Keyword blacklists: why our ad dollars aren't reaching the news
Boomer-era tech is cutting journalism at its knees
|Feb 18, 2020||3|
The advertising industry is marking a record-breaking $300 billion in ad spend this year. And, there are a lot of news outlets cultivating strong, engaged readerships. This should be a match made in marketer heaven, but there’s a problem:
Our ad dollars aren’t reaching the news.
This is bizarre but true: an industry-wide “brand safety” practice is shutting newspapers out of hundreds of millions of dollars (or more) of advertising revenues. And the ad industry has been operating like this for years.
The practice is called keyword blacklists.
What are keyword blacklists?
For years, agencies and adtech vendors have used keyword blacklists to automatically prevent ads from running on potentially inappropriate content.
These blacklists use Regex (basically an Excel file and some 1950’s technology tacked together with some duct tape) to scan for “bad words” like these:
The problem with this method is it can’t tell the difference between a human interest piece on Slate (“Is Lesbian Sex Real?”) and a PornHub title. In fact, words like transgender, bisexual and transgender are ranked more “unsafe” than the word porn.
So that means websites that cover anything from local news to LGBTQ+ issues are being blanket blacklisted across the industry. And that our ads are avoiding the higher quality eyeballs that are found on news sites.
So, we’re basically blacklisting journalism?
Yep. Newspapers are in an impossible position because they need to cover the news… and they need to use certain words to cover it accurately. They cannot, for example, replace “death” with “crossing the rainbow bridge.” (Under adtech keyword blacklists, the latter would be the more profitable word choice.)
This issue is only getting worse because brands are terrified of being caught up in a social media shitstorm. So they’re “over-blocking” content. Here’s what’s happening on the ground:
In 2018, Vice found out advertisers were flagging LGBTQ descriptors “gay,” “bisexual” and “transgender” - blacklisting their LGBTQ coverage (these words were listed higher risk than “killing”)
In August 2019, WSJ reported that Fidelity Investments blacklists the words “immigration” and “racism.”
In January 2020, The Guardian reported that brands and agencies are blacklisting “any negative words” (???) and even the very important preposition “is’’ to avoid appearing on ISIS terrorist content.
Two things make this problem worse:
We are blacklisting words at a higher and higher rate. Brands are trigger happy on words that might get associated with politics, bad subject matter, or anything listed in the Brand Safety Framework [this is a PDF].
Blacklists don’t often get reviewed. Adops insiders tell us that brands blacklisted “Parkland Florida” after the school shooting. This ban is probably still in place. Imagine being a local newspaper in Parkland.
Why is this happening?
We are misunderstanding brand safety. An ad for Pampers next to a news story about a car crash in a newspaper is collectively understood as brand safe - the reader does not think to themselves “Pampers obviously supports car crashes” when they see it. That’s because the news is different from every other website on the internet.
Putting news sites in the same arena as every other website on the internet makes brands unable to distinguish the level of threat between disinformation and news.
What can we do?
Publishers have been pleading with marketers to please upgrade to a less stupid mechanism than Regex.
“In this current digital model, there is little attention to the quality of the content or the provenance of that content,”said Newsworks chief executive Tracey De Groose, “There is no distinction between crafted journalism – that adheres to ethical standards, editorial codes, regulators and the law – and bedroom bloggers, amateur producers and at the extreme end, criminal content.”
If you have an ad budget, you can make sure you’re appearing on high-quality journalism by whitelisting legitimate news sources.
There are some companies working to solve the problem of keyword blacklisting by introducing AI and machine learning. This sounds great. And, it’s never going to be enough: we as marketers must always be watching the robots. 🤖
P.S. Nandini was interviewed in Fake: Searching for Truth in the Age of Misinformation, which was released last week. Check it out!