Last week, after being notified that our targeting capabilities allowed advertisers to reach people based on offensive self-reported entries for education and employment, we removed more than 3 million self-reported terms as targeting options in our ads interfaces. We did this after discovering a small percentage of people had entered offensive responses, in violation of our policies.
As an extra measure of precaution, we removed more than 3 million terms as targeting options that were derived from self-reported profile fields—school, field of study, employer, and job title. But we've also heard from advertisers how useful this targeting can be to reach the right people on our platforms.
What we're doing
After a thorough review, we're making additional changes to improve our policies, enforcement practices, and also adding back ~5,000 commonly used targeting terms that meet our Community Standards
We're updating our advertising policies
to make clear that content that goes against our Community Standards
cannot be used to target ads. This includes anything that directly attacks people based on their race, ethnicity, national origin, religious affiliation, sexual orientation, sex, gender or gender identity, or serious disabilities or diseases. This is not a new policy and this practice was already prohibited, but the policy is now more explicit.
Following a manual review, we'll be reinstating around 5,000 of the most commonly used terms for targeting use, after being deemed to be in compliance with our Community Standards
. The reinstated terms pertain to common use cases, such as 'student' for a university that hopes to enroll math students, or 'nurse' for a publication that wants to reach nursing professionals, or 'marketer' for a company that wants to hire marketers.
As of September 20, 2017:
These commonly used and approved terms will be available for ads creation and editing.
For any terms that are deemed to be in violation of our Community Standards
, we will remove these as targeting options.
We are pausing any ad campaigns that currently use the terms.
Affected advertisers will be notified via email, and advertisers will need to remove the disapproved targeting from their targeting criteria before they can get their campaign live again.
People receiving these, or any of our ads, can always use Ad Preferences
to opt out of specific advertising audiences, even if they've listed them on their profiles.
To prevent issues like this from happening again, we're adding more human review and oversight to our automated processes, which will review new ad targeting options prior to inclusion in our interfaces. Our goal is to improve our systems to help keep bad behaviors off of Facebook.