Last week, we explored the concept of keyword blacklisting and how bad it is for news sites. But you might still be thinking: “Better safe than sorry! I’m glad I’m one of 94% of brands with keyword blacklists. It’s my job to keep my brand safe, not worry about funding the news industry.”
Well, we have some bad news for you. Actually, it’s bullshit news.
“Fake news” is a term that has come to mean different things to different people,” says a University of Michigan research guide. We agree.
For this reason, we prefer the more scientific term: bullshit news. ????
This helps us center the idea that we’re not talking about “conservative” or “liberal” media - or any other media bias we may not agree with - but bullshit purposely crafted to:
This distinction matters because the goal of real news or “journalism” is to communicate what happened, based on facts and reporting. Journalists adhere to journalistic ethics, like transparency and accountability.
Some argue that journalism has gotten sloppier and more sensationalist over the past decade. We would agree that yes, quality has definitely suffered in the perverse world created by adtech and social media: a world in which newsrooms must compete with propagandists and bullshit artists for ad dollars.
Adtech products can’t (or refuse to?) tell the difference, and so news and bullshit merge together simply as “monetizable content,” which brands (unsuccessfully) try to filter it out with keyword blacklists.
What the news industry calls “news” is held together by a high standard of expected behavior and ethics. What Google calls “news” is basically just content with a date & time stamp. Let’s look under the hood for a moment.
In the news industry, there are real, peer-reviewed standards that come with rewards and consequences. Here is The Society of Professional Journalists’ Code of Ethics (click the link for the extended version). These include:
Now, here for example, is what Google’s algorithm requires from its news sources:
On Google, you can also crawl your way up to greater visibility in a few different ways. Accuracy is not one of them. Here are some of Google’s ranking factors, determined algorithmically:
It’s super problematic for brands that the adtech industry cannot differentiate journalism from “bullshit.” If they can’t tell the difference, they cannot protect you from it. And indeed, they have not.
Having articles with dates and time stamps is how Breitbart, a once-fringe white supremacist media outlet, was accepted as a verified news source across 20 ad networks including Facebook and Google.
Even as Editor-in-Chief Steve Bannon shouted from the rooftops that they were a propaganda operation aligned with white nationalism, Breitbart was given a place alongside The New York Times and The Washington Post. This gave readers the illusion they were reading the news.
Breitbart was slated to make $8 million in annual ad revenue before the Sleeping Giants campaign made it one of the most toxic sites on the internet. (Full disclosure: One of your BRANDED co-authors continues to lead that effort.)
The point here is that you can be a literal bullshit burrito that calls itself “news” and still have an amazingly lucrative business on your hands. That’s because...
Most adtech companies will work with bullshit publishers regardless of their content. Many claim to use a combination of tech and humans to weed out bad sites. Our understanding is that it’s much more tech than humans.
Fake news sites know which keywords to avoid. Dr. Augustine Fou of Marketing Science says fake sites with "algorithm-generated pages are optimized with keywords that attract free organic traffic” and avoid keywords that get the pages banned by brand-safety tech."
All kinds of brands can be found funding disinformation sites. You can check out examples of these sites for yourself if you want to see what they look like (we’re not going to link to them):
Buzzfeed reporter Craig Silverman investigated Canadian fake news sites and reported on November 6, 2019. Some of those sites may have been shut down, but many more remain unknown so the problem persists.
“You've got to a whole bunch of advertisers who aren't really paying very close attention. It's a license to pretend, and to print money” says April Lindgren, a professor at Toronto’s Ryerson School of Journalism.
Here are some steps we recommend taking right away:
Thanks for reading!
Nandini and Claire
P.S. Thank you to those who responded to the keyword blacklists issue with examples of how blacklists (aka blocklists) are hurting brands:
Notre Dame as a keyword. “When the cathedral of Notre Dame burned to the ground last year, many advertisers put “Notre Dame” on their black lists. Their intention was their ads not to be brought up with any news regarding the fire, however, their ads didn’t show up around any news regarding the 2019 Notre Dame NFL Draft Picks either... missed opportunity!” (from Stefan Bergmeier)
Pot as a keyword. “Last year, we got called out for providing a non-brand safe environment to an advertiser - it turned out that in their blacklist, there was the term “smoking pot”. However, in German language “pot” is the basic term for a kitchen tool, where you put all your ingredients for cooking in. As my broadcaster is airing some cooking shows (and is putting all related info on the internet), we basically ended up being non-brand safe for showing people how to cook ????
Content moderation works better. "One thing I’m seeing as a future tech is the growth of “content moderation” instead of keyword tech to solve for brand safety. You can see the challenge is so complex it really still needs humans to assist. But [private equity]-backed firms have a vested interest to keep the keyword industrial complex going; it’s just so expensive to design and spin up a new tech stack." (emphasis ours)