20 comments

[ 7.7 ms ] story [ 54.7 ms ] thread
"A Like that doesn't come from someone truly interested in connecting with a Page benefits no one."

That is clearly untrue. If that were the case, fake Likes would not be a problem.

Is this only for fake likes going forward, or are they going to be retroactively be removing fake friends?
They already active remove fake accounts. I've had some of my fake accounts closed down. In my case they were only fake because a service required I log-in with Facebook, and I don't want everything and anything associated with my main facebook account.
How do you think they identified your account as fake?
I honestly don't know. What I do know is that the account was associated only with an email, and the only activity was the site or app I wanted to approve. I didn't add friends, and didn't accept friend requests.
You didn't add friends is why... And I assume you accessed it from the same ip as your real one?
That's the thing though, I don't want to add friends. This is a fake account, it has no real friends. Facebook says they value the social experience of connecting with real friends ...
Not being phone verified, not adding new friends, IP address is definitely a consideration, but you used to be able to have up to 10-20 on the same IP...
"On average, less than 1% of Likes on any given Page will be removed"

Since this 'likefraud' is concentrated on branded/corporate accounts, shouldn't Facebook disclose the average impact to likes on these important pages? Or at least break down the impact of the fraud on a more granular basis.

The way that Facebook is handling this situation seems evasive and they are clearly trying to soften the PR blow here and spin things so they don't sound so bad. With reports coming out of major advertisers seeing over 10% of their likes disappearing, the Facebook team should really publicly address those cases, even if they are not the 'average.'

> These newly improved automated efforts will remove those Likes gained by malware, compromised accounts, deceived users, or purchased bulk Likes.

What about ransom Likes? The ones that happen when a user tries to interact with a site, only to be presented with "Like us on Facebook to see this content!" messages. Those don't fit into any category listed above, but it's hardly an honest reflection of the relationship the user has with a brand.

i dont understand how they could verify that the user clicked on the like button - does facebook have some sort of callback api that allows the developer who wishes to embed a like into their app to know about users clicking on the like button?
Then perhaps you need two different mechanisms for "bookmark" and "like"?
What about to like using hidden iframe Likejaking?

There are many pages having huge Likes using this technique, I can't see any correction in Likes there.

Why can't they make Like count to zero once a site caught using such technique.

I'm actually testing a theory of the benefits of fake share counts on one of site's right now:

http://TimeForZen.com (look at the social widgets up top)

ALL of the share counts for the social widgets are fake. I made them up. I want to see if having high share counts will create a bandwagon effect and encourage more people to share, or if it will keep people on the site longer due to a higher perceived usefulness.

So far it's not doing anything for me due to low traffic, I'll have to run the experiment with some other sites.

I did some testing with a client recently (small local business with relatively high customer engagement and satisfaction) and the difference in people willing to press a like button is substantial between 0-100 but stops mattering afterwards.
I find it funny that any time facebook post some news there will be hundreds of idiots posting their facebook tech support questions as comments. As if they'll ever be answered.
I noticed that too, I imagine there is a lot of pent up demand for support and for people looking for an avenue to voice their complaints.