CMX-Banner-Ad

Community Management Strategy: Data, Not Emotions

4363157652_a3e100861d

Not too long ago, I stumbled across a wonderful article by Steph Parker posted right here on TCM. The introduction to metrics is important for community managers.

I’ve felt that the trend has been approach to community management dealing in “feels”—and ignoring metrics. But once you get the hang of this basic calculation by SocialBakers:

1ENGAGEMENT

…you can start to tweak the equation to make more informed strategic decisions. And before I dive into this let me just make clear that this is a very high-level, introductory way to approach content strategy. In community management, there is a TON of data for you to consider.

Let’s make a claim: every social action is not created equal.

Depending on your goals, you can start to break social actions (or Edges) up into a priority hierarchy. Let’s define a basic hierarchy using the variables from the SocialBakers equation. Let’s call my goal “Brand Advocacy”, and we decide that the best indicator of “Brand Advocacy” through post-related Edges (actions one can do to a social post) is apparent in this hierarchy: Shares > Answers > Comments > Likes. Now that we’ve established a priority hierarchy we can apply a value (“weight”) to each action (“Edge”).

Let’s assign weights to each Edge:

Share = 5,
Answer =3,
Comments=2, and
Like=1.

This simple tweak will help you to establish content-type priorities.

Warning: Basic Math Ahead

Alright, so let’s see how we can put this into practice. Here are our two equations measuring our post effectiveness (let’s call this variable F) and Nd = Number of Fans on the day the Post was made:

Original:  F = (Likes+Comments+Answers+Shares)/Nd
Weighted:  F = ((Likes*1)+(Comments*2)+(Answers*3)+(Shares*5))/Nd

Let’s take 2 posts that generate the same amount of Edges – each with 10.

Post 1’s Edge Breakdown: 6 Likes, 2 Comments, 2 Shares
Post 2’s Edge Breakdown: 3 Likes, 3 Comments, 1 Answer, 3 Shares

Let’s say that there are 1,000 Fans. Using the Original equation, each of these posts have a total interaction sum of 10, and yield an effectiveness rating of 1%. Using the Weighted equation:

Post 1: F = ((6*1)+(2*2)+(5*2))/1,000 = 20/1000 = .02. 2% effectiveness
Post 2: F = ((3*1)+(3*2)+(1*3)+(3*5))/1,000 = 27/1000 = .027. 2.7% effectiveness.

Initially, it would seem that these posts are equally effective with the Original equation, but when we input the data into an equation that is more specific to our business/community goals – we see that Post 2 was marginally more effective than Post 1.

Let’s add a Post 3 with Edge Breakdown: 7 Shares
Post 3: F = (7*5)/1,000 = .035. 3.5% effectiveness.

Even though Post 3 had less interaction than Posts 1 and 2, it seems to be more effective using our Weighted equation.

Doing the math is one thing, analyzing correctly is another…

When someone executes an Edge of higher value, like a Share or Comment, you have to consider the impact that action makes for your community and/or brand:

Will a Comment start a conversation that will bring people back again and again to partake in a conversation?
Will a Share bring more people to your digital presence because a deliberate endorsement by the Sharer meant more to them than seeing that they “Liked” something in their Feed?

Figuring out what’s important to your community/brand is absolutely essential when defining the priority hierarchy.

Learning how to gather data—and then analyzing and acting on that data to achieve community/business goals—is a skill that every CM needs to have.  I’ve seen people dismiss data as “not being human” or as somehow being ethically wrong because you’re using tactics to “manipulate” the audience. I don’t buy it. This stuff sounds like a cop-out to me. You should not be shying away from the hard stuff!

Would you like to know more?

A couple of links to some of the concepts I’ve discussed here:

Academic Paper: The role of edge weights in social networks: modeling structure and dynamics

The blog on EdgeRankChecker.com is a great resource for Facebook’s EdgeRank algorithm, however – you can look for similar “behaviors” on other social networks. You can bet the concepts used to create EdgeRank  apply to more than just EdgeRank.

Photo cred: Trindade Joao

Enjoy this article?

Enter your email to receive free community management advice in your inbox:
About the author

Ryan Crowe is a social media strategist and budding theorist currently at Stealth Creative, a full-service creative ad agency located in St. Louis, Missouri. Ryan also sits on the Board of Directors for the NPO "LittleBigFund", where he acts as the Director of Social Strategy.

7 comments
doctorcrowe
doctorcrowe

@MindOverMatthew I find having data removes a lot of the "guessing" from planning and implementing a strategy. Flexibility still needed.

doctorcrowe
doctorcrowe

@melissabalsan Cheetah Gym in Andersonville? I worked there for about 2 weeks before they let me go for being in too many shows.