Engagement Rate: The Holy Grail of Social Media Content Analysis

A couple weeks ago, I was beyond thrilled to be a part of the Social Media Association of Michigan’s (#SMAMi) inaugural Ann Arbor #SOCIALCON, where I joined fantastic speakers like Nikki Sunstrum, director of social media at the University of Michigan; Kollin Currie, social media defensive team lead at Quicken Loans; and Chad Schaeffer, social customer service lead at Salesforce (killer company, right?!).

Banner to promote the Ann Arbor #SOCIALCON Conference

When the wonderful president and executive director of #SMAMi, Sola Obayan, initially asked if I would like to come and speak, she doled out several topics she knew Ann Arbor marketers would clear their schedules for. I chose social media measurement, data and analytics. Then I sat down to write it and realized that I might have bitten off the biggest chunk of subject matter in existence. I knew I had to narrow. So, I broke the presentation into two sections: everyone’s elusive favorite, social media ROI (another blog to follow!), and the main event: social media content analysis using engagement rate.

Why should I analyze my social media content?

There’s no beating around the bush here: social media content analysis takes time. It’s also not easy. There is not a button you can push that will detect different, non-text-based trends in top-performing and low-performing content (though this capability is emerging with tools like Curalate and Crimson Hexagon). So, why do it at all?

Theoretically (and I hope this is the case in practice), you should be producing content for your business that supports overall marketing objectives. Therefore, there is a purpose for your social content, supporting the customer journey at different points within the purchasing funnel. If content consumption is crucial to your social media marketing strategy (which it should be), then as social media marketers, it’s our job to produce content that our audience has told us they would like to consume. Trust me, sometimes the answer will surprise you.

Tools like Simply Measured, Socialbakers, Buffer, Sysomos Expion, etc. can help expedite this process. They can assist with data pull, documentation and visualization, and can even call out top-performing posts by a given metric.

What these types of tools also offer is the ability to analyze different components of the posts you’re publishing. Take the following post qualifiers, for instance:

  • Social network
  • Post type (photo, video, link, status update, Facebook Share, etc.)
  • Time of day
  • Day of week
  • Content category

Tools can analyze performance of all of your posts within a given time frame and tell you what works best overall. Does your audience respond best to content on Instagram? Are they more likely to engage on Sundays and during the evening? Maybe they prefer photos to video.

But tools don’t do all the heavy lifting for you. That’s where engagement rate comes in.

What is engagement rate?

What is engagement rate, and why is it so important for analyzing content? Engagement rate is actually a very simple calculation. In this context, we’re talking about it on a social media post-by-post basis:

Engagement Rate = Engagement / Reach * 100

Translated into English, that is the percentage of people that do something (e.g. like, comment, share, favorite, repin, +1, click, etc.) with the piece of content that they’re served on a social network.

Engagement rate is super important because it allows you to optimize your future content generation for what your unique audience has communicated that they care most about. It’s also a great metric because it eliminates a lot of other social media metrics that can skew your perception of post performance based on other variables; for instance, promoted post or tweet budget that inflates reach and associated engagement based on paid delivery, or a network’s propensity to favor a certain post type over another.

How do you use engagement rate to optimize future social media content?

Engagement rate can help you determine which post qualifiers (like the ones listed above) are the most successful for your brand. So, for example, if you were to analyze 10 tweets on Twitter, half of which were status updates and half of which were photos, you would add up the total engagement for status updates and divide that number by the sum of the reach for all status updates. You’d then do the same thing for photos before multiplying each value by 100 to get your engagement rate by post type. If status updates come in at a 3% engagement rate and photos at an 11% engagement rate, you can hypothesize that photos will perform better for your unique audience. Over time, these data-based hypotheses can be proven or disproven based on the analysis of future data.

Content categories are another part of this content optimization process. Essentially, they are buckets that classify the subject matter of your social media posts. They are developed to support marketing objectives and drive content strategy. Some tools like Sprout Social, Adobe Social, Rignite, LiftMetrix, Cortex and Rallio, among others, allow you to categorize or “tag” your different posts when publishing, so that the data pull and analysis process is more automated. In lieu of that, you can always calculate engagement rate by content category manually.

If you analyze the overall engagement rates of posts, you can also reveal certain characteristics that go beyond the commonly analyzed parameters above. For instance, are there certain hashtags that your audience responds to? Do they engage more with photos containing kittens versus puppies, etc.? Putting your best-performing and worst-performing content side-by-side and looking for commonalities can be a great way to detect more qualitative trends, create hypotheses and test over time.

These results can vary based on social network, so it’s best to keep that in mind during your analysis.

Putting theory into practice with social media engagement rates

I’m a big fan of empowering attendees when I speak. Sometimes you leave a conference, feeling really great and rewarded and then you ask yourself, so how the heck am I going to apply that monstrous, juicy glob of knowledge to my everyday work life? To bring the theoretical talk back down to earth, I opted for a mid-session workshop during #SOCIALCON, where I “made” my unsuspecting attendees break out their calculators and broach something that many creatives dread from the depths of their REM cycles: math.

Calculate engagement rates and perform social media content analysis on your own!

Introducing fictional brand Decision Made Ice Cream.

Screenshot of Fictional Decision Made Ice Cream Facebook page

These guys have a Facebook page and regularly post content to it that falls in one of three content categories:

  • Flavor Rage: Celebrating the product
  • Bribe with Prize: Incentivizing interaction with free scoops of ice cream through contests, sweepstakes or games
  • Food Challenges: Documenting the ridiculousness of eating bizarre combinations of foods and/or large quantities of it

Luckily, you’ve got access to six of their most recent Facebook posts (pictured below by content category). For each of the following, calculate the engagement rates (ER) for these posts:

  • All combined (overall ER)
  • Videos
  • Photos
  • Flavor Rage
  • Bribe with Prize
  • Food Challenges
  • Morning
  • Afternoon
  • Evening
Flavor Rage 
Fictional Orange Blaze Facebook PostFictional Clothe the Cone Facebook Post
• Post Type: Photo
• Time: 10:18 a.m.
• Engagements: 24
• Reach: 1,087
• ER: 2.2%
• Post Type: Photo
• Time: 9:36 p.m.
• Engagements: 47
• Reach: 1,398
• ER: 3.4%
Bribe with Prize 
Fictional Social Phrase of the Day Facebook PostTag a Friend Fictional Facebook Post
• Post Type: Video
• Time: 2:59 p.m.
• Engagements: 95
• Reach: 3,009
• ER: 3.2%
• Post Type: Photo
• Time: 7:07 p.m.
• Engagements: 202
• Reach: 4,365
• ER: 4.6%
Food Challenges 
Sweet Victory Fictional Facebook PostDecision Made Doozy Fictional Facebook Post
• Post Type: Photo
• Time: 6:52 p.m.
• Engagements: 102
• Reach: 3,009
• ER: 3.4%
• Post Type: Photo
• Time: 6:48 a.m.
• Engagements: 313
• Reach: 3,987
• ER: 7.9%

Generate your own insights from data!

Now that you have the numbers, it’s time to generate some insights. Answer the questions below:

  • What was the top content category by ER?
  • What was the top post by ER?
  • What was the bottom post by ER?
  • What did the top two posts by ER have in common?
  • What did the bottom two posts by ER have in common?

Time to hypothesize

With those insights, now form hypotheses to test in the future. I’ll start you off with some:

  • Posts do better when published in the morning.
  • Photos perform better than videos.
  • Decision Made’s customers respond better to posts with women in them.
  • Fill in
  • Fill in

Phew, now that you’re all done, let’s check your math!

Key Performance IndicatorsResult
Total Engagements783
Total Reach16,855
Overall ER4.6%
Video ER3.2%
Photo ER5.0%
Flavor Rage ER2.9%
Bribe with Prize ER4.0%
Food Challenges ER5.9%
Morning ER6.6%
Afternoon ER3.2%
Evening ER4.0%

Did you get them right? For answers to the insights question and for more justifiable hypotheses, just email socialmedia@regroup.us.

And if you want to check out my #SOCIALCON slides, they’re on SlideShare!

Tags: , , , , , ,