The study – by ad tech firm InSkin Media, in association with research firms Research Now and Sticky – involved nearly 4,300 UK consumers, and technologies including eye-tracking and viewability measurement.
Eye-tracking revealed that 25% of ads defined as viewable – i.e. meeting minimum industry guidelines of 50% of the pixels being on screen for least one second – were never looked at.
One third achieved a "gaze time" (time spent actually looking at the ad) of less than a second, while only 42% were looked at for at least a second.
The median time a viewable ad was actually gazed at was 0.7 seconds.
The study also revealed how long an ad needs to be viewable in the first place to hit certain levels of gaze time.
On average, to be looked at for up to a second an ad needs to be viewable for 14 seconds, it found.
Ads achieving at least one second of gaze time had to be viewable for an average of 26 seconds. For at least two seconds gaze time, the average viewability was 33 seconds, while for 3+ seconds gaze time, average viewability was 37 seconds.
The study further found that the time people spent looking at ads differed significantly across the four key formats covered. Most time was spent on page takeovers at 7.5 seconds, followed by billboard (3.3 seconds) and half pages (0.9 seconds) and finally MPUs (0.7 seconds).
Steve Doyle, InSkin Media's CCO, explained that a campaign should be assessed in three stages: did the ad have the opportunity to be seen, was it actually looked at and what was the impact.
Smaller formats have higher "opportunity to be seen" rates and many campaigns were optimised against these, he said. "But gaze time is very low – thus, it's optimising on low engagement and low impact."
Publishers also need to address the issue of clutter, which, the research found, can reduce ad gaze time by 37% on average across formats.
Data sourced from InSkin; additional content by Warc staff