If you’re in our industry, you’ll doubtless have spent the past couple years trying to avoid an endless barrage of gloomy predictions concerning Facebook’s ‘meaningful interactions’ algorithm. More specifically, the ways it favours surfacing almost everything except publisher content in users’ timelines.
And yet, paradoxically, a recent Digiday survey indicates that with no alternatives of similar stature, 70% of publishers contend that Facebook still delivers the best content reach of all platforms.
Even for unpaid content circulated organically, 45% of publishers agreed that Facebook was still top dog.
So we sat down with our analyst Barney Perkins to work out how specifically to use the platform to your advantage.
Buffer, the social media management platform, have found that posting five times a day maximises post engagement. Read more about that here. However, we have decided to focus our investigation on the metrics most easily monetised – reach and sessions referred to the website.
Since the algorithm changes, Facebook’s official suggestion is that post rate should not have an effect on reach. This is because the way content is allocated to a user’s news feed depends on relevance.
“Post frequently – Don’t worry about over-posting. The goal of News Feed is to show each person the most relevant story so not all of your posts are guaranteed to show in their Feeds.” – Facebook
Even if this is true, post rate might still influence the visibility of a publication’s content by diluting its average relevance and/or increasing the competition in users’ News Feeds – from the publication and from competitors.
The below analysis looks at the daily total reach (taken from Facebook Insights) and number of posts of a publication’s Facebook page (from Social Insider). It also displays the number of sessions referred to the publication’s website (from Google Analytics). Though there is some ambiguity about whether reach or sessions are more valuable metrics, both are given equal preference.
By conducting a simple linear regression of both reach and sessions on the number of daily posts, we found that there was indeed a negative correlation between post rate and reach, as well as a positive correlation between post rate and the number of sessions.
Model 1: Reach
Model 2: Sessions
By aggregating the above graphs, we found that the optimal number of Facebook posts per day was 11. But, make some considerations here. Firstly, the model specification is not a perfect fit of the data. In other words, it’s unlikely that had the publication started posting 40 times a day, there would be <0 reach, indicated by the red line:
One model we experimented with suggested an optimal post rate of 16 posts per day but behaved very strangely in the extremes. As total followers increased the optimal number of posts would drop down to <1 with an absurdly high expected reach.
In conclusion, the model we’ve focussed on here suggests that publications should aim to post to Facebook 11 times per day, with an exception for content that might have a considerably broad appeal. Perhaps questions of optimum post rate are also questions of what content is most relevant to the highest number of the publication’s followers.