Facebook’s ad rules have changed
Are you prepared?
Originally published 2018
Take a close look at the following Facebook advertisement for a California law firm. The ad is promoting the services and expertise of a law firm that primarily helps older professionals create comprehensive wealth and estate plans. It appears to be a straightforward, informative advertisement. You can imagine it succeeded in attracting clicks from potential clients — and earning the firm new business.
And yet Facebook rejected it, and this version never appeared online.
What happened? How did such a seemingly innocuous ad get turned down? Welcome to the new reality of Facebook advertising. That reality began to manifest itself in the spring of 2018. That’s when the news broke that U.K. marketing firm Cambridge Analytica had used data gained from over 50 million Facebook users in an attempt to influence multiple U.S. elections.
The news quickly put Facebook in a hot political spotlight, culminating with high-profile congressional hearings on national television. Founder and CEO Mark Zuckerberg confronted questions about how Facebook could allow a politically motivated company to gain access to the sensitive information of millions of voters.
Central to critics’ concerns was that this wasn’t just a leak of information. It was a leak that Facebook’s rules specifically allowed to happen. Facebook was both the source of the information — and then the promotional platform used to exploit the private data of its users.
Facebook took the accusations very seriously and acted fast. After all, this scandal represented a monumental threat to its business model, which relies almost exclusively on advertising. Of the $40.65 billion in total revenue the social media giant achieved in 2017, $39.94 billion – or 98 percent of the media giant’s entire revenue – was generated exclusively through ads.¹ That advertising revenue has allowed Facebook to ride a wave of colossal growth, including a nearly 50 percent leap in revenue between 2016 and 2017 alone. In order to maintain its status as an advertising powerhouse, Facebook needed to work quickly to repair the damage.
And so, on May 25, 2018, Facebook changed its advertising policies in an effort to create structure and transparency to issues-based or “political” advertising. In so doing, Facebook is actually responding to two different issues at once. First, it is seeking to ensure better data privacy practices to prevent a future scandal like the one involving Cambridge Analytica. At the same time, it’s addressing the broader concern Congress and other critics have expressed over the use of the platform as a propaganda machine.
The Cambridge Analytica affair
The U.K.-based marketing firm’s manipulation of Facebook user data was the main reason that the social platform changed its advertising policies. A summary of Cambridge Analytica’s actions can easily be found online. The suddenly intense scrutiny that Facebook’s data practices received from this story caused it to tighten its algorithm regarding “political” advertisements.
Facebook advertising has changed a great deal over the last five years but never more than it has in just the past few months. As Facebook comes to terms with its own social impact and the way its service has been manipulated to sway public opinion, it has applied sweeping changes that have impacted every advertiser. And that includes law firms. True, it’s unlikely that a smaller practice would intentionally use Facebook marketing to make a political point. Still, as our customer discovered, nonpolitical businesses also are getting caught up in those changes.
Facebook has been one of the most powerful tools any business can wield in its digital promotion and marketing efforts. Even with the changes the social platform is making in advertising policies and processes, that’s still true. But those changes require law firms — including yours — to adapt in order to continue leveraging Facebook’s strengths.
The big change
One of the biggest advantages of using Facebook to advertise has been the platform’s ability to target highly specific sets of users. Facebook’s aggregated data allows advertisers to target their message based on a variety of demographic and behavioral patterns, including what users like, where they work, where they live, and what they are interested in. It’s little wonder that the platform has experienced such phenomenal revenue growth.
But in some ways, that massive repository of user data has also become a liability. Since Facebook’s tools help advertisers pick through and organize that information, any group can design an ad that could reach users with specific political leanings, income ranges, or past purchasing behaviors. (Or users sharing any combination of those categories.)
Those capabilities haven’t changed, but the way advertisers use them has changed significantly. Facebook now requires people and businesses seeking to post issues-based ads to verify their account and prove their identity. This verification includes requiring a copy of identification information such as a driver’s license, as well as requiring users to input a code Facebook sends by mail to a physical address. In addition, Facebook has created an archive where groups and individuals can review all ads that have been identified and approved as political. They also can see when the ads ran, how many people saw the ads, and how much was paid to run the ads.
