Why ad viewability matters and how to improve it
Ad viewability is a key factor that advertisers consider when deciding how much to bid for impressions. This in turn, makes viewability a crucial lever for publishers to optimize their ad revenue. In this guide, we’ll explain what ad viewability means, how it is measured, why it matters, and how publishers can work towards optimizing their viewability score.
What is ad viewability?
Ad viewability answers the following question: How many served ads are actually seen by users? The most widely used definition for viewability comes from the Interactive Advertising Bureau (IAB) and the Media Ratings Council (MRC), wherein an impression is considered viewable if:
- For display ads, 50% of the ad is visible for at least 1 second
- For video ads, 50% of the ad is visible for at least 2 seconds
According to data released by comScore in 2013, only 46% of all ads served were found to be viewable. Back then, almost everyone knew that some ads are never seen by users, but the numbers released by this study demonstrated the scale of the problem.
There were strong calls from marketing executives like Marc Pritchard of P&G and Keith Weed of Unilever, and media agencies like Group M, for creating more stringent viewability standards. In response, Google (among others) started offering tools for publishers to measure their viewability and for advertisers to target high-viewability inventory.
Ad verification vendors like Moat, DoubleVerify, IAS, comScore, Pixelate, Forensiq, and Fraudlogix also gained popularity, offering a way to filter the ad tech supply chain for viewability.
As a result of these industry developments, the average ad viewability has improved considerably in the last few years. According to a report published by IAS last year, global desktop display viewability improved from 69.2% in 2019 to 71.9% in 2020, while mobile display viewability increased 11.9% in the same time period to reach 67.6%.
So, what makes ads non-viewable?
If only 70% of all the ads served are seen by users, what happens to the other 30%? The most commonly cited example of a low-viewability placement are ads that are placed below-the-fold (BTF). Publishers who create long-form content will obviously want to monetize the bottom half of their web pages. However, since a lot of users drop off before ever getting there, BTF ads, on average, score less on viewability compared to above-the-fold (ATF) placements.
Based on the logic above, the top of the page should be the most viewable area, right? Data suggests otherwise. According to Google, the most viewable placement is right above the fold, not the top of the page. The reason being that while ads need to remain in-view for at least 1 second to count as viewable, most users have already scrolled past the top placement, a problem that is further exacerbated on web pages that take a long time to load.
This tells us something important about viewability: it is not always intuitive. A placement that you think is highly-viewable might score low in the real world setting, and vice versa. This is why it’s important to measure the viewability of your inventory and make optimization decisions about ad sizes and placements based on actual performance data instead of assumptions.
Active View vs. ad viewability
Publishers who use Google AdSense or AdX to monetize their website might see the “Active View” metric in their revenue reports. While all vendors rely on IAB’s standard definition of viewability, the way they calculate the score for ads served using their platform differs. Active View is Google’s method, but in most contexts, the terms can be used interchangeably.
Why bother optimizing for viewability?
Advertisers have a strong interest in bidding more on viewable inventory because it improves their return on ad spend (ROAS). Most demand-side platforms now provide advertisers the tools to forecast, set campaign pacing, and generate reports specifically targeting high-viewability inventory. In 2015, Google made its display network GDN 100% viewable, and stopped charging advertisers for impressions that were not viewable.
The relationship publishers have with viewability is a little more complex. During the initial push for viewability, many were worried that the move from being paid for impressions to being paid only for viewable impressions would mean revenue loss. Savvy publishers took this shift as an opportunity to position their inventory as “premium” and demand higher bids.
However publishers may feel about it, one thing is certain: the industry is not going to roll back on viewability. With time, high-viewability inventory is only going to become more valuable to advertisers, with vCPM (viewable CPM) replacing CPM as the currency of digital advertising. Publishers should optimize for viewability, so they can benefit from this trend and command better bids for their inventory. Since page design, speed, and content are the main levers used to optimize viewability, working to improve it often results in better overall UX too.
How to optimize viewability score
Focus on ATF, but avoid top page placements
On average, ATF placements outperform BTF placements in viewability. It’s important to note that the “fold” itself varies for users, as they access your website using different devices and screen resolutions.
