When it comes to privacy promises, what businesses say about the personal information they collect from you has to line up with their day-to-day procedures. That’s the message of the FTC’s proposed settlement announced today with Facebook. Where did the company go wrong? The agency’s 8-count complaint boils down to this: Facebook’s privacy practices often flew in the face of its stated policies and, as one count alleges, the company made significant retroactive changes to its privacy practices, without getting users’ consent.

Facebook’s privacy settings. First, the FTC charged that Facebook promised that users could restrict their information to a limited audience, using certain privacy settings. But the truth, says the FTC, is that even when users went to Facebook’s Central Privacy Page, clicked a link to “Control who can see your profile and personal information,” and limited access to certain people – say, to “Only Friends” – their choice was ineffective when it came to third-party apps that their friends used. What kind of information did the apps have access to? Things like a user’s birthday, hometown, interests, status updates, marital status, schools attended, where they worked, photos, and videos.

Privacy changes – material omission. Additional counts related to privacy changes Facebook made in December 2009. According to the FTC, Facebook claimed the changes gave users “more control” over their information and allowed them to preserve their “old settings” to protect the privacy of their profile information. But what really happened? Certain information that users had designated as private - like their Friend List - was made public under the new policy. The FTC charged that when Facebook implemented the December 2009 changes, the company overrode users’ existing privacy settings without adequately disclosing what it was up to.

Privacy changes – unfair practices.  Furthermore, according to the FTC, by designating certain user profile info as public when it had previously been subject to more restrictive privacy settings, Facebook overrode users’ existing privacy choices. In doing that, the company materially changed the privacy of users’ information and retroactively applied these changes to information that it previously collected. The FTC said that doing that without users’ informed consent was an unfair practice, in violation of the FTC Act.

What info apps had access to. According to the complaint, for a significant period of time after Facebook started featuring apps on its site, it deceived users about how much of their information was shared with the apps they used. Facebook said that when people authorized an app, the app would only have information about the users “that it requires to work.”  Not accurate, says the FTC.  According to the complaint, apps could access pretty much all of the user’s information - even info unrelated to the operation of the app. For example, an app with a TV quiz could access a user’s Relationship Status, as well as the URL for every photo and video the user had uploaded - information that went well beyond what the app “requires to work.”

What info Facebook shared with advertisers. Facebook also told users it wouldn’t share their personal information with advertisers. In Facebook’s Statement of Rights and Responsibilities, the company said, “We don’t share your information with advertisers unless you tell us to ([e.g.,] to get a sample, hear more, or enter a contest). Any assertion to the contrary is false.  Period . . . we never provide the advertiser any names or other information about the people who are shown, or even who click on, the ads.”  In fact, says the FTC, from at least September 2008 until May 2010, Facebook ran its site so that in many instances, the User ID of a person who clicked on an ad was shared with the advertiser. So much for “never.”

Facebook’s “Verified Apps” program.  The FTC also challenged the operation of Facebook’s Verified Apps program.  Facebook told people that the program involved a “detailed review process” and was “designed to offer extra assurances to help users identify applications they can trust - applications that are secure, respectful and transparent, and have demonstrated commitment to compliance with Platform policies.” About 250 apps paid between $175 and $375 for the seal. But according to the FTC, Facebook took no steps to verify either the security of a Verified App’s website or the security the app provided for the information it collected, beyond the steps it took for any other app.

Photo and video deletion. In addition, the FTC charged Facebook with making deceptive claims about its photo and video deletion policy. Each of the photos and videos a user uploads onto Facebook has a Content URL - a URL for its location on Facebook’s servers.  Facebook told users, “If you want to stop using your account you may deactivate it or delete it.  When you deactivate an account, no user will be able to see it, but it will not be deleted . . . When you delete an account, it is permanently deleted from Facebook.” But even after users followed Facebook’s procedure for deactivating or deleting an account, Facebook still served up these photos and videos to anyone who accessed them via the Content URL. That, said the FTC, rendered Facebook’s statements deceptive.

US-EU Safe Harbor program.  Finally, the FTC challenged what it said were deceptive statements Facebook made about its compliance with the US-EU Safe Harbor Framework, a mechanism by which U.S. companies may transfer data from the European Union to the United States consistent with European law.