After our last few posts, you now know how to build an online community and what steps you should take to maintain a strong community. Our first recommendation for maintaining a strong community was to listen, and in this blog post, we will explore in more depth how to make listening a measurable activity that yields valuable information about your community, in 3 simple steps.

The way you do this is through designing experiments with your content and measuring how your community reacts. There are three essential steps to getting this right. Each step requires a bit of planning and forethought, but the knowledge you'll gain about who your community is as a whole, what they expect to get out of interacting with you, what they love, and what they hate, will be well worth your effort.

We call this approach being a "metrics-based community manager." Rather than firing out content based only on your brand's needs, what you think your community might like, or even based on what some report tells you about what works for other people's communities, you use metrics to develop a sixth sense for listening to your community. You do this in a measurable way with proper planning, execution, and measurement, and then you take the lessons and apply them to your overall strategy and daily tactics.

 

generic plan

Step 1: Planning Your Experiment

First, we need to figure out what information we want to get from our experiment. This should come naturally, as you're sure to have some questions that, if answered, would benefit the way your brand approaches social media.

To make sure you get the most out of this blog post, I've included one of our own real-world examples of how we've used this method to learn something about one of our client's community and influence their behaviors for the better as a result.

Your plan should outline the following:

  • What problem are we trying to solve/question are we trying to answer?
  • What is our hypothesis? How do we think we can solve this problem, or what do we believe is the answer to our question?
  • What action could we take through our content to test our hypothesis? (Be sure to look for any factors you need to control that could effect the outcome, such as the time you post, for example.)
  • How will we measure the results? What metrics are available that could give us a better idea, beyond what's been made painfully obvious to us, of how our community received our post? (Be sure to take the time to really dig through all of the metrics that your platform and any third-party tools you use make available to you. The more data you can gather, the more you can learn!)

 

Real-World Example: For the client in question, engagement on posts had dropped during the holidays and was slowly picking up again at the start of the new year, however the majority of the comments were junk, rather than relevant discussion. This had never been a problem before with this community, but a couple of weeks into the year, it wasn't really getting any better. So, we theorized that a new post structure developed by our own John Maver could encourage more relevant discussion by hooking readers with a title, giving a very short summary of the included link, and including a call to action at the end that was highlighted with an alt-code character. All of this was created on a post-by-post basis in order to make sure that all of the post's content would remain above the character cutoff limit, after which readers would have to click the "See more" link to read any additional characters in a post. If any additional information needed to be included in the post, it could come after the cutoff, but all of this essential information had to come before the cutoff point.

Here's what our plan would look like:

  • Problem: The majority of responses on content are currently spam and unrelated support requests. We need to get our community back to having real discussions with us again.
  • Potential Solution: The community is still inactive following the holidays. If we improve the way we present our discussions, we may be able to hook more of the people who see our posts into commenting on them or sharing them.
  • Action: We will develop an improved posting structure that hooks our reader, presents everything they need to know to immediately jump into the discussion, and a highly visible call to action to join in on the discussion. Details like how to best highlight our call to action, what our hook should be (title, question, something else?), will be worked out over a series of additional experiments.
  • How Do We Measure The Results?: We should measure the total interactions on each individual post to see if our posting structure improves the overall number of total interactions. We should also measure the average percentage of relevant comments versus spam and unrelated discussion on our current posts, and compare them to the same percentage in our experiment. It would also be interesting to see if the prominent positioning of the call to action influences more community members to take the desired action, and if this applies to multiple calls to action, or only a specific few.

 

Business  people discuss an idea in front of the computer

Step 2: Executing The Plan

As long as you did a great job with your plan in step one, this should be the easiest part of the process. Post your content, interact with your community as you normally would (unless a change there is a part of the plan), and follow your plan to the letter. You should make note of any anecdotal observations during the execution stage. "Wow, X% of these comments are relevant discussion, rather than spam or customer service and support requests. That's way more relevant responses than our other posts have been getting!" However, don't forget that while this supplemental information is really valuable when you get to step 3, it doesn't replace using any relevant metrics that are available to give you hard data on how your community reacted.

 

Real-World Example: We tried multiple forms of the structure and different alt-code characters to determine which were the most effective. As always, we were active in the discussions on these posts, and during these conversations, as we made note of the effects our changes to the post structure caused. We quickly noticed that the number of relevant comments to the discussion at hand were increasing dramatically.

 

bigstock-Analyzing-data-31328702

Step 3: Measuring The Results

It's tempting to think of this last step as something that happens after the experiment, a reactionary step that can be put off until after. However, I would caution against this. As I mentioned before, understanding how you will measure the results of your experiment should be a part of planning the experiment. Without being able to properly measure the results, all of your preparation and execution is worthless, because even if you got great results, you need to understand why you got great results. If you don't know how you did it, you won't be able to repeat those results again when you need to, like in the middle of an expensive campaign. Even worse, in some cases, if you don't look at the metrics to understand what's going on, you may not even realize anything significant happened at all!

 

Real-World Example: We could already see from looking at the comments themselves that our experiment had a big impact on the number of relevant comments, but we didn't stop there. By looking at our available metrics we were able to also determine that the posting structure lead to an overall increase in interactions per post, that our prominently positioned calls to action lead to a measurable increase in the number of people who heeded our call to action and took the desired action (such as sharing the post, or clicking the included link), and that our posts using this structure were less often marked as spam or hidden by our audience.

Simply reaching our initial objective and proving our hypothesis was a win, but by looking at additional metrics we were able to identify other ways in which our post structure was benefiting our client and their community! When you're done, make sure you don't forget to properly analyze your metrics so you understand why things turned out the way they did, and be sure that you apply these lessons to your future activities so that you're always pushing your social media strategy and tactics to new heights!

 

There is no shortage of great information about what days and times are best to post, what content people care about, how to plan content that will naturally benefit from some degree of viral spread, and other data pieced together from studies of tens, hundreds, or thousands of online communities.

That data is great, but realistically none of this is  the same across multiple industries. Even communities hosted by brands in the same industries can be vastly different in how they interact, what they are interested in as a whole, and which content they will love and hate. That's what makes these experiments so valuable, they give you information you can't get anywhere else, they teach you who your community is, not who an average of 100 other similar communities that aren't yours are! That is the power of listening to your community!