Advertisement

Here's Why Twitter's Porn Problem Freaked Out Brands

10 million accounts could be sharing NSFW content

Twitter is trying to manage the problem of NSFW content. Illustration: Matthew Billington

There are some dark corners of Twitter—accounts that take "not safe for work" content to new depths and that can spell big trouble for brands whose ads end up running next to them. It's a serious concern that advertisers and Wall Street insiders raised with Adweek last week, and one that Twitter has been quietly trying to manage.

The problem is largely a technological one: how to police hundreds of millions of accounts, 10 million of which could be devoted to pornography, according to one Wall Streeter's estimates.

In late April, Robert Peck of SunTrust Robinson Humphrey reported on Twitter's dirty parts. He warned that once brands found out ads could run alongside these posts, the advertisers would pull back on spending.

"Since it is unlikely that Twitter would knowingly allow this, it must be a function of Twitter's technology," Peck wrote in a note to investors. "Twitter's technology should hopefully be able to identify these users, but it may take some time. One can perform basic searches to see the issue."

Last week, advertisers Nielsen, Oikos, Gatorade, Duane Reade, NBCUniversal and Best Buy were all alerted that their Promoted Tweets appeared when a consumer searched pornographic pages. Twitter said it was developing a fix. Nielsen said it would temporarily suspend its campaign. The other advertisers declined to comment.

Twitter is not alone, it is important to note. Facebook has also come under fire for offensive content, while Tumblr has been known to host X-rated material and Reddit users have a reputation for posting NSFW material—just to name a few.

Twitter's porn problem comes at sensitive time. CEO Dick Costolo's effectiveness is in question as the company recently missed ad sales forecasts.

The CEO of one mobile advertising network found it shocking that Twitter still faces this kind of problem. "Our brands freak out at any hint their ads show up in situations like this," the CEO said.

Facebook, which has developed a reputation for its hard-line stance against pornography, is years ahead of Twitter in detecting this type of content, according to industry watchers.

Costolo has admitted that one of his biggest failures was not acting aggressively enough to combat harassment and threats from trollish users.

When it comes to adult content, the Internet often employs third-party technology to ensure ads aren't seen in an unfortunate context. Kiril Tsemekhman, chief data officer at one such company, Integral Ad Science, said platforms like Twitter can only police themselves, as it does not give outside firms the access they'd need to help. "It's difficult for anyone outside the platform to manage it," he explained. "You'd need to be able to quantify each tweet into a certain category that advertisers would want to stay away from."

Ad tech experts propose that for better control, Twitter would need to sift through all its content, identify the good from the bad, flag offensive photos and videos—and do all that in every language.

A signal that more advanced technology is imminent came by way of Twitter's recent deal with Google, which can now help target ads on the site. Ad industry sources said that the search giant might be able to help in areas where Twitter's technology falls short.

"You'll never solve 100 percent of the problem," said John Murphy, vp of marketplace quality at ad exchange OpenX. "But there are things you can do to minimize exposure."

Advertisement
Advertisement
Adweek Blog Network