How to Win on Facebook: 8 Lessons Learned From Analyzing 1 Billion Posts

If you want to know what works when it comes to marketing, you need to study the competition.

This includes the tough competition, but the weak competition as well.

By doing this, you can determine why the best succeed, and why the others fail in their efforts.

To do so effectively, you need to look at one particular area of interest.

The BuzzSumo team decided to compile as much data as possible on marketing on Facebook, and then complete an in-depth analysis.

Overall, they analyzed over 1 billion Facebook posts and came up with findings that will change the way you market on Facebook. 

They were kind enough to share their data with me, and allow me to walk you through it.

I’ll break down the findings, and let you know exactly how you should use them.

But first…

Why Facebook? Why not any one of the hundreds of other social networks?

While it would be worth analyzing other networks as well, Facebook is the place to start due to its size.

As of the start of 2016, Facebook has 1.59 billion active users per month. That’s over 20% of the global population.

This means that customers for just about every imaginable business are on Facebook.

Finally, no other social site drives even close to as much referral traffic as Facebook. It’s a unique platform because users are more than willing to click on links to visit content on outside websites (like yours).

Okay, so we have the right network, and a ton of data.

Let’s look at the findings.

1. The best time to post isn’t when you think it is

When you make posts on Facebook, they are shown to some users that have already “liked” or followed your page.

The exact number (or percentage) will depend on your organic reach.

Your organic reach is determined by a bunch of different factors. The most important ones are the amount of engagement you get on your posts, as well as how much competition there is from other posts.

Competition is a crucial factor. If there are tons of posts made from friends and other pages a user has liked made around the same time, Facebook can only show so many of them. The more competition, the less likely yours will be shown.

Intuitively, it makes sense to post when the most users are online, which is during the day. That way, a greater percentage of your followers could potentially be exposed to your posts.

But the BuzzSumo analysis actually found the complete opposite.

The red line in the following graph represents your competition; the amount of other posts being made around the same time.

The blue line shows you the engagement that posts receive at different times in the day. Engagement includes commenting, sharing, and liking posts.

The pattern is extremely clear.

Posts get the most engagement when the total amount of posts is at its lowest.

Conversely, there’s too much competition during peak times (during the day), which leads to low engagement.

The conclusion from this data is to post between 9 and 11 PM in the timezone that most of your followers lives in.

BuzzSumo mainly looked at U.S. pages, but you can apply the findings for any timezone.

Why? Because they tested it for another country – France.

They found that posts made between 11 PM and midnight had the highest average engagement (240.06 shares). Likewise, posts made from 10 AM until noon performed the worst.

Almost identical trends.

2. Certain types of posts perform over 5 times better than others

There are 6 main types of posts you can make on your page as a business:

  1. Questions
  2. Image posts
  3. Videos
  4. Links
  5. Giveaways
  6. Coupons/Discounts

As part of the analysis, BuzzSumo looked at the performance of each type. Note that interactions means the same thing as engagement (total number of likes, shares, comments).

There’s one type of content that isn’t on the graph that actually performs better than all of them.

No, I didn’t lie to you before, there are only 6 main types. But there are a few different subtypes of each of the types.

Most importantly, the BuzzSumo team dug in different types of question posts, and found something huge.

Question posts combined with images resulted in an average of 616.70 interactions. However, questions posts that were only text only received 144.45 interactions (terrible).

Finally, the analysis didn’t just look at the total number of engagements. It also included the breakdown by each type of engagement (comments, likes, and shares).

There’s a few interesting points of notes on that graph:

  • videos get the highest number of average shares (83.87)
  • giveaways get the highest number of average comments (33.91)

While the high number of comments is interesting, keep in mind that comments are usually required to enter giveaways, and are almost always low quality.

A more practical analysis would need to look at the average length of comments for each post type (maybe an idea for a future research project for BuzzSumo).

Overall, we can make the conclusion that images and videos perform best when it comes to overall interaction, while videos are best if you’re looking for shares in particular.

Additionally, combining questions with images (or videos) will yield the best results; do so whenever possible.

3. Sunday is a special day, especially for businesses on Facebook

Not all your posts will be as important as others.

You’re going to want to make sure that important posts (like links to your website or product announcements) are as optimized as possible, so that as many people see them as possible.

We’ve already determined a time range when you should post, but we haven’t looked at the best day.

The analysis revealed that posts on weekdays all receive nearly the same amount of interactions (on average).

However, posts on the weekend receive a significant amount more:

Sunday is the best day to post, with an average of 401.75 interactions, followed by Saturday at 365.30 interactions.

The research didn’t reveal the reasons why this happens, but you can take your own guess. Perhaps people use Facebook more on weekends, or businesses don’t post as often (less competition).

Regardless, you now know that posts on Saturday and Sunday will get more engagement than the weekdays. Save your important posts for the weekend.

