Measuring Success

I’ve been thinking about success in social media a lot lately. What is a successful social media action or campaign, how do you measure social media and what are some examples of good measurement. I’ve been thinking about it for a number of reasons, but one of them is you guys. I like to give you good advice and back it up with either stories (anecdata) or data (real data). That way you can use it as ammunition against those who seek to keep your agency in the stone age (or the late nineties). A key part of touting those anecdata and real data is determining what is successful, and how do you do that with social media?

In the past, I’ve reported out counts. Numbers of followers, numbers of likes, numbers of retweets. The rationale was sound, as it did the same as I’d seen from traditional marketing efforts: number of impressions. But is someone following your Twitter account really indicative of you successfully reaching them? (Similarly, is someone driving by your billboard really indicative of you advertising to them?) Given the number of zombie and resold accounts, probably not. And that’s the issue this article on Defamer makes:

Impressions and uniques mumbo-jumbo aside, there is one major takeaway here: Twitter metrics simply cannot account for whether the show is actually well-received, only how much it’s tweeted about.

So there’s no way to tell how many of the 178,500 authors tweeting about Miley: The Movement—reaching more than three million fellow tweeters—actually like Miley and would buy products advertised against the show, and how many—like me—were simply begging for Miley to go away.

On the other hand, there is new research out showing that Facebook Likes might actually mean something, if only in persuading others within a person’s social circle to react positively. (Which, if you know anything about government outreach campaigns, is a HUGE deal.) See this article from

According to the report, “liking” an article will almost certainly encourage friends to “like” it as well, even if it’s not all that good. Unfairly criticizing an article won’t have the same effect.

Now, that’s all well and good, but it doesn’t really tell us what is successful, just that something Liked is more likely to become successful.

What would be perfect is a way to tie a specific social media message to a specific action. When I post this tweet, did anyone resolve to stop smoking? Did anyone put down their Snickers bar? NSA-like powers that all government agencies are purported to have could help with that, but honestly, we don’t have those powers. In lieu of that, though, how do you define success in social media?

One interesting way that I’ve seen it done was through a partnership of Major League Baseball and Twitter. Instead of tying a social media message or campaign to a specific activity and declaring black or white, failure or success, they measured social media messages against each other and saw which type got the most traction. Which worked best. They asked five teams to rotate through five different types of livetweeting and measure their interactions with their public. They even used the teams non-tested games as controls to see if the tested methods were an improvement. The article has a wealth of data, but here’s the high points:

1) Takeovers (of the team’s Twitter account by a famous person) bring more followers.
2) Vine videos get more retweets.
3) Vine videos get more mentions.
4) Vine videos get more favorites.
5) The use of official (team-sponsored) hashtags were highest with play-by-play

While this might seem like an advertisement for using Vine, MLB baseball and Vine seem to go together quite nicely. What I really wanted to point out was the testing that went into this. Measuring success by counting eyeballs isn’t great, but measuring success by win/fail/black/white isn’t great either. Showing levels of success and what types of messages are more likely to succeed might be where we should focus our measurement efforts.


One thought on “Measuring Success

  1. Hi, interesting topic for discussion. In our EPAAC “I’m a fan of life” social media & gaming campaign on cancer prevention ( we noticed that paid FB advertisement has greatly improved exposure of our content. It’s not what we were hoping for – we hoped that celebrity branding and gamification would do the trick – but that’s how it was.
    The issue of impact measurement (i.e. actual lifestyle change or at least persistent awareness of the issues presented) remains unsolved for us and we will keep working on that. More details here:
    Best, Matic

Comments are closed.