Major brands are unwittingly funding disinformation domains.

The GDI estimates that a quarter billion dollars (US$235 million) is paid annually to our database of 20,000 disinformation sites by ad tech companies placing adverts for many well-known brands. These brands include household names that many of us know: Audi and Honda, American Airlines and OfficeMax, among others. The sites in our sample include www.addictinginfo.com, www.rt.com, www.twitchy.com and www.zerohedge.com.

This happens because much of online advertising is now placed “programmatically” (i.e. by computers). Every time we load a web page an instant auction happens to sell off the advertising space to the highest (computer) bidder. The GDI has explained how this system works in a white paper that we published in May.

Programmatic advertising is a fast and easy way to target potential customers but brands too often have no clear information about the sites where their adverts wind up. This means that brands discover only after the fact that their adverts have appeared next to very “brand unsafe” content. And currently there is little brands, or the ad tech companies that support them, can do to prevent this outcome.

This is a whole-of-industry problem that requires a whole-of-industry solution. And it must start with the ad tech companies.

For the universe of ad tech companies studied, the study estimates:

• Google serves 70% of the disinformation sites in the GDI study and provides 37% of the revenue (US$86 million annually).
• AppNexus (a Xandr Company), in second place, serves 8 percent of the sample disinformation sites and provides 25% of the revenue (US$59 million annually).
• Google serves the long-tail of low-traffic, low-revenue disinformation sites while AppNexus and others serve a group of high-traffic, high-revenue sites.
• Other major players funding disinformation sites include Amazon, Criteo, OpenX, The Rubicon Project, and The Trade Desk.

The findings are based on a sample of over 1,700 known disinformation domains. The list of disinformation domains was compiled from respected third parties working to combat disinformation, including Politifact and Le Monde. Our report outlines in more detail the methodology that we used.

Our findings are well-based estimates. Only the ad tech firms know which bids they win and what sites they serve. Making this information more transparent is one of the things we hope to change.

At the same time, brands and ad tech firms need real-time updates of disinformation domains as part of their brand safety tools.

Lastly, brands should have more control over ad placements based on a domain’s disinformation risk levels.

It is in the interest of everyone - but most importantly brands - to remove risky domains from their ad buys - and to defund disinformation. This is why the GDI is developing an index of real-time risk ratings to help ad tech and brands better direct ad spend.

The GDI wants to see disinformation criteria implemented by brands and ad tech firms as part of programmatic advertising.

We hope that leaders in the ad tech industry will join us in our efforts to defund and disrupt disinformation.

For more on the full study: Download report here