Measuring scroll depth may be a better indicator for defining where users spend more time, instead of choosing an arbitrary point on the page.
Placements that are created on-the-fold and just above-the-fold tend to be the most viewable. Most sites have top-of-the-page placements, but these ads tend to underperform in viewability, so consider avoiding them.
Choose ad sizes that outperform on viewability
Ad size also affects viewability. The most viewable ad sizes are vertical because they tend to stay on the screen longer as users move around the page.
Based on data reported by eMarketer, the most viewable ad sizes for mobile are 120×240, 240×400, 234×60, and 320×50. On desktop, the most viewable ad sizes are 970×250, 300×600, 728×90, and 300×250.
One size, 160×600, was found to be a high performer on both device types. That said, most research on ad sizes and viewability is old, so the best thing to do is generate your own measurement data.
Measure and improve Core Web Vitals
Core Web Vitals are a collection of metrics (Largest Contentful Paint, First Input Delay, and Cumulative Layout Shift) that Google thinks have the biggest impact on UX.
Improving Core Web Vitals will also increase your viewability score because it fixes page speed and minimizes “layout shift”, a phenomenon where the user’s position on the page abruptly shifts during page load.
Want another reason to improve Core Web Vitals? Starting May, Google will update its search algorithm and start driving more traffic to websites that have higher-than-average Core Web Vital scores. Read more about it in our guide about optimizing Core Web Vitals.
Work on reducing bounce rate
Bounce rate measures the percentage of users who exit a website without engaging in any meaningful interaction. High bounce rate can signify underlying problems with your website design, page speed, or content quality, and negatively affects almost all key business metrics.
If bounce rate hovers anywhere above 70%, it means that a majority of your users are existing the site before they ever have a chance to see an ad, often even before the ads load. It can therefore be a leading contributor to low viewability.
A bounce rate of anywhere between 25% to 40% is considered good, 41% to 55% is average, and 56% to 70% implies room for improvement.
Optimize your header bidding stack
According to Kevel’s header bidding industry index (HBIX), 70% of all US publishers were using header bidding by Q4 2020. While header bidding is a great tool, it does tend to be a big contributor to page latency, especially client-side header bidding.
While server-side header bidding, for instance Google’s Open Bidding, does provide an answer to the page latency problem, giving up client-side completely might not be the best solution for all publishers.
Still, there are things that you can do to optimize your client-side header bidding stack, like choosing a fast wrapper, limiting the number of bidders, setting optimal auction timeouts, using asynchronous loading, and removing bidders that take too long to respond.
Start lazy loading your ad units
Lazy loading is a technique used to defer loading page elements before they enter the user’s viewport. Applied to content, lazy loading can conserve bandwidth and increase page load speed by prioritizing the loading of elements that are in-view.
However, lazy loading can also be applied to ad units and many publishers are already doing so. Apart from the usual benefits, lazy loading also improves viewability by restricting ads that are outside the fold from loading.
Use sticky ads and adhesion units (in moderation)
Sticky ads, as the name suggests, remain fixed on a specific part of the web page regardless of where the user scrolls. Due to this, they remain in-view for longer than most other ads and score high on viewability and achieve better clickthrough rates (CTR).
Even though Google Ad Manager supports sticky ads, using sticky ads that are too large are considered obstructive by industry standards like Acceptable Ads Standard and Better Ads Standard. Using too many sticky ads can also hurt UX and is therefore not a good idea.
Bottom-line is that sticky ads can help improve CPM and viewability score, but they should only be used in moderation.
Only refresh ads that pass viewability criteria
In the last few years, ad refresh has gained popularity as a way to increase Session RPM by displaying more than one set of ads to a user within a single browsing session. IAB recommends that refreshed inventory should be declared as such for buyers and healthy time intervals should be used between subsequent refreshes.
However, when ad refresh is configured without taking viewability into account, it can have a disastrous effect on the overall viewability score, as ads continue to be refreshed outside the user’s viewport. The fix for this is to make sure that ads are refreshed only after they have already met the industry standard viewability criteria.