4. The optimal length of articles for Facebook users is…

If you’re using Facebook as a business, you’re going to be linking to your content (you better be).

And as I mentioned at the beginning, most Facebook users are happy to consume content on other websites.

However, it turns out they have a preference for this content.

BuzzSumo found a strong correlation between the performance of a Facebook post and the length of the article that is being linked to.

As you can see, when a post linked to a short article (of under 1,000 words), it received the fewest number of interactions (171.65).

Posts that included 1,000-2,000 word articles performed the best (277.37 interactions), followed closely by 2,000-3,000 word articles (274.06).

The most interesting finding to me is that posts with articles of over 3,000 words performed about 18% worse than the other in-depth articles (225.02 interactions).

Since I use Facebook heavily for Quick Sprout, and most of my posts fall into this final category, I might not be getting the most out of Facebook.

This post (that you’re reading) falls under 3,000 words, do you like it better?

The bottom line is that Facebook users enjoy in-depth posts, but they also don’t want to spend hours reading one. Keep your posts between 1,000-3,000 words when possible.

5. Users aren’t on Facebook to read

There are 2 main components to a Facebook post.

There’s the description of the post (the blurb you type), as well as any link or media you attach.

It turns out that the shorter your description, the better the post will do.

It’s clear that posts with 0-50 characters (that’s about 0-10 words usually) get the most interactions by far (411.16).

The number of interactions a post get go down as the number of characters in a post increase.

The simple takeaway is to make your descriptions of posts as short as possible. Leave any detailed explanations in the content that you link to instead.

6. Whatever you do, don’t post YouTube videos

As we saw earlier, videos get a good amount of engagement on Facebook (and the most shares).

However, there’s a certain type of video post that far exceeds the others.

There are 2 popular options:

  • Embed YouTube video:
  • Directly upload/embed a video on Facebook:

You wouldn’t think there’d be a big difference, but the results from BuzzSumo’s analysis showed otherwise:

When it comes to the number of interactions that video posts get, embedding a YouTube video only gets an average of 140.75 interactions, a fraction of the 951.48 interactions that direct embed videos get.

You might suspect that there’s not a sufficient sample size of directly uploaded videos on Facebook, but BuzzSumo accounted for that.

For this particular segment of the analysis, they analyzed over 53 million YouTube video posts, but also over 25 million direct embed posts. While that’s not as many, it’s a great sample size.

The bottom line is that if you make video posts on Facebook, take the time to upload the source video onto Facebook itself.

When you’re making a post, click the “upload photos/video” tab at the top of the text area, then choose the file from your computer. Then fill out the options like adding the title.

7. Facebook users like Instagram

Facebook integrates nicely with other social networks.

If you use Instagram as well, you can check the Facebook option while posting a photo and it will also be posted to Facebook.

It looks like a normal Facebook picture post, with a small difference of saying (“from Instagram”).

You wouldn’t think this would make a difference, but the analysis proved otherwise.

Pictures that are posted on Facebook through Instagram received 23% more interactions than images uploaded directly on Facebook.

I honestly don’t know why this happens (theories welcome in comments), but the data is clear.

If you’re posting pictures on Facebook, and also use Instagram, use the post to Facebook option that Instagram offers.

8. The effect of hashtags probably isn’t what you expected

The final big area of the massive analysis looked at the effect of including hashtags in posts.

You know what hashtags are, right? Any tag that is preceded by a “#” is a hashtag (e.g. #QuickSproutisgreat, #GamesofThrones).

In the past, hashtags almost always improve the amount of engagement that you get on most networks.

Well, the research says otherwise on Facebook:

You can’t argue with data: Posts without hashtags received 34% more interactions than posts with hashtags.

Unless you have a specific reason to include a hashtag, leave them off  your Facebook posts.

Don’t apply these findings until you read this

There’s one key thing that you need to remember to apply these findings effectively.

Understand that these findings are correlations. They look at the average effect of different variables.

What this means is that the findings are best practices.

If your audience behaves significantly different than the average audience, your optimal Facebook posts may look different.

These findings are great starting points, but they may or may not be right for you. Start by implementing them, but then test other options as well to confirm if they are the best or not.

If you need help to do that, read through my guide to optimization.

Conclusion

Let me quickly summarize the 8 best practices that came from BuzzSumo’s research:

  1. The best time to post is between 10 PM and Midnight of your audience’s local time
  2. Posts with questions and images get the most engagement
  3. The best day to post is on Sunday
  4. Posts that link to articles between 1,000-3,000 words get the most interaction
  5. Short post descriptions get by far the most interaction
  6. Directly embedded videos get over 6x the interaction of YouTube videos
  7. Images posted through Instagram get a 23% boost in engagement
  8. Posts with hashtags get less interaction than posts without hashtags

These are all backed by an insane amount of data, and are great best practices to follow for your Facebook marketing.

If you have any questions about the findings or research, leave me a comment as usual below.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>