This is a small part of a larger marketing dashboard completed using Facebook page data. The goal of this dashboard was to create some actionable insights for the company. Visualizations are created using JavaScript, Python, and HighCharts. One of the benefits of using JavaScript to embed web visualizations is that the dashboard can be viewed on any device, anywhere in the world.
The first chart (the scatterplot) shows us that for this particular organization, posting a photo is by far the best option. Photos produce more engaged users and result in more consumptions than any other post type for the groups. This result may not be the same for other companies - it depends on the audience. Some audiences may prefer videos, for example. So it's important to look at your own company's individual social media data to understand how to best engage with your particular audience.
The section chart (stacked bar) tells this company how they can better reach out to users who do not currently "Like" their page. Once again, photo wins out here - the highest percentage of non-likers engage with a company's post if it's a photo.
In the Engagement Decision table, we can look for patterns going across or up/down to make decisions. There aren't too many strong patterns (and this happens sometimes!) but we could state that the "Unique Likes" column has the most dark blue, indicating that unique likes are easiest to get regardless of post type. We can also see that the Photo and Link rows have a lot of blue, indicating that they are popular post types in terms of unique shares, unique likes, and unique comments. You can also use this table to pick out particular values - for example, answering the question "How many unique comments do I get on status posts?"
Finally, the sliceable line chart at the bottom of the page shows us how our engaged users and engagement rate change over time. Clicking "Engaged Users" changes the line to show users, and clicking "Engagement Rate" changes the line to show the engagement rate.