Slashdot Mirror
Facebook Flat-Out 'Lies' About How Many People Can See Its Ads, Lawsuit Alleges (theregister.co.uk)
A new lawsuit claims that Facebook exaggerates how many people can see its ads, thereby defrauding advertisers. "In other words, it is alleged not quite as many eyeballs are seeing Facebook's ads as its salespeople charge for," writes Thomas Claburn via The Register. From the report: In a complaint filed on Wednesday in a US district court in Oakland, California, plaintiffs Danielle Singer and her company Project Therapy, LLC claim the Potential Reach and Estimated Daily Reach figures that Facebook provides to advertisers are wildly inflated. As an example, the complaint claims that Facebook's purported Potential Reach among 18-to-34-year-olds in each U.S. state is greater the actual population of 18-to-34-year-olds in each of those states.
"Based on a combination of publicly available research and Plaintiffs' own analysis, among 18-34 years-olds in Chicago, for example, Facebook asserted its Potential Reach was approximately 4 times (400 per cent) higher than the number of real 18-34 year-olds with Facebook accounts in Chicago," the complaint states. And in Kansas City, the complaint asserts, the number provided by Facebook was 200 per cent higher than the actual number of 18-to-54-year-olds with Facebook accounts in the area. What's more, the court filing contends that former Facebook employees, described as confidential witnesses, have acknowledged that Facebook is fine with inflated numbers. The attorneys representing Singer and her biz, which supposedly spent over $14,000 on Facebook ads, are seeking class-action certification in order to represent other affected Facebook advertisers. According to the complaint, "a former Facebook employee who worked in the infrastructure/mapping team stated that those who were responsible for ensuring the accuracy of the Potential Reach at Facebook were indifferent to the actual numbers and in fact 'did not give a sh--.'" They also said the "Potential Reach" statistic is "like a made-up PR number."
"Based on a combination of publicly available research and Plaintiffs' own analysis, among 18-34 years-olds in Chicago, for example, Facebook asserted its Potential Reach was approximately 4 times (400 per cent) higher than the number of real 18-34 year-olds with Facebook accounts in Chicago," the complaint states. And in Kansas City, the complaint asserts, the number provided by Facebook was 200 per cent higher than the actual number of 18-to-54-year-olds with Facebook accounts in the area. What's more, the court filing contends that former Facebook employees, described as confidential witnesses, have acknowledged that Facebook is fine with inflated numbers. The attorneys representing Singer and her biz, which supposedly spent over $14,000 on Facebook ads, are seeking class-action certification in order to represent other affected Facebook advertisers. According to the complaint, "a former Facebook employee who worked in the infrastructure/mapping team stated that those who were responsible for ensuring the accuracy of the Potential Reach at Facebook were indifferent to the actual numbers and in fact 'did not give a sh--.'" They also said the "Potential Reach" statistic is "like a made-up PR number."
This can't possibly be true! A major corporation would never ever lie about a completely unverifiable "fact" just to make money!
Sounds like a potential reach around!
Smart crooks at least make sure the numbers are somewhat plausible. They probably started small and found that nobody actually noticed. Then they just kept inflating the numbers, completely unaware that there is a hard upper bound. It really does not get much more stupid than this.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
No argument there.
Nominates story for Captain Obvious award.
Donald Trump, on a crusade to make Nixon look respectable
Back in the 80's, fake London newpapers would call advertisers and offer them advertising print runs in the "Islington & Highgate Gazette" of 50,000. Newspapers like this were fake, non-existent, they had plausible sounding names but little else. They'd ring advertisers seen in real media, get bookings, print off 50,000 copies, send one to each of the advertisers, and one for their lawyers, then pulp the rest. Then the same company would ring them about advertising in the "Chiswick and Hounslow Courier" with a print run of 100,000.... The Newspapers were a few stock articles and mostly adverts from companies suckered in.
