Facebook Reveals 8.7pct, or 83.09 Million Accounts Are Fake

[engine]

Facebook Reveals 8.7pct, or 83.09 Million Accounts Are Fake
[news.cnet.com]

Last week, during its first quarterly earnings report as a public company, Facebook revealed it had 955 million monthly active users and 543 million monthly active mobile users. In the social-networking giant's 10-Q filing published last night, the company disclosed that nearly 20 percent of the latter number, 102 million users, accessed the social network in June solely from their mobile device.

So, how many of these accounts are fake? Facebook estimates 8.7 percent, or 83.09 million accounts.

That's a huge jump, both in raw numbers and as a percentage, from Facebook's last estimate. Back in March, Facebook said 5 to 6 percent of accounts are false or duplicate. At the time, this meant between 42.25 million and 50.70 million users.

[Panthro]

lolol

[bhartzer]

As a comparison, do we know how many Google+ and Twitter accounts are fake?

Most social networks have fake accounts--and I wouldn't be surprised if 8 percent of Twitter accounts are fake, as well.

[Leosghost]

I would be highly surprised, if the real figure for "fake accounts" is below 10% on any of the "social media"..especially on those who have either already sold shares or who are yet to make their IPO..

It is quite possibly on the higher side of 15% or even near 20%..

Facebook's own figures ( released as a damage limitation PR exercise ) for "fake accounts", I believe about as much as I believed their own ( and their IPO "handlers" ) "valuation" figures pre IPO..

[Strapworks]

I agree with Leosghost, there is no way the actual number is under 10%. I think the 8.7% is there effort of admitting to fake accounts without sounding out of control. Remove fake accounts, duplicate accounts, and accounts only used once and you only have 75%ish left (if that).
I had a friend the other day still think Facebook was a good buy because it went up $.06 while he was on his lunch break, I laughed and told him to go and waste his money, 4 hours later it dropped over $3, LOL
I give it 3 years before someone else comes along and becomes the new fad.

[gethan]

...and these are the accounts that "like" your page when you pay for likes.

Last week, digital distribution firm Limited Press alleged that, based on its own analytics software, 80% of clicks on its advertisements within Facebook had come from fake users.

[bbc.co.uk...]

[chicagohh]

I think 8.7% is low. It's laughably easy to fake social signals. People want to be liked and most people find it hard to offend others... this means that for the cost of a yahoo email and a few hours anyone can build a FB profile with 100 friends. If you do this off and on over a few months it's not hard to have 50 hand built FB accounts with people you don't know responding to your updates or wishing you a happy birthday.

Twitter is even easier.

Mind you - this is all done by hand. There are programs that do this in bulk.

8% - LOL!

[ponyboy96]

Ah yes, the new link spam which will be fake accounts and likes. Easier to game than links.

[g1smd]

Look at some of the forums that have been on the web for a very long time, and you'll see in some cases that well over 50% of all user names ever registered have made only 0 or 1 post.

I think ZenCart or some other open source software was such an example.

[ChanandlerBong]

I want to know how many fake accounts there are at WebmasterWorld. Anyone spot any bots posting?

[Digmen1]

And what about duplicate accounts at ebay !

And what about 13 year old son Facebook ?

[buckworks]

Yes, we do sometimes spot bots posting on WebmasterWorld. They are shot on sight.

Unfortunately it's harder to deal with humans who are so clueless they look like bots. ;)

[topr8]

>>I want to know how many fake accounts there are at WebmasterWorld. Anyone spot any bots posting?

i spot them all the time - also 'human' bots so to speak ... the mods are very good at deleting the posts quickly and i would also bet there is good bot spotting programmed into the software too.

>>And what about duplicate accounts at ebay !

i don't get this? what relevance has it got? ebay has got a tangible number of sales made through the system, which is how they make money - who cares if they are duplicate accounts?

[Clarence]

Fake accounts don't affect Advertiser Impressions or clicks, so it's not a big deal to Facebook's Bottom line.

Advertising pays for Facebook. Actual logged in people viewing impressions, and actual people clicking on ads.

Don't confuse Facebook's press with Facebook's numbers, and Facebook Advertiser base is increasing Exponentially "Faster then Google".

[Clarence]

Last week, digital distribution firm Limited Press alleged that, based on its own analytics software, 80% of clicks on its advertisements within Facebook had come from fake users.

This story is a complete lie. Anyone who has ever created a Facebook ad would know it's a complete lie. Why would this company go through all this to lie?

When targeting an ad, you can choose who will see it, down to there location. If you advertise your VirtualBagel page to men in India, that's who will click.

BUT, No only are Fake accounts not clicking your ads, it's hard enough to get anyone click your ads.

[wolfadeus]

What do bots post on WW? I can see the use of fake accounts on Facebook (and agree with posts above that they are likely to be higher than 8%), but what is the point for this on a niche website like this?

[hal12b]

<<What do bots post on WW? I can see the use of fake accounts on Facebook (and agree with posts above that they are likely to be higher than 8%), but what is the point for this on a niche website like this? >>

Probably to post a b.s. link to their website in hopes of clicks or ads based on page views.

[Robert Charlton]

Last week, digital distribution firm Limited Press (sic) alleged that, based on its own analytics software, 80% of clicks on its advertisements within Facebook had come from fake users.

Hard to know where to begin on this one, and whether it merits its own discussion. It's also part of this story, and of Facebook's mishandling of customer relations overall.

Limited Run, a niche music marketing company wanted to change its FB page name from Limited Pressing to its new Limited Run, and along the way did some testing which on the surface seemed careful... but it apparently wasn't careful enough... and it came up with some sensational figures that got widely publicized, and perhaps resulted in some major corporations dropping their Facebook advertising.

CNET reports the basic background, and quotes the now deleted Limited Run blog about what Limited Run had observed....

Firm ditches Facebook for Twitter, claims clicks are bots
|July 30, 2012
[news.cnet.com...]

The quoted blog post is worth looking at, because it illustrates the frustration of a small company trying to deal with Facebook.

In reaction, AimClear (which handles a lot of Facebook advertising), brought some stricter analysis to the situation and was critical of the 80% bot click figure in a SearchEngineWatch post...

Facebook Ads 80% Bot Claim, Examined! Why You Shouldn’t Lose Faith Just Yet
August 10, 2012
[searchenginewatch.com...]

AimClear was apparently more careful in identifying bot traffic, looking, eg, at time and date stamps across time zones... and also took issue with the 2% javascript disabled figure. Finally, AimClear also took issue with the possible sample size and challenged Limited Run to provide more data.

On its Update blog, on July 30th, Limited Run posted the following....

When we posted about leaving Facebook on Monday, we only intended our small group of customers and followers to know what was happening, and why. We had no clue it was going to explode like it did. But now, we’re just a very small company, that wants nothing more than to go back to work. We don’t want to be known for this, and we’re going to keep turning down requests for interviews. Facebook is in contact with us, and they’re going to look into the issues we had.