines are remarkably similar to the MRC’s desktop viewability guidelines. These are the same guidelines which critics led by Unilever’s Keith Wood argues are too lenient.

However, the MRC guidelines should considered the base standard, rather than the gold standard, according to Mark Finney, director of media and advertising, ISBA (Incorporated Society of British Advertisers):
It is a misconception that the MRC standard is a “target” to be adhered to, it is really more of a baseline. A baseline is needed because all advertisers have their own viewability targets, but you have to have a definition of a “view” for comparison purposes and to calibrate measurement tools. What you choose to pay for is a different matter. However, some people think that MRC is advocating 50% for one second – it is not.

Setting a lower threshold gives advertisers the flexibility to buy the type of inventory they require. So while some brand advertises call for 100% viewability to maximize eyeballs, this is less of a priority for performance advertisers who are interested in consumer interaction with their ads, but do not what to pay over the odds.

Mark Finney:

“Unilever has been very vocal that everything should be 100% viewable. Group M will not pay for anything less on behalf of its clients, and this would seem sensible, but there are downsides.”

Not all advertisers share the “100% “view. Advertisers who are more focused on direct marketing care less about this than CPA (cost per acquisition). E.g. if the price is right and something is 40% viewable but had a measurable effect on sales they may be happy with that. To get 100% viewable ads carries a hefty premium.

A good indication of the bewildering array of metrics / definitions currently used by agencies and advertisers to measure viewable impressions is to look at the data dictionary of a measurement company, such as Moat Analytics. ClickZ was shown a copy by InMobi. There are dozens of variants, depending on percentage of ad in view, time in view, position on the page and behavior of the user, all defined with different methods of calculation.

How are mobile ads verified as viewable?

Publishers, and increasingly agencies and advertisers, use independent verification companies.

The list of vendors accredited by the MRC (PDF) currently includes 14 vendors. Check with the MRC which ones of these are ‘Compliant – Self Asserted’ and which are ‘Compliant¬ – Audited’.

Only two of the 14 are currently accredited for mobile web and mobile app:

  • Integral Ad Science
  • Moat

What proportion of ads are viewable?

Unfortunately no company could supply any stats to show how many mobile ads are viewable. So we will have to make do with a chart for general display viewability from measurement company Meetrics. This suggests that more than half of display ads in the UK currently fail viewability tests.

dna38_visability_of_display_ads

You might expect publishers to refute such numbers. But they do not.

So what would it take to get publishers to improve viewability?

That’s an easy one: advertisers will have to pay a premium for guaranteed viewability.

Ben Price, group product manager, global media brands, at UK-based publishing house, Future:

I think there are two things to understand – 1) Increasing viewability will always increase CTR (click through rate) – that’s a fact – because if the ad isn’t seen, then it isn’t going to be clicked. However, 2) Advertisers need to be aware that publishers aren’t currently flocking to drive viewability up, because the yield doesn’t make sense right now.

For example, if advertisers were just to pay for viewable impressions, at a viewability rate of 50%, the publisher needs to see a rise of 100% in CPMs to achieve the same revenues as before. In reality, the publisher CPM increases are in the 20-30% range, which is why we haven’t seen widespread adoption from the publisher side.

This also poses the question of how important viewability and CTR really are to advertisers, if the publisher CPMs aren’t in direct correlation to viewability and CTR, what other quality metrics should be measured?

Does improving mobile viewability improve effectiveness?

Tests run with one campaign by Future found that doubling the viewability of mobile web ads, increased CTR by 145%.

Ben Price explains:

“We focused on mobile ads on one of Future sites – and subsequently rolled it out across the portfolio. We did two main things. 1) Added a sticky footer (ad remains on screen, when scrolling) 2) Adjusted the placement of MPUs (Mid Page Units) to higher viewability locations on the page.”

The CTR was pretty evenly increased across all campaigns and all ad sizes on mobile as a result. We compared to a previous 3 month average and saw similar sizes of increase in CTR as we rolled out on the other sites as well.

Off the back of this test, we’ve been building out a dynamic ad serving technology that automatically optimizes the placement of ads across the sites, depending on the content, user and engagement, optimizing for viewability and time-in-view. We plan on launching in Q2 2017.

This does not mean viewability is the same as effectiveness, points out Price:

Viewability is a minimum metric and only indicates a minimum level of quality, we need to make sure that we are working to optimize quality at an advertiser, agency and publisher level.

ISBA’s Mark Finney agrees. Confusing these two is a common mistake:

“Don’t confuse viewability with effectiveness – they are two very different concepts. Viewability measurement is based on a minimum threshold to enable the validation of an exposure to an ad impression, but it is not a KPI (key performance indicator) in itself. It should be a hygiene factor.”

Why mobile viewability is different and more difficult than desktop

The first thing to note is that the smaller form factor means that if publishers, under pressure from advertisers, use more intrusive tactics to increase viewability, it will have an even greater impact on user experience. For example interstitial (interruptive) ads will take up the full-screen and (if badly designed) can be difficult to close, and large sticky ads take up far more of the real estate on mobile than on desktop.

Mark Finney:

“Trying to optimize to viewability can lead to creative decisions that can negatively impact the user experience and therefore the brand (intrusive overlays, interstitials etc.). This may also drive more users to ad-blocking.”

Remember the context and the importance of creative – factors often overlooked in mobile where it is routine to re-use (or should that be “misuse”) creative designed for other platforms.

The second thing to note is mobile connection times may be slower, so the ad may not even loaded fully by the time visitor has moved on.

Verifying mobile web viewability is more difficult than desktop:

The issue is that the technologies that vendors use to verify ads on the desktop are not necessarily applicable on mobile web.

Integral’s Jason Cooper:

“The over-arching issue is that in mobile we have moved away from an environment where URL’s (referrers), cookies and Flash were ubiquitous across all major desktop browsers, to an environment where none of these elements can be relied upon. Using a Flash approach solved measurement of unfriendly iframes in the desktop environment, but Flash is redundant on mobile, so for mobile web viewability we have to seek out new solutions for iframes.”

The situation with in-app ads is even more complicated as they use a mixture of proprietary “native” technologies a long with web technologies.

One effort to cure this is the IAB’s on-going MRAID  (Mobile Rich Media Ad Interface Definitions) standard.

Jason Cooper:

“Broadly speaking there are two approaches for measuring in-app viewability: a native approach, which is the MRC accredited route, and MRAID, which offers more scale, but in its existing form, isn’t MRC accredited.”

Another major misconception is that all in-app ads are always in view because they have a fixed placeholder within the app. However, there are numerous use cases where this isn’t true, one case is with ads that are preloaded waiting to be shown at intervals dictated by the app developer such as between levels of a game. In this scenario the ad downloads and fully renders in the background, in a preloaded state, and of course all tracking pixels load along with it. However, if the user doesn’t get to that certain point in the game, then the preloaded ads are discarded, this is an important point to understand especially when an agency is trying to make sense of the data in their reporting.

Mobile ad viewability – and viewability in general is a contentious and evolving subject. It helps if all members of the ad ecosystem are well informed. Get involved with whatever initiatives are running in your country and attend any events.

This columnist recently attended a JICWEBS town hall that was well attended and informative (but the only advertiser speaker, from Unilever, wanted to keep his opinions secret).

This is Part 38 of the ClickZ ‘DNA of mobile-friendly web’ series.
Here are the recent ones:
How agencies and advertisers can spot and combat mobile ad fraud
Mobile advertising accounts for nearly half of digital spend, but it comes at a price: ad fraud
Where is Google heading with mobile local search?
Is Google killing mobile organic search

 

Andy Favell is ClickZ columnist on mobile. He is a London-based freelance mobile/digital consultant, journalist and web editor.
Contact him via LinkedIn or Twitter @Andy_Favell

Related reading