Brand risk in digital video and display advertising dropped significantly in the first half of 2021 compared with the same period in 2020 across 19 major markets, the latest Media Quality Report from Integral Ad Science shows.
Globally, brand risk fell by 40% to 50% across desktop display, desktop video, mobile web display and mobile web video between the two periods, based on trillions of data events analysed by IAS for the report.
In desktop display, rates of brand risk fell from 4.2% to 2.4%, with similar proportion declines in mobile web display (2.6%, down from 4.7%), desktop video (3.6%, from 6.3%) and mobile web video (3.6%, from 6.7%).
In video advertising, the UK saw the greatest improvement across both types of device, with risk on desktop video falling almost three quarters from 5.7% to 1.6%, and almost two-thirds on mobile web from 6.3% to 2.2%.
While the US also saw an improvement in both desktop and mobile web video, it continues to have the highest level of risk for advertisers outside the US, UK, Germany, France, Italy, Spain, Canada and Australia, with 5.5% on desktop (down from 8%) and 7.1% on mobile web (down from 9.9%).
When it comes to display advertising, however, the UK saw little change, with brand risk at 3% on desktop (compared with 3.1% last year) and 3.4% on mobile web (up from 3.3%). The US, meanwhile, enjoyed significant improvement across both, though brand risk remains slightly higher than in the UK: 3.5% on desktop (down from 5.9%) and 4.3% on mobile from (from 6.7%).
Video inventory achieved even lower rates of brand risk when purchased programmatically, IAS said – in mobile web video, programmatic risk across all markets fell almost two-thirds from 5.6% to 2%, while on desktop video it fell almost three-quarters, from 5.3% to 1.4%.
IAS defines the brand risk rate as the proportion of impressions on page that are “flagged for posing various levels of harm to brand image and/or reputation through association”, based on eight categories of risky content: adult, alcohol, gambling, hate speech, illegal downloads, illegal drugs, offensive language and violence.
It said that the breakdown in the types of content creating brand risk had changed significantly over the past year, with a large drop in the proportion of alcohol-related impressions flagged across all inventory types, while the proportion of disinformation content also fell.
However, adult, hate speech and violence contributed a greater share of brand risk than a year earlier. On desktop video, the risk posed by hate speech rocketed, from just 2% of all instances last year, to 15.4% this year.
Nick Morley, managing director EMEA at Integral Ad Science, said: “It’s interesting to compare brand safety risks to those of a year ago, when we were all navigating the unforeseen circumstances during the pandemic and a rise in misinformation. The distinct reduction in brand risk this year suggests that industry stability is resuming, with confidence increasing towards video ad inventory as the economy recovers.
“The improvement around misinformation reflects the industry’s commitment to supporting high-quality media outlets and publishers. More generally, we believe the move away from a blunt approach to brand suitability towards a nuanced contextual strategy means advertisers won’t inadvertently miss positive opportunities to interact with engaged audiences.”