Meta Makes $16 Billion a Year From Scam Ads and Budgets $1 Billion for Fines. Do the Math.
When getting caught is just a line item
By Cat Casey | The Shield Blog
Part 3 of 12 in the "Who Sold My Data?" Series
I want you to do some arithmetic with me.
Meta makes approximately $16 billion per year from what internal documents call "higher-risk" advertisers. That's scam ads, fraud schemes, crypto garbage, fake celebrity endorsements, and the kind of targeted manipulation that ruins people's lives.
Meta budgets approximately $1 billion per year for regulatory fines and legal settlements.
$16 billion in. $1 billion out.
That's not a compliance problem. That's a 1,500% return on investment.
Welcome to the scam economy. Meta built it. Your grandmother is funding it.
The Reuters Investigation
In October 2024, Reuters published an investigation that should have ended careers. They obtained internal Meta documents, interviewed current and former employees, and documented what many suspected but couldn't prove:
Meta knows it's delivering scam ads. It does it anyway. It's wildly profitable.
The numbers from the investigation:
- $16 billion annually from "higher-risk" advertisers
- $3.5 billion every six months from ads Meta's own systems flag as potentially fraudulent
- 10% of Meta's total revenue comes from this category
- 15 billion scam ads delivered daily across Facebook, Instagram, and WhatsApp
- Internal employees estimate Meta platforms are involved in one-third of all successful scams in the United States
One-third. Of all scams. In America.
Meta isn't just failing to stop fraud. Meta is the infrastructure that makes fraud scale.
How the Algorithm Targets Your Grandmother
Here's what makes this different from a newspaper that accidentally runs a bad ad.
Meta's algorithm doesn't just deliver ads. It optimizes them. It learns who clicks. It learns who converts. It learns who falls for things.
When a scammer buys ads on Meta, the algorithm studies the results and finds more people like the ones who got scammed. It identifies the patterns—age, isolation, recent life changes, cognitive decline, financial anxiety—and serves more scam ads to more people who match that profile.
The system isn't neutral. The system is a scam-targeting machine.
Reuters found internal discussions where employees raised concerns about this exact dynamic. The response? The ads were too profitable to restrict.
When your platform makes $16 billion a year from a category of advertiser, you don't crack down on that category. You optimize for it.
The Fine Math
Let's talk about what happens when Meta gets caught.
In 2019, Meta paid a $5 billion FTC fine for privacy violations related to Cambridge Analytica. At the time, it was the largest privacy fine in history.
Meta's stock price went up after the settlement was announced. Investors were relieved it wasn't worse.
The company makes about $135 billion in annual revenue. A $5 billion fine is 3.7% of one year's revenue. It's a speeding ticket.
Meta has learned to budget for this. They set aside money for fines the way you set aside money for car insurance. It's a known cost of doing business, not a deterrent.
The equation is simple:
Revenue from bad behavior > Cost of getting caught = Keep doing it
Until that math changes, nothing changes.
The Victims
Let's make this concrete.
Elder fraud in the United States:
- $4.9 billion lost in 2023—up 43% from the year before
- Average loss per victim over 80: $9,000+
- 72% of elder fraud is enabled by data broker targeting profiles
Those data broker profiles have names. Real names that real companies use:
- "Financially Challenged"
- "Probably Diabetic"
- "Gullible Elderly"
- "Rural and Barely Making It"
These aren't jokes. These are actual marketing segments, designed to identify people at their most vulnerable and sell access to them.
Meta's algorithm ingests this data. A scammer buys ads. The algorithm finds the "Gullible Elderly." The scam ad lands in their Facebook feed, sandwiched between photos of grandchildren and posts from church groups.
It looks legitimate because it's in a legitimate place. It works because the targeting is precise. And when grandma loses her retirement savings to a crypto scam endorsed by a fake Elon Musk, Meta books the ad revenue and moves on.
The Celebrity Deepfake Industrial Complex
You've seen them. Fake videos of celebrities endorsing investment schemes. AI-generated Tom Hanks selling dental plans. Deepfake Oprah promoting weight loss pills.
Meta is the primary distribution channel for this content.
The platforms have policies against deepfakes and unauthorized celebrity endorsements. The policies exist on paper. In practice, the ads run for days or weeks, reach millions of people, and generate massive revenue before anyone takes them down.
By then, the scammer has moved on to the next campaign. The money is collected. The victims are scammed. The celebrity files a complaint. Meta issues a statement about "continuing to improve our systems."
Rinse. Repeat. Profit.
Martin Lewis, a British financial journalist, has been fighting Meta for years over fake ads using his likeness. He's documented hundreds of scam ads with his face attached to crypto schemes and investment frauds. Meta keeps running them. He keeps reporting them. They keep coming back.
At some point, you have to conclude that the system is working as designed.
The "Higher-Risk" Euphemism
I want to pause on that phrase: "higher-risk advertisers."
This is corporate-speak for "advertisers who are probably scamming people." Meta knows who they are. Meta has systems that flag them. Meta tracks the complaints, the chargebacks, the reports of fraud.
And Meta keeps taking their money.
$3.5 billion every six months. From advertisers Meta's own systems flag as problematic.
The company isn't ignorant. The company has made a business decision. The decision is that $16 billion a year is worth the reputational damage, the regulatory fines, the lawsuits, and the human wreckage.
That's not a failure of content moderation. That's a revenue strategy.
What Meta Says
When confronted with this reporting, Meta issues statements about their commitment to user safety. They cite the number of scam ads they've removed. They point to their AI systems that detect fraud. They announce new initiatives and partnerships.
None of it addresses the core problem: Meta makes more money from scam ads than it loses from enforcing against them.
Until that equation flips—until the cost of running scam ads exceeds the revenue from running scam ads—Meta has no financial incentive to solve this problem.
They have every incentive to appear to be solving it while continuing to profit from it.
And that's exactly what they're doing.
The Scale of the Machine
Let me put 15 billion daily scam ads in perspective.
That's:
- 625 million scam ads per hour
- 10 million scam ads per minute
- 174,000 scam ads per second
Every second of every day, Meta delivers 174,000 ads that its own systems have flagged as potentially fraudulent.
The company that built the most sophisticated ad targeting system in human history—the company that can identify your political leanings, your relationship status, your health conditions, your financial anxieties, and your likelihood to make impulse purchases—somehow can't figure out how to stop running ads for obvious scams.
They can find you. They can't stop finding people to scam.
The technology exists. The will doesn't. Because the will would cost $16 billion a year.
The Regulatory Failure
You might wonder where the government is in all this.
The FTC has enforcement authority. They've used it—occasionally, slowly, usually after the damage is done.
But the penalties aren't calibrated to the problem. A billion-dollar fine sounds massive until you realize it's less than one month's scam ad revenue. The company can pay it out of petty cash and keep going.
What would actually change behavior?
- Percentage-of-revenue fines that scale with the violation (the EU is starting to do this)
- Personal liability for executives who approve fraudulent ad categories
- Algorithmic auditing requirements that force transparency in targeting
- Private right of action letting victims sue for damages
- Criminal referrals for knowing facilitation of fraud
None of these exist in meaningful form in the United States. So Meta keeps doing the math. And the math keeps working out in their favor.
The Advertising Ecosystem
Meta isn't alone in this. They're just the biggest.
Google runs scam ads too. So does TikTok. So does X (Twitter). The entire digital advertising ecosystem is built on a model that prioritizes revenue over verification.
Advertisers pay. Platforms deliver. If the advertiser is a scammer, the platform still gets paid. The incentives are aligned against user protection.
But Meta deserves special attention because of scale. Facebook and Instagram reach 3 billion people. WhatsApp adds another 2 billion. No other platform has the reach, the targeting sophistication, or the documented evidence of knowingly profiting from fraud.
When you're responsible for one-third of all successful scams in America, you don't get to hide behind "industry-wide challenges."
You ARE the industry-wide challenge.
What You Can Do
Protect yourself:
Assume every ad is lying. Especially investment opportunities, health products, and celebrity endorsements. If it came through a feed, treat it as suspicious until proven otherwise.
Never click financial ads on social media. If a company is legitimate, you can find them through a direct search. If they're only reachable through a Facebook ad, that's a red flag.
Verify celebrity endorsements independently. Before believing that Tom Hanks wants you to invest in crypto, check his official channels. Spoiler: he doesn't.
Report scam ads. Yes, it feels futile. Do it anyway. It creates a paper trail. It adds to regulatory pressure. It documents the problem.
Talk to the elders in your life. The people most vulnerable to these scams often don't know they're being targeted. Have the conversation. Show them examples. Make it specific.
Protect everyone:
Support regulatory reform. The FTC needs more funding, more authority, and penalty structures that actually deter. That requires congressional action.
Demand platform accountability. Meta responds to advertiser pressure and public relations crises. Make this a PR crisis. Make advertisers uncomfortable sharing a platform with scammers.
Use protective tools. Apps like ZoraSafe can help identify which platforms are exposing you to the most risk and provide real-time warnings about suspicious content.
The Bottom Line
Meta has built the most effective advertising platform in history. It can target you based on your psychology, your vulnerabilities, your recent searches, your life changes, and your likelihood to believe what you're told.
The same system that helps legitimate businesses find customers helps scammers find victims. And Meta knows this. They've known it for years. They've chosen to keep the money.
$16 billion a year.
$1 billion budgeted for fines.
Do the math.
The math says this won't stop until we make it stop. Until the fines are larger than the profits. Until executives face personal consequences. Until the public pressure becomes unbearable.
Until then, Meta will keep optimizing. The algorithm will keep learning. The scam ads will keep flowing.
174,000 per second.
Your grandmother's retirement savings, one click at a time.
Next week: "They Tracked 9,000 Phones at a Synagogue. Then They Sold the Data." — How your most protected activities became products.
About the Author: Cat Casey is the CEO of ZoraSafe and has spent 20+ years in cybersecurity. She wrote this article angry and stayed angry through every revision.
The Shield Blog publishes weekly investigations into the companies profiting from your data and the practical steps you can take to fight back. Subscribe to get each article in your inbox.
Sources: Reuters investigation "Inside Meta's Broken Promises on Stopping Scam Ads" (October 2024); FTC enforcement data; FBI Internet Crime Complaint Center 2023 report; data broker industry documentation; Meta investor filings and earnings reports.