yego.me
💡 Stop wasting time. Read Youtube instead of watch. Download Chrome Extension

Facebook Fraud


6m read
·Nov 10, 2024

Have you heard of Virtual Bagel? Their Facebook page has over 4,000 likes. They use the page to promote their brilliant business model: "we send you bagels via the Internet -- just download and enjoy." It sounds like a joke, and it is, sort of. This page was set up by BBC technology correspondent Rory Cellan-Jones in 2012. He wanted to find out what is the worth of a like on a Facebook page, so he bought some likes for Virtual Bagel.

Now there are two ways to buy likes: the legitimate way and the illegitimate way. The illegitimate way is to go to a website like BoostLikes.com and purchase some likes. You can get 1000 for $70. Sites like these use clickfarms in developing countries like India, the Philippines, Nepal, Sri Lanka, Egypt, Indonesia, and Bangladesh. Here, employees are routinely paid just 1 dollar per thousand clicks of the like button. So, Facebook explicitly forbids buying likes this way.

Instead, they offer the legitimate way to pay for likes by advertising your page. Prominently displayed is a link to "Get more likes" with the promise: "Connect with more of the people who matter to you." And this is how Virtual Bagel got its 4000 likes. Rory Cellan-Jones paid 100 dollars to Facebook, and the likes rolled in. He targeted his ad to the UK and the United States, but also to countries like Egypt, Indonesia, and the Philippines.

Now, where do you think Virtual Bagel was most popular? I'll give you a hint: it wasn't the US or the UK. But within a day, he had over 1600 likes, mostly from developing countries. Now, what was more problematic was that the people who followed Virtual Bagel looked suspicious. For example, there was one Cairo-based follower whose name was Ahmed Ronaldo. His profile consisted almost exclusively of pictures of Cristiano Ronaldo, and he liked 3,000 pages.

Cellan-Jones also observed that his new throng of fans was particularly disengaged, just as you'd imagine those from a click-farm would be. But he hadn't hired a click-farm; he had paid for Facebook ads. This story was reported in July 2012. In August, Facebook reported it had identified and deleted 83 million fake accounts (that was 9% of the total at the time). This resulted in noticeable drops for popular singers and celebrities.

So did they delete all of the fake likes? Nope, not even close. I know because most of the likes on my Facebook page are not genuine. In May 2012, I received a number of emails from Facebook offering me $50 worth of free promotion of my page, which at the time had only 2,000 likes. My YouTube channel had twenty times that following, so I thought surely this free 'paid' promotion could help me reach more of the people who mattered to me.

And immediately, I could see results. Within just a few days, my likes had tripled, and they kept on growing, thousands per day. And after a few months, I had about 70,000 Facebook likes, which matched my YouTube subscribers at the time. Now, what was weird was my posts on Facebook didn't seem to be getting any more engagement than when I had 2,000. If anything, they were getting less engagement.

I didn't understand why at the time, but I have since realized it's because most of those likes I was gaining through Facebook ads were not from people who were genuinely interested in Veritasium. How do I know? Well, because fake likes behave very differently from real followers. Have a look at this graph of the engagement of my Facebook followers. Here I'm plotting countries as bubbles, so this is Canada, and the size represents the number of likes I've received from that country. So this is the United States; it's a nice big bubble.

Now I'm ranking these countries on the horizontal axis based on what percentage of those likes have engaged with my page this month. So, as you can see, roughly 30% of Canadians and Americans have engaged with my page, but they're not as active as the Germans, where over 40% of my likes have engaged, and they are not as active as the Austrians, a small but passionate group of Veritasium fans at nearly 60%. These are all of the other Western countries, so you can see that it's common for between 25% and 35% of my page likes to engage with my page every month.

Now here is Egypt, where less than 1% of my likes have engaged with my page. Now this is India, the Philippines, Pakistan, Bangladesh, Indonesia, Nepal, and Sri Lanka. That's big followings, but no engagement. Together, all of these countries make up 80,000 likes; that's roughly 75% of the total likes I had before the last video. And these are the profiles that followed me when I used Facebook advertising.

And they are worse than useless. Here's why: When you make a post, Facebook distributes it to a small fraction of the people who like your page just to gauge their reaction. If they engage with it by liking, commenting, and sharing, then Facebook distributes the post to more of your likes and even their friends. Now, if you somehow accumulate fake likes, Facebook's initial distribution goes out to fewer real fans, and therefore, it receives less engagement, and so consequently, you reach a smaller number of people.