According to Facebook, 98% of its entire revenue was generated exclusively through ads in 2017.
Transparency is nothing new for political advertising. Anyone who has seen an ad on TV for a given candidate or referendum will be familiar with language that indicates the organization that paid for the ad. But the way social media in general and Facebook in particular has been used to influence the electorate over the past few years has thrown numerous wrinkles into the situation. Even the Federal Election Commission (FEC) is struggling to navigate the new digital landscape.² Facebook’s application of the rules is broader and doesn’t apply only to promotions for specific candidates or upcoming elections. The social platform’s threshold for political advertising now includes anything that falls under a standard referred to as “issues advertising.”
The upshot: Facebook has changed its rules, and previously commonplace ads are now being rejected as political. Again, that has included many advertisements posted by law firms.
Since one of FindLaw’s services to its attorney clients is creating such ads, we quickly discovered that getting each one cleared through Facebook’s approval process was not feasible. Taking a proactive approach, FindLaw undertook a deeper analysis of content that might trigger a rejection. Much of what we uncovered was as surprising as it was useful.
What we found out about Facebook
In 2017, FindLaw, on behalf of thousands of clients, published around 25,000 ads on Facebook promoting legal services. With millions of engagements and interactions by users with these thousands of ads, we were able to compile a comprehensive wealth of data on the performance of Facebook legal advertising.
While FindLaw typically uses this data to further refine our strategies for social advertising, the information also has helped us understand the changes that Facebook itself makes in its advertising platform. Most businesses have been able to analyze the impact only in terms of their own specific campaigns. Because FindLaw’s model uniquely provides service to thousands of different marketers in one industry, we have been able to analyze the data more broadly.
More specifically, we can quantify to advertisers in the legal market both the challenges and the benefits of including Facebook promotions in their digital marketing strategy.
During the first two months, the new ad rules were in place, FindLaw documented more than 250 instances of Facebook rejecting ads due to alleged political or issues-based content. That number represents roughly 10 percent of all ads we created during that time period. Of those 250 total instances, however, 84 of them happened over a three-day stretch from June 18 to 20. Since June 20, daily rejections have dropped by half.
Why the decline? FindLaw’s theory was that, around June 20, Facebook was transitioning its ad approval process to a new, more predictable algorithm. This was largely suggested by the consistency of practices that have since been flagged by Facebook’s system. During the first few weeks after Facebook instituted the changes in May, we found it much more difficult to anticipate what would and would not trip the rejection action.
With Facebook’s system more predictable, the rejection patterns became clearer, and FindLaw was then able to conduct a more useful analysis. We soon found that the most common areas of law receiving Facebook rejections for ads containing political content were Criminal Defense, Employment Law, and Family Law. In addition, we discovered a high number of rejected ads in what we classified as “Other.” The majority of rejected ads in that category were associated with Immigration and Housing Law.
However, those numbers represent a pure volume, and do not take into account the difference in the total volume of ads published per area of practice. So, while there were more ads rejected in the area of Criminal Defense compared with, say, Tax Law, we also needed to consider that many more ads are created under the heading of Criminal Defense than Tax Law.
Looking at the total number of ads rejected only as a percentage of that area of practice, a different pattern emerges:
This chart reveals that the most likely areas of practice for political ad rejections are Tax Law, Employment Law, Criminal Defense, and Other (Immigration and Housing).
What we found surprising in this result is the diversity in the areas of practice that seem to be affected by Facebook’s changes. It wasn’t a surprise to find Immigration Law ads rejected as political. But the fact that Employment, Criminal Defense, and even Business Law represented such likely rejections implies that the content Facebook deems “political” has become very broad.
Law firms simply seeking to attract new business might think Facebook is erring far too much on the side of caution. Perhaps it is. But these changes suggest this caution is directly related to the critical attention the digital platform’s data practices have received. Facebook hopes to plug any holes that might leave it vulnerable to manipulation – holes that could allow future ad revenue to drain out.
