Actually, Social Media Buttons Work Really Well

2012-05-31

Yesterday, an epic rant against social media buttons titled Sweep the Sleaze made its rounds on the Web. The gist: Don’t bombard your audience with buttons to share stuff on social networks. Instead, make good content that they’ll be compelled to share organically.

The advice is well-reasoned and generally correct, and was covered by The Oatmeal a few months back and by 37signals a few years back.

But this part smelled fishy to me:

But do these buttons work? It’s hard to say.

And this:

Some people probably do use those buttons. Maybe even a lot of people. And maybe you do and think I’m dead wrong about this. Maybe I am. And maybe someone needs to do some serious research to know for sure.

It’s quite easy to do a bit of research and find out just how many people use those buttons, particularly the Twitter button thanks to Twitter’s data-generous API. So I sought to answer the question:

If a website displays those annoying Twitter buttons, just how many tweets linking to it are made by people using those sleazy-ass buttons?

Here’s a gist of the script I used:

It scans through the latest 500 or so tweets that link to a particular website, and checks the source field of each tweet, looking for the clearly marked Tweet Button.

I first ran the analysis on a new website called Politwoops that was launched yesterday by my former employer, The Sunlight Foundation. The site is a good test subject because it displays the Twitter button prominently in the header, it’s about Twitter, and it went a bit viral yesterday.

Search term: politwoops.sunlightfoundation.com
Total tweets analyzed: 498
Tweets from button: 111 (22.3%)

So here we have a Twitter-savvy audience, and 22% who linked to the site did so using that flippin’ Twitter button on the website.

How about the 37signals blog, an even savvier audience made up of folks who’ve been on Twitter for half a decade? Surely, none of them will find use for that superfluous tweet button that appears on the bottom of each post:

Search term: 37signals.com/svn
Total tweets: 499
Tweets from button: 70 (14.0%)

Smaller, but still significant.

Switching audiences, let’s look at viral media website BuzzFeed. They serve content optimized for sharing on social networks, and they prominently display social media buttons on each post.

Search term: buzzfeed.com
Total tweets: 495
Tweets from button: 191 (38.6%)

Huge! BuzzFeed’s audience consists of more normal folks, not the ones who get worked up over a topic like this. The buttons are meant to remove friction, and here we see that they really appear to be doing so.

It’s time for a confession: I knew those buttons were effective well before running this analysis, because my startup Upworthy is all about spreading stuff that matters through social media. Here’s the analysis for our website:

Search term: upworthy.com
Total tweets: 499
Tweets from button: 319 (63.9%)

You can tell by viewing one of our nuggets that we really, really optimize for sharing. Yes, we have buttons that are hard to miss. But we’re also very thoughtful about crafting share text that will (a) make you want to use it to share and (b) make others want to click.

It’s true that a social media strategy simply isn’t about sticking buttons on a website. It requires high quality content, and just as importantly, packaging that content in a shareable way. It’s not sleazy to try to get your audience to share your stuff. Share buttons are well-placed reminders that sharing is an option.

You can look at a page on Upworthy and think, “Boy, that share button stuff is so offputting. People who fall for it must be gullible buffoons!” Or you can look at it and think, “This content is good and important and should be shared. Buttons that make it easier to do so are worth the pixels and bandwidth.” If you’re in the latter camp, I should mention that we’re hiring.

And if you think this piece of content is worth sharing on Twitter, please go ahead and click the button below…