Why Facebook Button Counts are Wrong

by Ari Herzog on Jan. 29, 2011 · 26 comments


Facebook count inconsistencyUsing yesterday’s blog article as an example of the type of Facebook buttons you see on many websites and blogs enabling you to click and like or click and share content with your friends, you can see 12 people clicked the Facebook Share button (in the bottom right of the accompanying screen shot).

However, the breakdown indicates only six people shared the article on writing negative comments; which increases to 12 when you add five people who clicked the recommend button at the bottom of the page and one person who added a comment to that recommendation.

Someone at Facebook HQ goofed up.

Kevin Sablan is responsible for the reality check of how many shares and likes were on the page, for he created a browser bookmarklet called RealShare which you can read about and watch a video how to install it on your favorite web browser.

Is the difference between a share and a like enough of a difference to make a fuss? That’s your call, but if you run any metrics on how many people like your pages, you may now need to edit those metrics to breakdown people who merely like what you write versus others who share it.

Related articles you may enjoy:

{ 26 comments }

Marcus Sheridan-The Sales Lion | Tweet @
January 29, 2011 at 2:02 PM

Hmm, good observation here Ari….I wonder how Google’s great ‘algorithm’ will account for that ;-)

Marcus

Ari Herzog January 29, 2011 at 6:32 PM

What’s Google got to do with Facebook?

Marcus Sheridan-The Sales Lion | Tweet @
January 29, 2011 at 8:01 PM

Just that there has been ‘talk’ of ‘Google factoring in more social media stuff, you know– likes, tweets, all that jazz. But you know how ‘rumors’ work. ;-)

Kevin Sablan January 29, 2011 at 4:40 PM

Hey there Ari, I’m glad that you too are concerned with this accuracy issue. You’re right, it’s totally up to each person watching their metrics to decide whether or not the difference matters to them. For me, the number of comments is the biggest issue, since some of those comments might be saying negative things about the page … which is the polar opposite of “liking” or recommending it.

Ari Herzog January 29, 2011 at 6:33 PM

Why would someone write negatively if they choose to LIKE/RECOMMEND something? It could happen, but…? ;)

Kevin Sablan January 30, 2011 at 10:57 AM

Ah, that’s the problem. Let’s say I share an link on Facebook. That counts as a share/like/recommend. Now let’s say my friend Ari sees that link on replies with a comment: “That is the worst article I have ever seen. I recommend that none of my friends click on that link.” As far as the counts on the buttons are concerned, that just counted as another share/like/recommend. In no way is that truly a recommendation. On the back end, FB knows that it is only a comment. On the front end, they don’t make that distinction.

Wayne @ Southern California | Tweet @
January 29, 2011 at 6:22 PM

“Someone at Facebook HQ goofed up.”

I can’t help but smile when I read that. It wouldn’t be the first time, unfortunately. They certainly don’t inspire confidence in that regard, do they?

Ari Herzog January 29, 2011 at 6:34 PM

Heh, I added that sentence at the end. Glad you appreciate.

Vee Sweeney @ Internet Marketing Trends January 30, 2011 at 4:33 PM

I agree that it is downright confusing and it makes it difficult for those who want to track their social media statistics easy without trying to pick through them to get a rough estimate that they can go off of. In this case, the number or difference is small but it does make me wonder about how large of a margin of error can happen. Is this just a one time fluke (I doubt it) or can’t people trust Facebook’s stats all together? If the post had 1000 reactions, would the difference then be 100? That’s what I am worried about.

Graphic Design Melbourne | Tweet @
January 30, 2011 at 7:50 PM

Hi, Ari. A marketer who is seriously watching out for his metrics would definitely notice the distinction between the number of likes, shares and comments he got and he would want these logged correctly. However, the systems that keep track of these things are not perfect. They goof up more often than we like. So, what are we to do then? Do we just ignore them and look in on other metrics? Or, do we call up the attention of the people behind the systems, just like what you are doing now? I think, you are always among those persons who are not afraid to do the latter. :)

Christopher Roberts@Philosophy Blog January 31, 2011 at 5:03 AM

Have you been following me Ari?
I read that article just over a week ago, and I commented as well!

I will say what I said on Almighty Link again, for the benefit of your blogging community:
“I never knew that! That is madness and Facebook really should fix it. Just checked my site and indeed you are correct. I installed the Facebook Like button earlier this month and it says I have 18 likes. Go and check it out yourself, as you will see it’s actually only 6 :(

I am not going to disable it yet, but I do agree with you that it is a bit odd.”