How Facebook (apparently) decides what to reject
As FindLaw dug into the specific rejections and analyzed the individual components of the ads, we developed a deeper understanding of what seemed to drive a Facebook rejection. Some factors seemed sufficient to initiate a rejection on their own. But most of the advertisements seemed to be rejected because they contained multiple small signals that pushed them over Facebook’s rejection threshold. For example, simply using the word “tax” in an ad wouldn’t immediately or necessarily lead to a rejection but doing so while trying to target an audience within a specific net-worth range would far more likely cross the threshold.
In other words, Facebook appeared to look at all elements of the ad to determine whether it crossed the “political” line. Those elements included the topic of the ad, the image it used, and the presence of significant political content in the destination page where visitors were directed. However, the two things that appeared most likely to trigger a rejection were (1) certain keywords in the ad copy, and (2) red flags in the targeting strategy. Let’s look at those two factors more closely.
One of the primary methods that Facebook likely uses to identify political advertising is pinpointing specific keywords in the ad copy. FindLaw conducted a deeper analysis of roughly 70 ads to more closely identify keywords that might trigger rejection. While the ads spanned a wide range of topics and approached them in different ways, patterns arose.
First, a few general keywords relatively common in legal use seemed more likely to trigger a rejection. Those keywords include the following terms and references, among others:
- Marijuana or drug terminology / slang
- Specific gov’t institutions (e.g., the Supreme Court)
- Tax / IRS
- Immigrants and immigration
- Specific law (e.g., the Civil Rights Act)
- Legal / legalization
Looking at this list, it’s easy to see the challenges a law firm could face when creating an “acceptable” Facebook advertisement. Again, even the word “law” itself seems to be part of the calculus Facebook applies to its approval decisions. Add to that obstacle the mention of specific legislation, particularly those most commonly recognized by the public at large, such as the Americans With Disabilities Act or Family and Medical Leave Act, and it’s almost surprising the rejection rate isn’t higher.
Despite the noted constrictions, Facebook’s algorithm appears to seek some degree of nuance in determining whether a specific keyword in isolation and without associated charged language is enough to reject. This means that the presence of most of these keywords alone isn’t enough to trigger a rejection. But when used in conjunction with other red flags, they increase the likelihood.
Also, it’s necessary to look at keywords in terms of both volume and frequency. That the terms “tax” or “IRS” triggered 10 of those 70 instances of rejections is highly significant. As we noted before, within the overall ad population those terms occur far less frequently than the word “law” – which, of course, constantly appears in legal advertising. That high rejection rate suggests the word “tax,” even appearing alone, is more likely to trigger a rejection than the word “law” by itself.
Just as important as keywords is the way an advertisement uses Facebook’s targeting tools to identify its audience. Highly specialized targeting has long been among Facebook’s strengths, but it has also proven to be its Achilles’ heel. This precision-targeting capability was the open door, allowing agenda advertisers to use the social media platform to influence and inflame political opinion within specific demographics. Having come under fire for the practice, Facebook has understandably elected to restrict and more closely monitor the use of certain kinds of targeting.
Under Facebook’s new monitoring regime, significant changes have and will continue to come to the types of targeting options Facebook provides. In August 2018, the social platform discontinued access to its “managed custom audiences” and “partner categories.” In essence, these were data sources that Facebook licensed from Experian, Acxiom, and other third-party data brokers. This data provided a wide range of information such as purchasing habits, income, and net-worth data.
As FindLaw examined the targeting behaviors that seemed most likely to flag a rejection, we found that the more an advertiser attempted to isolate specific communities within larger populations, the more likely the advertiser would trigger a political rejection.
Most notably, income and net-worth targeting options (targeting an ad based on how much money a person makes or has) seemed especially likely to trigger a rejection. While targeting didn’t seem to be as crucial a factor as keywords in attracting the algorithm’s scrutiny, when it became a factor it was a more significant one.
