In 2019, WFA invited key verification suppliers to join forces and participate in an exercise to share market-by-market data on levels of impressions. Twenty months since the launch, we are now able to assess how these trends are moving over time.
Share this post
It’s extraordinarily difficult to know exactly how many impressions are technically viewable, how many are fraudulent or how many are brand-safe. No single company sees the entire market.
But the market needs more objectivity. It needs more benchmarks.
Last year WFA invited key verification suppliers to join forces and participate in an exercise to share market-by-market data on levels of impressions which meet the MRCs’ standards for viewability, Sophisticated Invalid Traffic (SIVT) and brand safety. The MRC accreditation is critically important. For these companies to meet the same, high standard, we know there is a degree of consistency to the digital ad quality numbers they report.
Adloox, DoubleVerify, Integral Ad Science, Meetrics and MOAT are all contributing data to fuel the benchmarks. WFA’s strategic partner on Accountability, Ebiquity daughter company Digital Decisions, does all data processing, quality control and interface maintenance.
Twenty months since the launch, we are now able to assess how these trends are moving over time. The below are just some of the observations we’ve made between 2019 and 2020:
Overall, ads are becoming more viewable year-on-year
- Video leads the pack here with an average global viewability rate of 69% in 2020 (up from 67.7% in 2019). And some markets, for example, Russia, Spain and Turkey, see rates well above this (in the high 70%s).
- But the biggest YoY increases are coming from display and in-app formats which have both seen around a 4% point increase globally (i.e. 55% in 2019 to 59% in 2020). At 67%, the display viewability rate in Brazil is 8% points above the global average.
- But it’s not all good news. Video viewability in the UK was at 69% in 2019 (2 points ahead of the global average) and has now dropped down to 65%, 4 points below the average.
Ad fraud is increasing considerably within China and driving the global average
- Interestingly, across all formats and markets, fraud is more or less flat YoY.
- But this conceals the fact that it has increased considerably in certain markets, China in particular.
- China is (by far) the market most exposed to ad fraud within the DMB, with up to 15% of impressions found to be either General or Sophisticated Invalid traffic. The level of exposure has almost doubled YoY.
Levels of ‘unsafe’ impressions not growing significantly overall but many markets changing places
- As much as 18% of impressions are considered ‘unsafe’ in the worst affected markets, but the headline global average of unsafe impressions has only increased marginally YoY.
- The main finding is that markets are changing places in the ranking. Markets such as UAE, Turkey, Israel and the UK are now worst affected, replacing a contingent which included Russia, Malaysia and Switzerland at the top.
There is some logic behind these figures, of course. Publishers worldwide have been optimising towards viewability for some time so a slow and steady improvement is anticipated. Third-party ad serving and verification is not widespread in China so it’s expected that levels of fraud would be higher than average in this market.
Now we have the benchmarks to prove our assumptions and we will continue to update them on a quarterly basis to track the trends.
But while the data provides utility, we want to urge clients to maintain close stewardship over their digital media. After all these benchmarks are not a panacea.
The verification technology deployed by vendors participating in the DMB meets MRC’s high standards but the sophistication of criminals executing ad fraud is ever-increasing and all vendors remain susceptible to emerging ad fraud techniques.
Additionally, although the Digital Media Benchmark dataset is considerable, by definition, it only includes impressions from advertisers that have deployed ad verification technology. As a consequence this is a view of the quality of the most vigilant advertisers. Not the whole market.
Finally (and perhaps most important of all) due to the arrangements verification vendors have in place with Large Digital Enterprises (including Facebook), data from some key platforms is not included in the DMB.
In summary, this data should not necessarily be considered the whole picture and WFA asks advertisers to consider the limitations of the data and to use with caution.
But with these considerations in mind, the use cases for these benchmarks are various:
- Advertisers that set global guidelines or performance standards can now refer to a quarterly up-to-date resource that provides granularity without constant revisions or re-negotiations
- Understanding performance versus the market by easily identifying outliers and underperforming regions or channels is a valuable way of assigning priorities for improvement for marketers overseeing regional or global investments
- Advertisers have started referring to DMB rates in their agency contracts and performance related fee agreements, which incentivises agencies to continuously improve against a realistic target, delivered by an impartial consortium of the leading industry vendors in ad verification