L'analyse de données à la portée de tous.
Une solution sur site pour les entreprises.
Une solution hébergée pour les entreprises.
A little over a year ago, I started playing bass for a band called Golden Idols. Aside from my extreme bass riffage and overt super-cool-rockstar energy, the band was excited for me to join and bring the skills I had learned over the past couple of years as a marketer. As a result, I got put in charge of the social media efforts of the band.
Our main social media efforts are two-fold: Twitter to connect with other bands, bookers, and music writers, and Facebook/Instagram to connect with fans. I decided to focus on the latter, as Facebook likes are actually an important metric when talking to bookers (they are a lot more likely to book you if it looks like you have a lot of fans).
Early on, every member of the band actively posted on our Facebook page. That led to a few problems. Mostly, the voice was often inconsistent (I/my versus we/ours) and someone was posting a lot of cryptic statuses. I wanted to convince my bandmates that plain-status posts by pages do not get prioritized very well by Facebook and therefore get less engagement, so I visualized the data. I was able to convince them that we should avoid status updates without an image or link, as shown by the dropoff of status posts here:
You can also see that we ramped up our posting of photos and videos. I saw a dramatic difference in engagement for pictures and videos, so I set up our Instagram account with an IFTTT recipe to autopost to our Facebook page, ensuring a steady stream of content.
Now, you can see that cumulatively, we’ve posted photos more than any other type of post and that those get the most consistent number of likes (it probably helps that we are all super-cute).
Working in the marketing world, I had heard that Facebook started giving a lot of priority to native videos in the newsfeed, especially in its app. I really wanted to make a music video, but the boys were less enthused by all the work that would go in it. So I did a little proof of concept by making a short video of dancing GIFs for one of our songs. Looking at the share metrics now, it’s easy to see how I convinced the guys that videos were a good idea.
I used to have to go to my Facebook page insights to download all the data for this kind of analysis. But now, with Tableau’s Web Data Connector, getting the data is super-easy. I used the Facebook WDC from Alex Ross at Tableau Junkie. You can select any Facebook Page (this is different from a standard Facebook profile) that you manage and pull all the page feed data with it. The data you get is perfect for analyzing engagement on your Facebook posts as it contains:
It contains a lot more than that, but I found that a lot of it wasn’t what I expected. For example, Page Likes showed the current number of Likes we have for every post, even if we didn’t have that many followers. Page New Likes showed zero for each post. So, for analysis on the growth of our page, I will stick with Facebook insights data.
Use the Web Data Connector to connect to your own data, and share with us any cool insights you find!