5 things we learned from our November Facebook experiment

In an effort to practice more of what we preach, we spent the month of November trying some different things out on our Union Metrics Facebook Page to see what works for us organically – we didn’t boost any posts for the whole month – and we thought we’d share those results with you here.

We’re also going to share how we did it, in case you want to try an experiment of your own! (And we can help you measure, too.)

How we did it.

We built a specific Facebook content calendar to plan out what time of day we were going to post- twice a day, usually once in the morning and once in the evening- with the type of content we were posting. A few things got shuffled or changed as happens in most content calendars, but this way we knew we weren’t posting three status updates in a row or always at noon.

FB Nov Experiment


And color-coding lets you know at a glance what’s left to be done. Find a system that works for you. 

At the end of the month we looked over our metrics to see which posts had performed well and looked for patterns in the type of content, time of day, and other factors to keep planning going forward.  You’ll be more successful boosting posts if you boost the ones that already do well organically. But we’re getting ahead of ourselves.

What we learned.

1. Video and photo posts performed the best, especially video. 

This didn’t surprise us given the popular nature of visual content on social media in general, but as we stressed in our post about how the News Feed works, it all comes down to your audience, their preferences and the always-changing News Feed algorithm. It seems like our audience does prefer visual content, at least for the moment.

Always use general best practices as a base line for testing with your own audience.

2. Natively uploading visual content is key to performance.

Another result that wasn’t surprising, but is now confirmed for our particular audience. Uploading photos and videos directly to Facebook gets better results than sharing them from another site with a link.

3. Those videos don’t have to be Hollywood film quality, as long as they’re useful.

We’re a small team with limited resources, so producing high-end video several times a week isn’t something we’re equipped to do. Instead we opted for a series of quick tip videos from our Social Media Manager, featuring tips from our own original research and blog posts.

FB Videos

This was inspired by a presentation at SocialPro from the First Entertainment Credit Union Social Media Manager, who found a series of quick videos filmed around the office performed much better on their Facebook Page than a more expensive, professionally produced video did.

4. However, the post with the top reach overall wasn’t visual content— it was a link to our blog.

This was our biggest takeaway: Providing a variety of useful content in different forms made the blog posts we did share perform better than if we just shared every blog post we produced mixed in with some third party links and the occasional Instagram photo.

It makes sense for us to keep up this mix of posting and promote these links in the future, so that’s what we’ll test going forward.

5. The best time for us to post is in the evenings.

Seven of our top 10 posts with the highest reach were posted between 5 and 7pm CT. That’s when the majority of our audience is on Facebook, so it makes sense. But it’s still something to keep an eye on in case it shifts seasonally.

Anything else?

We learned a lot more than just these five things – our audience really isn’t into status updates, for example – but they were the big highlights of our month-long experiment. We encourage you to run a similar experiment on your own page, and at the end of the month you should have a good idea of what kind of content performs best with your audience and therefore which posts will be the most worth you promoting.

And if you need the best analytics you can get for Facebook for your experiment, we can help with that.