FindLaw’s research found that targeting factored into only about 40 percent of the ads. Using the same sample of 70 ads, we found those most likely to elicit a “political” rejection included targeting based on:
- Income and net worth
- Association with government agencies and employees
- Specific political affiliations
- Specific employers and industries
- Association with specific cultural, racial, political or industry groups
These categories line up with Facebook’s overall goal of better controlling and documenting issues promotion where entities routinely target specific groups of people and provide them a political message. An organization wishing to influence labor union politics might, for example, target specific industries. A group expounding a fiscally conservative agenda might target specific incomes and net worth. Facebook knows that political advertisers will try to isolate segments of the population — and that these advertisers will potentially game the system Facebook has put in place.
Facebook’s new limits on target options comes into conflict with legal advertising in some pretty obvious ways. As we mentioned earlier, one of the areas of practice most likely to get caught up in political ad rejections is Employment Law. The biggest reason? Their ads reasonably target very specific industries and local employers. Quite often, the industrial workers most likely to get hurt on the job, or to bring an employment complaint against their employer, are exactly the same people a political advertiser targets to promote the “jobs” agenda of a given candidate or party.
All is not lost
By now, you might be wondering whether your firm should simply forgo Facebook as a marketing medium, but that would be shooting yourself in the foot.
For one thing, these changes are seeking to create transparency in Facebook advertising, which may help elevate the confidence users place in the ads they end up seeing. The new Facebook rules require a more thoughtful approach to targeting — and that approach could help your firm’s ad avoid rejection and potentially improve the way you select your audience.
In cases where your firm seeks to target specific segments, you can apply many different targeting values to reach the desired audience. For example, for an ad promoting immigration services, a law firm wouldn’t want to simply target people with an interest in immigration.
Such a broad approach would be more likely to attract an audience that sees immigration as a political issue. To reach immigrants themselves, the firm would probably instead wish to target specific kinds of visas, and specific cultural populations, as well as people who work in industries more likely to employ immigrants.
It appears that Facebook’s new algorithm casts a wide net, catching anything that looks like it might be politically motivated. That said, we’ve also discovered that Facebook has a relatively robust appeal process. If an ad gets caught in that net, Facebook will respond to a direct appeal that an advertisement is intended to legitimately promote a service, not to influence public policy.
With all this in mind, let’s look at specific examples of some rejected legal services ads — and how they were reworked to pass muster or, in one case, were successfully appealed.
Taking a case-by-case look
We’ll start by revisiting the Facebook advertisement that opens this paper. It links to a post that seeks to inform potential clients about changes coming to the federal Pension Retirement Act. In other words, it’s providing useful information – an excellent way for a firm with this specialty to build trust among potential clients. It certainly doesn’t seek to sway political opinion.
But as FindLaw research discovered, calling out particular pieces of legislation signals to Facebook that the promotion might be political in nature. This ad refers to a specific law. What’s more, the phrase “2006 bill” in the copy text sends another signal that becomes more significant with its relationship to the specific legislation language. It’s reasonable to assume Facebook’s algorithm tied these keywords together and inferred this might be a political ad.
In addition, the advertisement targeted a specific group and income demographic — in this case, retirees interested in managing their funds. That’s another red flag for Facebook, one that ultimately led to this ad being rejected as political.
The solution primarily required adjusting the copy to clarify its objective and remove some of the likely keywords. The targeting then became less problematic. Facebook approved the revised ad seen here:
What to take away. Facebook evaluates both keywords and targeting when determining whether to approve an advertisement. While there was no single obvious red flag here, a combination of factors pushed the ad over the rejection threshold.
Again, some areas of practice present more challenges than others when navigating through the new rules, and immigration law is at or near the top of that list. Knowing what we know now about the ways Facebook applies its standards for political advertising, it shouldn’t surprise you that the following ad was initially rejected:
This advertisement is legitimately promoting a business that has a substantive interest in advocating for the rights of its customers. If you’re an immigration lawyer, you want to educate potential clients on their rights and demonstrate your own expertise.
First, let’s talk about what likely got the original ad flagged. In this case, it’s almost entirely the ad copy. It contains many “political” keywords, including “immigration,” “civil rights” and “ICE Agents.” Other probable red flags include a strong advocacy tone and an unclear objective — meaning it’s not immediately obvious that this advertisement promotes a legal service.
