Andrew Teman

Posts tagged stats
Engagement Is Flawed

I had the opportunity to sit on a fun panel the other night with some smart people, including Anne, Eric, George, Janet, and Ted where we discussed lots of different social media related stuff. The final question of the night, was something to the effect of “what will be the big trend or thing to watch for in 2012?”.

The first part of my answer was that I think most media plans are going to start BEGINNING with mobile, rather than sprinkling mobile on as an additive. The second part of my answer was part soap-boxing and part wishful thinking, and I basically said that in 2012, I think the smart marketers out there are going to begin really asking what all of this social media stuff really means. Metrics and methods are really going to be questioned, as brands and marketers begin to scale WAY back on social, focusing only on the parts of the space where gains can truly be seen and measured with real confidence.

Along these lines, my most sincere hope for 2012, is that we once and for all abandon the “engagement” number as a measure of success.

And I say this, because quite simply and literally, “engagement” is at best directionally flawed, and at worst, harmfully misleading.

For Facebook specifically, engagement is generally defined mathematically as the sum of comments and likes on a given post, over the impressions that the post received (this stat is actually called “Feedback” on Facebook now). So if a post was seen by 100,000 people, and received 200 comments and 300 likes, your engagement formula would be (200+300)/100,000. And your engagement rate would be 0.5% (we’ve multiplied it by 100 to get the final %).

Simple enough, right? And I think we can all agree, that we’ve seen (and even used) this stat as evidence of a post, page, or campaign “working” or being successful. High engagement equals social media success! Done and done.

But here’s the problem, those comments that you’re tallying up as part of the numerator, most likely contain some nastiness. Some of those people in that group are probably complaining or hating on your brand. It’s even possible that most of those comments are negative. So if within those 200 comments, 175 of them are telling you that the brand sucks, do we still count those as “engaged” users?

This is a comment on the page of local cable provider RCN (name has been removed).

RCN Facebook Comment

This is a comment on the page of TiVo (name has also been removed).

TiVo Facebook Comment

Should these both be counted as carrying equal weight? I personally don’t think so. But they are counted equally under the engagement-as-a-measure-of-success methodology.

So in 2012, let’s start thinking beyond the top-line numbers and the tweet-able headlines, and really start to understand what is actually going on in social. Because if we don’t, the bell is going to be tolling for this whole industry really soon…