There are lots of proposals for “the best time to publish your content on Facebook / Twitter / Instagram / LinkedIn” circulating the social networks. Where do they come from? Can they be trusted? How do you use them in practice? Prof_OG, Over-Graph’s Mr. Data, has the answers.

Everyone who communicates on social networks has asked themselves at least once “When should I publish my content in order to have the best possible impact?”

In response to this question, numerous recommendations, meant for everyone, have appeared on the web over the last few years. But can they be trusted? We asked Professor OG, the man who developed Over-Graph‘s engagement forecasting module, to enlighten us on the subject.

  • The “best moment.” OK, but best according to what criteria?

Prof_OG: “First of all, the best moment for what? You might want to be the “best” in terms of visibility (reach, impressions), social engagement (weak interactions, comments), or traffic (number of clicks). The best might also be “right now” if the information is really urgent.

So, even the definition of the best moment for posting varies according to the brand or entity, but once the objectives are set, you can begin to determine the best time to publish content.”


  • Is there a best time to publish? Could it possibly be the same for all communicators?

Prof_OG: “Several factors have to be taken into account when you are making the calculation, the most important of which is your community. Each community is unique and you have to take that into consideration when deciding the best time to deliver content to it. It’s also necessary to consider the type of publication (photos and links on Facebook are subjected to different filters), the time, the day, and the habits of the publication.

Good content, that is, content that is appreciated by the community, will generate interaction no matter when it is posted. And conversely for bad content. So to suggest the best time to publish, it is important be based in the data and not be biased by “extreme” posts.”


  • What is your advice about the generic recommendations we often see when surfing the net?

Prof_OG: “In my opinion, they are practically useless, because the recommendations are based on moments when engagement or reach peaks were registered over the whole of the social network. It could prove to be a good practice for really big profiles / accounts with more than 1 million fans / followers, because with such a large community they may find themselves there. But even here you need to be sure that all the community is in the same place. So, if it’s recommended that you publish, let’s say, at 6 p.m., which country are you talking about? What happens if 50% of the page’s users are in a different time zone? And also, network connection habits vary according to country, age, etc.

It’s a global trend, we were able to do it with our study on Instagram, but it doesn’t have the force of personalized content. And if everyone took this advice it wouldn’t work anymore because the newsfeeds would be overwhelmed.”

  • What would you recommend for small communities? Will following the generic recommendations fit for them?

Prof_OG: “It’s quite simple. If the community is well-identified, homogeneous, and all have the same busy periods, in this case the recommendations can be followed. If it’s a movable target, that is, the community does not behave in a regular and heterogeneous way, in this case the result will be more unpredictable.”


  • So it seems that the recommendation needs to be personalized. Can you explain how you have integrated this into Over-Graph?

Prof_OG: “We have taken 3 important steps:

1/ The best moment is calculated as a function of the number of engagements on the publications. This means we can have a big volume of data that is trustworthy, controlled, and across all the networks.
2/ The more a user is active in Over-Graph and on his social profiles, the more we can refine our recommendations.
3/ The best times are calculated for each of the social networks > then each of the selected profiles > then possibly for each format that you want to publish in.

This is how the recommendations about redistribution of monitoring work were implemented by our Chrome extension.


It suggests time slots that are very close together, to limit delay, and further apart for the best time to publish in terms of engagement (likes, comments, shares on Facebook; Retweets, favorites and replies on Twitter). Perhaps some of our calculations will surprise you. Some users are recommended to publish on Sunday at 10 p.m., but a publication at that time has a long life (little competition, lots of connection) which means that engagement will normally be very large.”

So, there’s nothing to stop you from using the generic publication hours that you can find on every side. Or, you can trust our engagement forecasting module which will recommend times that are best suited to your community.

Article published by Xavier BK in SocialMedia

the 12 March 2015