The BBC, set up a anonymous facebook page with zero advertising and zero reason for people to like it, and they got loads of likes from across the world. What struck me is how Facebook must be behind that, because how else would the people in third world countries *know* about the new pages, let alone who would *pay* them to like these pages? I assumed it was to puff up FB's numbers prior to its IPO.
https://www.bbc.co.uk/news/29505104
Facebook claimed it was companies paying third parties to like the page to make it more popular.... but the BBC DID NOT PAY OR TELL ANYONE ABOUT ITS 'Virtual Bagel' empty FB page. So Facebooks explanation about these likes seems false.
So now you're telling me they lie to advertisers and the lies are whoppers, easily verified to be fake.... well yeh, but FB were never prosecuted for the alleged securities fraud last time, because there was no proof, and they won't be this time, because they'll pretend a 'reach' number is some vague metric without legal meaning.
Just like the newspapers I mentioned.... they never said "sales of 50000", they said "print run of 50000" .
ublock origin and FB purity for the win...
"times higher":
4x + x
It's not a phrasing that's used all that much, honestly. But 4x a number is not 4x higher than the original number. What if you reduced 1.01 times higher to 1 time higher. Like I said, it's an awkward phrasing that isn't used much. But it would be 1 time and 1 time again (2x).
"x% higher"
(x/100) * x + x
It's the exact same number.
Spreads beyond Hollywood. News at 11. Frank Zappa was right - again.
Why guess when you can know? Measure!
Democrat voters.
how can you trust the view/clicks number unless your site keeps the same metrics? They shouldn't charge for views as you have no idea if the person actually saw it, it could have scrolled off the screen without being noted.
"The hands that help are better far than lips that pray." - Robert Ingersoll (1833-1899)
If you counted all the 18-34 year olds that are currently in Chicago, you'll find a number higher than the census reports.
Tourists, workers who live out of town, etc, could easily swell up to 4x the amount.
...
I have bought ads on Facebook for my project. When you are creating an ad you specify your audience. This can be geographic and / or demographic. As you refine your audience Facebook will show you the potential reach. If the reach is too low (say 70-80 year old males in zipcode 90210) it will warn you that your audience is too narrow and your ad may not reach many people, conversely if it is too large it will also warn you. So it is just a ballpark tool to give an idea of how your filtering has narrowed down the potential audience.
Here's the thing - they don't bill by the potential audience. They bill by the actual impressions and responses served (supposedly, but that is an entirely different matter). So the point is that inaccuracies in the potential audience does not have a financial impact on the person buying ads either way. In fact, if anything, inflated numbers means you will spend *less* money on your ads, because your audience is smaller than you thought and thus FB can't display as many ads.
Better known as 318230.
I remember when the web was first starting up, people were wondering which model it would follow: newspapers, television, radio, or libraries.
I suggested the library model since in my view, there was no money to be made off of the net in the way that would support a whole industry.
It seemed at first that I was wrong, and then these studies came out:
These tell us that 8% of the users account for 85% of the ad clicks, and these users tend to be from households with yearly income under $40,000.
In other words, advertising on the internet does not reach the audience it wants, but instead is mostly taken up by the people who spend a lot of time on the internet because they have no other form of recreation.
This has been exacerbated by the bots which take up 28% of internet traffic, the use of ad blockers, and the tendency of experienced people whose time is valuable to avoid the internet since its audience now seems like daytime TV watchers after the mobile era began in 2007.
Since those studies have come out, we have seen the big companies trying to jockey "we have a lot of warm bodies" into "our advertising is valuable," when all credible data suggests the opposite.
In other words, assume the crash position.
Alternative Right.
5 times is not equal to 5 times higher.
It kills me that only now, at long last, someone actually spending money on the ads decided to do the math to see if they're getting what they paid for.
"Higher" means it is over and above the base.
Attendance IS "200% of last year's" is equivalent to "a 100% increase over last year" or, in this example, "attendance is 100% higher than last year's".
According to previous poster, saying "attendance is 100% higher this year" means this year's attendance equals last year's attendance - simply incorrect.
Ken