Google continues to work hard to keep bad ads off the internet. On Wednesday it provided a look at how the company fought bad advertising practices in 2017, sharing details on technology and policies added to protect consumers, advertisers and publishers.
Providing some perspective on the enormity of the challenge, Google removed 100 bad ads per second in 2017, Scott Spencer, Google's director of sustainable ads, wrote in a blog post. “We took down more than 3.2 billion ads that violated our advertising policies," he wrote -- up from 1.7 billion in 2016.
Google's ad network operates at a massive scale. The company blocked 79 million ads in its network for attempting to send people to malware-infested sites, and removed 400,000 of these types of sites in the past year. The company also removed 66 million of what it calls “trick-to-click” ads, as well as 48 million ads that attempted to get users to install unwanted software.
Meanwhile, Google worked to develop better technology to protect advertisers, removing 320,000 publishers from its ad network for violating the company's publisher policies. It also blacklisted nearly 90,000 websites and 700,000 mobile apps.
In 2017, Google removed 2 million pages for policy violations each month. After expanding its policy against dangerous and derogatory content in April to cover additional forms of discrimination and intolerance, Google removed ads from 8,700 pages that violated the expanded policy.
The ad network is open to publishers that do not abuse the platform. Google paid AdSense publishing partners $12.6 billion last year. But to make money from Google ads, publishers must follow the rules that prohibit websites in the ad network from serving ads with content that is misrepresentative.
For example, a publisher is prohibited from serving ads if it pretends to be a legitimate news website based in London when it’s actually a content scammer in California. “We found that a small number of publishers were responsible for the majority of these violations,” Spencer wrote. “Of the 11,000 websites we reviewed for potentially violating the misrepresentative content policy, we blocked over 650 of those sites and terminated 90 publishers from our network.”
Google continues to write and enforce new policies to protect consumers and publishers. Too often, Google identifies violations of its scraping content policy. This type of policy violation occurs when bad actors try to make money as quickly as possible by copying news or content from other sites, Spencer wrote. So last year, Google estimates that it blocked more than “12,000 websites for “’scraping,’ duplicating and copying content from other sites, up from 10,000 in 2016.”
Aside from the fact that the numbers of "bad ads" deleted is up these stats tell us very little. The most important question would be---in my mind, at least---what percentage of all ads served are "bad ads"?