What do you do when what you know stops working?

Last year we spent the month of November experimenting with our approach to Facebook and we wrote about our results in the appropriately titled 5 things we learned from our November Facebook experiment. A lot has changed with Facebook since then and we’ve been experimenting with new approaches and features accordingly. Here’s what we’ve learned so far.

Got any questions or want to share your own Facebook marketing experiences? Come tell us about it on our Page!

What changed on Facebook

Since last November, Facebook has introduced new features- Live video being most prevalent among them- and changes to the News Feed algorithm that favor posts from friends and family over that of brands. At the announcement of the latter, many brands were understandably concerned about continuing to reach their audiences on Facebook organically.

We took our own advice for this platform update, asking what the update changed, if it should change our goals and if we had the resources to experiment with it. Organic reach was definitely lower and there were some new features Facebook itself seemed pretty excited about, so our next step seemed pretty obvious: How could we better our visual content strategy on Facebook while still leaving enough room to experiment?

What we did differently

Based on Facebook’s updates, we decided to take our video experimentation further, branching out from just sharing short pre-recorded tip videos to some Facebook Live videos that would act as “office hours” sessions, addressing our new product updates and sharing social media tips and tricks.

Office Hours 5

Throughout the third quarter, we tested different times and days to broadcast on Facebook Live, eventually settling on a regular broadcast every other Thursday at 4pm CT. We’re hoping as Facebook rolls out new features to all users- like the ability to schedule Live sessions and have a waiting room for viewers before they start- we’ll build a more consistent audience who can get their questions answered around not only our products but social media marketing in general.

What we learned

Building an audience isn’t something that happens overnight, so we’re still experimenting with our approach in our Live and other videos to see what our audience likes the best and making sure that video is still the thing they’re most interested in. That means paying attention to the metrics, of course. When you take into account our boosted Facebook posts, links are our most successful post types in terms of both reach and engagement for Q3:

FB Post Types I Q3 2016

But when you take boosted posts out of the equation, video posts were our most successful in terms of both reach and engagement, followed by photos.

FB Post Types II Q3 2016

That means our audience is still very organically interested in the visual content we have to offer, so we’ll continue to make that a large part of our Facebook content strategy while still leaving room to experiment with other post types as we think of new ways to present them.

How we can go deeper

The screencaps you see above are from Facebook’s native analytics, which we use in conjunction with our own Union Metrics’ Facebook analytics for maximum feedback on our performance. Using our fan and audience reports, we can dive into the times of day our audience is most active, as well as pay attention to factors such as age, gender, geography and language that can affect the types of content people like to see.

When FB fans are online Q3 2016

So what’s the big takeaway? Video is still working for us, so we’ll stick to that for the moment while also leaving a little room for experimentation in our Facebook content strategy, especially visually.

Have you run a Facebook experiment of your own? How did it go? Tell us about it!