I see what Facebook have done, they are trying to show how popular a site/blog is ‘all in one’ but that does slightly deceive readers. However what is the alternative – a ‘Like’ and a ‘Comment’ and a ‘Share’ button? That’s a lot of buttons which means longer page load time and a more cluttered blog…

What’s your answer to the problem Ari, as do you really feel that just the number of ‘Likes’ is enough to display a sites/blogs popularity?

Are comments and shares not important too?

Delena Silverfox@Coupon Codes January 31, 2011 at 6:31 PM

This really makes me wonder now. There are so many pages out there with tens of thousands of “Likes,” and I wonder if it’s as important now to focus on the number of fans who Like a page or the number of people who are taking action in the page’s favor? With so many products out there promising to show you how to grow your fan list on Facebook, your article shows more than ever the importance of knowing exactly what you want out of people in regards to interacting with your page.

Thanks again for such wonderful info, Ari!
Delena

John February 1, 2011 at 7:58 AM

Hey Ari,

That’s something I have never noticed myself. From what you have shared here I think the difference between a share and a like is enough to make a fuss. It seems as if Facebook has been making a lot of goof ups. Thanks for sharing your observations Ari.

Jeff Esposito | Tweet @
February 1, 2011 at 3:21 PM

The same can also be said about the Tweet button. Much less reliable than the old TweetMeme

Christopher Roberts@Technology Blog February 2, 2011 at 12:22 PM

Really, I never knew that…

What does the Twitter button include Jeff?

Jeff Esposito | Tweet @
February 2, 2011 at 12:49 PM

Chris – the Tweet button shows the number of times something is tweeted, but misses ones all the time by not counting Tweets that you can find on search.twitter.com

Christopher Roberts@Technology Blog February 2, 2011 at 1:39 PM

Right… so that is more likely to display less than the true value, whilst the FB one is likely to display more… CONFUSING!

Melz February 3, 2011 at 2:27 AM

Seeing the breakdown of it, there’s really a big difference. Summing up it’s 12..which goes on shares..of which other thinks that it was shared to 12 person…but if we are to look at the breakdown, those 12 ..goes to different factors… shares, likes and recommend.. hmmmmmm…

Ken Cook February 3, 2011 at 8:57 PM

Thanks for addressing this. I had thought something wasn’t right with the buttons but never took the time to really dig into it. I love the tool from Kevin Sablan – thanks, too, for pointing that out. Would love to have you come on SocialMediaEdge.com on BTR to discuss this (and your services).

Cassy Jhon February 5, 2011 at 4:01 AM

HI, Ari Herzog,

This really makes me wonder now. There are so many pages out there with tens of thousands of “Likes,” and I wonder if it’s as important now to focus on the number of fans who Like a page or the number of people who are taking action in the page’s favor? With so many products out there promising to show you how to grow your fan list on Facebook, your article shows more than ever the importance of knowing exactly what you want out of people in regards to interacting with your page.

Thanks again for such wonderful info, Ari!

Ari Herzog February 10, 2011 at 3:36 AM

Quality always trumps quantity. The number of fans, followers, connections, etc you have is meaningless in the greater scheme of things. If I like your page but I never interact on that page when you ask questions or share information, what value is my like to you?

Parekh February 6, 2011 at 2:04 PM

I believe the whole point (or remedy that can be done) is to have something like a unique filter. Once the user already triggered a recommendation, that person will not be allowed to recommend something in the blog. The “Edit” functionality also helps out in updating information (if available).

Cassy Jhon February 9, 2011 at 6:37 AM

I never knew that! That is madness and Facebook really should fix it.
From what you have shared here I think the difference between a share and a like is enough to make a fuss.

http://getridofacoldsoresfast.com.com” rel=”nofollow”>

Udegbunam Chukwudi@Patchwork Of Tips | Tweet @
February 19, 2011 at 1:57 PM

Of late I’ve noticed that Facebook shares and likes have the same number of counts thus I prefer having just one of them on my blogs than both ;-) .

Besides I’ve noticed that having the Facebook Share button and Facebook Comments plugin creates some sort of conflict.

Melinda February 20, 2011 at 2:13 AM

Thanks Ari for the great post. I agree that it is downright confusing and it makes it difficult for those who want to track their social media statistics easy without trying to pick through them to get a rough estimate that they can go off of. Best of luck.

Cheolsu June 21, 2011 at 12:48 AM

I understand that this article was published much before Google’s “+1” came along. Does anyone know if Google “+1” counts are wrong too? I have just inserted it to my website.

Comments on this entry are closed.

Previous post:

Next post: