How to Improve Your Facebook Engagement Using Controlled Experiments

social media how to Are you wondering what type of Facebook post works best?

Are you struggling with getting engagement or reach on your Facebook posts?

You may have heard that you need to post more photos or more questions or more photos with questions…

But what really works best for your audience?

In this article, you’ll learn how to find out which types of Facebook posts work best for your page.

Why Experiments?

Running controlled experiments from time to time will help you discover what works on your page so you won’t have to take someone elses word for it.

It’s also a great way to help you come up with a posting structure that may help you streamline your content creation in the future.

#1: View Insights for Ideas

In your Insights section, you do have some data on your best post types (under Posts), but that can be a function of what types of content you have been posting lately. So you may not get the whole picture if you haven’t been posting many links recently. But it’s a good place to start your investigation.

best post types

Take a look at the best post types for your page.

Another good place to start is to look at the Posts section and click on one of the posts. You will get a lot of information on your individual posts. The example shown here is a photo tip, which typically does very well on my page.

post level data

Clicking on a post will give you a lot of information.

What I wanted to see is how posts did in a more controlled experiment. If I posted similar information with a link or a text post, would it do as well? Here’s an idea of how you can set up an experiment on your own page to do a little investigation.

#2: Set Up Your Experiment

Start with an outline of types of posts that you want to test and length of time. I ran two different time periods to test the number of posts per day. The longer your testing period, the better your results. But for the purpose of this post, I chose to run my test this way:

Length: Posting 5 times per day for 3 days

Types of Posts: Text, link, photo

Post times and formats:

  • 6 am: Informative post with a link and a tag of another Facebook page (text post would remove the link preview, photo post would have the link in the status area)
  • 9 am: Tip (link post would include a link to the tip, photo post would include a photographic representation of the tip)
  • Noon: Longer post (more text, but also would include a photo or link if needed)
  • 3 pm: Question
  • 6 pm: Humorous post

For the second experiment I posted 3 times a day with slightly later times: 10 am, 1 pm and 7 pm.

If you’re going to be a purist about this testing, you would post the same information each time and vary the post type each time. But that might not be so fun for your fans.

Once you’ve done your posts, you should wait a day to allow for all the results and then you can start assembling your data.

First, download your post-level data in your Facebook Insights area. Click Export Data, then select your date range and make sure you’ve selected the Post-level data button.

export post level

Download your post-level data.

Now you’re going to need to gather the data on four different tabs of the Excel spreadsheet.

excel tabs

Use the different tabs on the Excel spreadsheet to see the data.

#3: Gather the Data

I wanted to know about comments, likes and shares, as well as the clicks, reach and hides for each post. I found that information on each of these tabs within the spreadsheet:

  • Lifetime Talking About This: Comments, likes and shares
  • Key Metrics: Reach
  • Lifetime Post Consumers by Type: Clicks
  • Lifetime Negative Feedback: Hides and unlikes

One thing to note on the Lifetime Post Consumers by Type is that you’ll see Link Clicks, Other Clicks and Photo Views. The Other Clicks incorporate when someone clicks anywhere on the post that isn’t the photo or the link—they may click on the headline, the text in the status area or on a See More link if you have a long post.

And a note on the Lifetime Negative Feedback: You’ll have a column for Unlikes of Your Page and the “xbutton_clicks”, which is the Hide button. I take the sum of these columns since they have a similar effect—the person is not consuming your content in the news feed anymore.

Now put these statistics into one spreadsheet so you can compare.

statistics

Assemble the data into one spreadsheet.

Sum up the days of data so that you can see what type of post “won” the day. Highlight the winners for each day. Obviously in terms of reach, you may be seeing the same people for each post, but it still equates to potential eyeballs on your posts.

#4: Draw Some Conclusions

From this data, you may see patterns. One thing that’s obvious is that the status updates are the best at comments, likes, reach and other clicks for the two posting schedules. So I need to keep that in mind when doing my posting. But I did see that the link posts did OK in the reach for their posting schedules. I’m not sure how the link posts got photo views in the second round of testing so that was unusual.

The other thing you can do is compare the time slots between the posts and see if a certain type of post does better in one time slot. I noticed that the midday photo posts seemed to perform well compared to the other times during the day that I posted photos. After running this experiment, I might make sure my midday post has a photo.

Final Thoughts

By doing more controlled experiments, you can also find out what works best for your page. I would recommend using more data and trying different types of posts—maybe add a video post or test some different tagging strategies.

In any case, it was a fun exercise and I felt like I learned a few new things. I also confirmed my thoughts about the status updates getting a little bit more engagement and reach.

What do you think? What types of experiments have you tried running on your page? How did they turn out? Share your thoughts with us in the comments below!

Tags: , , , , , , , , , ,

About the Author, Andrea Vahl

Andrea Vahl is co-author of Facebook Marketing All-in-One for Dummies and also uses her improv comedy skills to blog as Grandma Mary, Social Media Edutainer. Connect with her at www.andreavahl.com. Other posts by »




More Info
  • Theodore T

    What a great reminder. It’s easy to get into a routine and just stick with it, even when great information is available for improvement. Thanks.

  • http://www.ftlcollective.com/ Miguel Bravo

    I do this using Facebook Ads. 2 or 3 dark posts with effectively the same message but different status formats, targeted to fans only.

    Most recently, I was able to determine for a certain bigger client of mine that sharing recipe pages on his site as “photo + link in status” got more LINK clicks than link posts. Really surprised me.

  • Chris Picanzo

    Thanks for the valuable info Andrea, it’s always interesting to see the stats and sometimes even puzzling to what works better. I have to say It’s a struggle for me with two particular clients both being consignment shops and they post pics regularly which their audience does like but I feel there could be more engagement on both pages but worry about posting too much as it would require additional posts on top of their already frequent pictures of items. Just had to throw that out there lol One is furniture and they post their own pics all at once maybe twice a week and the other is vintage fashion who posts only a couple times a day which is fine. I’d love to get more interaction from the furniture page but it’s a tough audience :)

  • http://www.AndreaVahl.com/ Andrea Vahl

    Thanks Theodore! Yes, I definitely think it helps to mix things up so you can truly make sure you are on the right track!

  • http://www.AndreaVahl.com/ Andrea Vahl

    That is very interesting to know Mig! Glad to hear other’s experiences with tests.

  • AnnMullen

    How do you get post details like that? I tried clicking all over the place and never got it. Help please.

  • http://sproutsocial.com/features/social-media-engagement Sarah @ Sprout Social

    This is really great advice, Andrea. I think experiments like this should be done frequently and consistently. While Facebook may be changing their algorithm every so often, what should remain (relatively) consistent is your community. Every one is different and determining a post type that will provide value to them is what a brand should strive for. The experiment of analyzing post type success could be replicated across any social channel too!

  • Maurisico

    As a testpurist i can’t allow these kinds of testing :-) Just kidding, is a good way to know your dialogues with your customers. But when you want to do clean testing, it is my opinion that you use timeline ads to test. You can test timeline ads with different formats in different ads.

  • AmandahBlackwell

    Great post!

    I’ve been experimenting with different posting times and information. For example, I started posting more videos. But it’s been wonky. Why? Because I like to schedule posts ahead of time. When I upload a video and schedule it for another day, it back posts to the day I uploaded to Facebook. Huh? I don’t understand this. The good news is that fans like the videos.

  • DonnaGilliland

    Good information Andrea. You always write great posts.

  • http://www.EntrepreneurOnFire.com/ John Lee Dumas

    Andrea, these steps are perfect for finding out what the best posts for your fan page
    are. Asking a question works really well, but we continue to test
    everything to see what works best as you’ve outlined above. Thanks for
    this!

  • faigie

    is it worth it to try if I only have 250 fans so far?

  • http://increaseyoursocialreach.com/ Sofia

    On your FB page go to insights and you’ve got all the data there. I believe insights only works after a certain number of likes.

  • Joe

    My Facebook page was reviewed 1/5 star by my (friend) competitor. I am not very sure how to see list of the page review and how to answer it. Any idea how to avoid such thing?

  • http://navneetsau.com/ NavNeet Sau

    Brilliant post Andrea. Facebook insights is really an awesome invention for business owners and marketers to help them about what works perfect for Facebook marketing. Thank you for letting us know about this deep knowledge.

  • http://www.AndreaVahl.com/ Andrea Vahl

    Yes, some audiences are a little tougher than others to get interaction from. I would definitely make sure you are varying the types of posts (text vs pictures) and do some experiments. But I know it can be challenging! Thanks for your thoughts!

  • http://www.AndreaVahl.com/ Andrea Vahl

    If you go to the Posts section of Insights, click on the link for the individual post. You do have to have over 30 fans to see the Insights as Sofia said. Hope that helps!

  • http://www.AndreaVahl.com/ Andrea Vahl

    Great point Sarah! Yes, these experiments are just for Facebook :)

  • http://www.AndreaVahl.com/ Andrea Vahl

    Yes, I know this is a very “unscientific” test since there isn’t enough data here but it’s a good start :) I definitely agree with testing ads as well – great way to see what works. Thanks!

  • http://www.AndreaVahl.com/ Andrea Vahl

    I’ve heard from many people who are having trouble with the scheduling feature lately – frustrating! I haven’t tried scheduling video posts so I’m not sure if it’s a problem with the recent changes or if it’s always been wonky for videos. Good to know!

  • http://www.AndreaVahl.com/ Andrea Vahl

    Awww thanks Donna!

  • http://www.AndreaVahl.com/ Andrea Vahl

    Thanks John!

  • http://www.AndreaVahl.com/ Andrea Vahl

    I think it is – sometimes pages with fewer fans get better engagement overall so it’s worth experimenting. Good luck!

  • http://www.AndreaVahl.com/ Andrea Vahl

    If you switch your business category anything other than a Local Business it will remove the Review. That is frustrating if you want to keep that classification. The only other solution is to actively seek better reviews to balance that out. There is no way to prevent a certain person from reviewing your page if the review option is open. Hope that helps!

  • http://www.AndreaVahl.com/ Andrea Vahl

    Glad you enjoyed it!

  • http://www.bloggingfromparadise.com/ Ryan Biddulph

    Hi Andrea,

    Sensational advice. Just by digging into metrics and experimenting for a few days you can see what works and what does not.

    As for me, my travel pictures do really well. So I post ‘em ;)

    Thanks!

    Ryan

  • Samantha Dickson

    Great post, Andrea. An essential part in using social media, especially platforms like Facebook, is remembering to utilize the metric tools available. Many companies/organizations optimize their social media presence through continuous interaction, but then forget to benchmark this against their metrics to see what’s working, and perhaps what is not. By completing these controlled experiments, you’re truly analyzing the patterns of your audiences. This is a great way to monitor what your users engage with the most, and to create content strategies that play off of these insights. Thanks for sharing!

  • Dan Linstroth

    This is great. Thank you for the tips. I am curious about the optimal frequency for Facebook posting. 1,3, 5 times per day? Using your advice, I could post with each level of frequency over a five day period and then compare engagement metrics. Week 1: post just once a day. Week 2: post three times per day. Week 3: Post five times per day. Then compare. Does that make sense? Any advice? Thank you.

  • Elizabeth Hall

    I have experimented a bit. I also noticed that status updates did better overall on my page. I really like the detail you went into on this tutorial. I’m definitely going to try to do more specific testing using some of your tips. Thanks so much for the info.

  • http://intesols.com.au/ Moin Shaikh

    You can hover over to your page’s insight data by clicking View Insight button at very top in your admin panel area.

    To be very simple: Go to this url (MAKE SURE YOU REPLACE ‘YOUR_PAGE_USERNAME’ WITH YOUR PAGE’S USERNAME) https://www.facebook.com/Your_Page_Username?sk=page_insights

  • Pingback: 6 Action-Inspiring Ways to End Your Blog Post (and 12 Examples) - Zen Optimise

  • Chet

    :) simple? it doesn’t even make any sense. Replace your page username with your page’s username?

  • http://intesols.com.au/ Moin Shaikh

    I meant replace the USERNAME of your page in the url i posted above. For example if the page username is smexaminer the your insight url would be https://www.facebook.com/SMExaminer?sk=page_insights

    That’s simple, i guess.

  • Pingback: E2M Eagle’s Eye - 29 Nov 2013 | E2M Blog

  • kumea

    I have learnt so much from reading your articles. Question, what if you are selling a product, i have noticed that once i do a post with the product to get it sold, the likes and shares ate very minimal. Any tips or suggestions for sales?

  • Dennis Fischman

    Andrea, thanks for the information. I wonder if this is one of those cases where the act of observing changes what’s being observed. If you notice that something is working–a type of post, or a time of day– and then you go heavy on that, might people get tired of it? Then you would have to test again to catch the change in what works. Lather, rinse, repeat?

  • Pingback: Content Marketing Tips for 2014 and Beyond | DragonSearch Digital Marketing Services & Consultants

  • David Gitonga

    Thanks for the tips. I also noticed that status updates on my page get the most engagement, followed by photo, and finally the link. I will be sure to use your method to experiment.







Check out the Social Media Marketing Podcast!
Check out our Blogging workshop!
Join our Social Media Marketing Networking Club
Download the free Social Media Marketing Industry Report
Check out our sister publication: My Kids' Adventures

Get Your FREE Copy of the 2014 Social Media Marketing Industry Report

Wondering how your peers are using social media? Get this free report (50 pages, 80+ charts) and never miss another great article from Social Media Examiner.