This is the edited ad that Facebook ultimately approved:
What to take away. The biggest and most obvious change is the removal of some hot- button immigration language. This actually improved the advertisement, because the linked material is now far more about how to interact with law enforcement than it is about immigration at large. Along with a new headline that clarifies the objective of the linked material to provide information for users, it’s clearer why this ad exists and who it is for.
When originally published, the ad below was rejected by Facebook as being political advertising.
The reasons for the initial rejection probably include the use of keywords such as “federal employer” and “federal agencies.”
The advertisement also targeted specific agencies and employees, including the U.S. Postal Service and the American Federation of Government Employees.
The solution here wasn’t an edit, though. After all, this ad belongs to an employment law firm that represents government employees. That’s the target audience. The goal of the ad isn’t to change political opinions, but to educate a legitimate segment of likely consumers.
So, FindLaw sent a friendly email to Facebook appealing the decision:
“Hi Facebook! This particular ad was rejected for political reasons. However, it merely shares information regarding how some federal employees may be discriminated against in the workplace, which is related to a service that our customer (a law firm) provides to its clients. If you still feel that this is political upon further review, please let us know why that’s the case. Thanks!”
The appeal succeeded. Facebook reversed the rejection and the ad ran unchanged.
What to take away. Facebook’s appeals process can be worth pursuing. While it appears that the majority, if not all, initial rejections result from of a decision made by a computer algorithm, the appeals process seems to get a human involved. And a person is better equipped to understand and respond to the nuance of these ads. FindLaw has reached nearly a 90 percent success rate with its appeals. That rate comes from FindLaw’s understanding of Facebook’s rationale in classifying political advertising, and our ability to make the case that a given ad shouldn’t fall under that classification.
A sharper tool (so handle with care)
Even with all the changes Facebook has put into place, it remains one of the best tools a law firm can use to promote its services to the types of people it wants to reach. In fact, you could make the case that these changes have made it stronger. By developing methods to make political posts more transparent, Facebook is shoring up its credibility as a marketing platform. And that could help keep users — and advertisers — from leaving.
But in order to make use of this sharpened tool, law firms advertising on Facebook must tailor their messages carefully. And that requires knowing the words and targeting methods most likely to trigger an ad’s rejection.
Ethics rules relating to social media advertising
Social media sites are designed to connect people, and most sites allow users to communicate with each other or the public at large through one or more means. The rules prohibiting solicitations require lawyers to consider the intended recipient’s identity as a potential client. Hyper-targeting your advertising may misidentify it as solicitation. Unfortunately, most states haven’t had to investigate these issues yet, so there is little precedent establishing where the line between conversation and solicitation is.
Lawyers wishing to advertise on social media are most likely aware that these ads are subject to their jurisdiction’s rules on attorney advertising. The normal prohibitions still apply to making false or misleading statements, presenting oneself as an expert, revealing attorney-client privileged information, and the use of testimonials. Some jurisdictions also consider Facebook posts and even an attorney’s Facebook page as “lawyer” advertising subject to these rules. If you are in one of these jurisdictions, make sure to review the entirety of your profile and timeline before inviting additional visitors to your page.
Further, having a social media advertisement reach potential prospects invites questions and comments on the ad itself. Lawyers must understand the potential to form an attorney-client relationship when responding to any questions or comments on the site. It is best to take those communications to a familiar medium, like a phone call or email, where you can properly manage the communication and advise the client of the conditions surrounding your discussion.
- “Facebook Reports Fourth Quarter and Full Year 2017 Results” https://investor.fb.com/investor-news/press-release-details/2018/facebook-reports-fourth-quarter-and- full-year-2017-results/default.aspx
- “FEC struggles to craft new rules for political ads in the digital space,” Washington Post, June 28, 2018, https://www.washingtonpost.com/politics/fec-struggles-to-craft-new-rules-for-political-ads-in-the-digital-space/2018/06/28/c749a234- 7af9-11e8-aeee-4d04c8ac6158_story.html
Make social media work for you.
Discover what FindLaw’s social media solutions can do for your firm.