That's how a rising number of fans can result in a drop in engagement. And from this, Facebook makes money twice over -- once to help you acquire new fans, and then again when you try to reach them. I mean, your organic reach may be so restricted by the lack of engagement that your only option is to pay to promote the post.

What's worse, there is no way to delete fake likes in bulk -- all you can do is target posts around them. And I should re-iterate I never bought fake likes. I used Facebook's legitimate advertising, but the results are as if I had paid for fake likes from a clickfarm. Now, you might think the solution to all this is just to exclude countries with click-farms from your ad campaigns. But unfortunately, the problem goes much deeper.

Meet Virtual Cat, a virtual pet like none other. Its page is committed to supplying only the worst, most annoying drivel you can imagine. Only an idiot would like this page. And that's not just my opinion; that's actually what it says in the page description. And I should know because I wrote it. I created this page yesterday, and I then paid $10 to advertise the page through Facebook, targeted only to cat-lovers in the United States, Canada, Australia, and the UK.

Now, I expected that because I had excluded all of the big click-farm countries and because the page is so terrible, I basically wouldn't get any likes. But within 20 minutes, I had blown through my whole budget, and I got 39 likes. So who are these people liking a blank page and costing me 25 cents a piece? All of the profiles were all from the places I had targeted, mostly the US, but there was something strange about them.

All of these people liked a LOT of things, like hundreds and thousands of things. And a lot of the things they liked were odd too. Like in one account, this person liked T-mobile, AT&T, and Verizon. They liked Jeep and Lexus and Mercedes and Volvo and Volkswagen. They liked everything. Other accounts I saw liked kitchen scrubbers, and they liked mouthwash. Who reports that on their Facebook page? It just baffles me.

So the real mystery to me is why someone, somewhere would click on ads they didn't care about without making money from them. I mean, I don't think these likes came from bots -- they are too easy to identify and eliminate. And I also don't think for a second Facebook would pay click-farms to click on those ads to generate revenue for them, so it really seems like a mystery.

And then, in this article, I found what I think is the most reasonable hypothesis. Click-farms click the ads for free. In order to avoid detection by Facebook's fraud algorithms, they like pages other than the ones they've been paid for to seem more genuine. I mean, you can imagine 1000 likes on a particular page coming from one geographic area in a short period of time would seem suspicious. But buried in a torrent of other 'like' activity? They would be impossible to identify.

So workers at these click-farms will literally click anything. I mean, where do you think Facebook's Security page is most popular? Dhaka, Bangladesh. What about Google? Dhaka. What about soccer star David Beckham? It's actually Cairo, but you take my point. So wherever you're targeting, advertising your page on Facebook is a waste of money.

I wish Facebook would remove the fake likes from my page and all the others. But that would mean admitting that they have generated significant ad revenue from clicks that weren't genuine, which then suppressed the reach of pages that had low engagement, forcing those pages to pay again to reach inauthentic fans. So the truth is Facebook benefits by maintaining this status quo, because the reality is nobody likes this many things.

More Articles

View All
Running Water - Thaw Project | Life Below Zero
My nephew Jesse is coming in on the flight. I’m bringing the Young Gun in. “Hey Jesse, how’s it going? How are you doing? Thank you for helping me out.” Mhm, I am Jesse Moore, and I am from Fairbanks, Alaska. I’m here because, uh, my auntie happened to …
Half the universe was missing... until now
This episode was sponsored by KiwiCo. More about them at the end of the show. Until recently, half the universe was missing or hidden or just… undetected. And no, I’m not talking about dark matter or dark energy, which make up 27 and 68 percent of our un…
Checking bus fares with if statements | Intro to CS - Python | Khan Academy
Let’s design a program using Boolean expressions and if statements. The public transit system wants to build an app that determines a passenger’s bus fare. The standard bus fare is $4.25; however, they offer discounts for certain age groups. Kids under fi…
Comparing P-values to different significance levels | AP Statistics | Khan Academy
What we’re going to do in this video is talk about significance levels, which are denoted by the Greek letter alpha. We’re going to talk about two things: the different conclusions you might make based on the different significance levels that you might …
Second derivatives (implicit equations): find expression | AP Calculus AB | Khan Academy
Let’s say that we’re given the equation that (y^2 - x^2 = 4), and our goal is to find the second derivative of (y) with respect to (x). We want to find an expression for it in terms of (x) and (y). So pause this video and see if you can work through this.…
Constructing exponential models: percent change | Mathematics II | High School Math | Khan Academy
Cheppy is an ecologist who studies the change in the narwhal population of the Arctic Ocean over time. She observed that the population loses 5.6% of its size every 2.8 months. The population of narwhals can be modeled by a function n, which depends on th…