Author Archive: Sara Miller

The Bigger the Facebook Page, the Harder it is to Reach Fans

Have you ever wondered if the size of your Facebook page’s fan base plays a part in the distribution of posts from your page? According to a recent analysis done using Facebook post data, the rate of distribution on a post decreases as the size page’s fan base increases.

Generally, which fans are reached by the average post depends on the Facebook sorting algorithm Edgerank, which takes into account not only the content preferences of your fans, but also the affinity between your fan base and your page.  Whenever your page posts, the message automatically shows up in the newsfeed of a subset of the fan base, the impressions garnered from this is called ‘organic reach’.

In order determine what the average percentage of a page’s fan base is reached organically, over 40K posts across multiple verticals from Q4 2012 were analyzed.  The sample data was then grouped by fan base size, and the average organic reach per post was compared to the total fan base size.

Facebook has previously reported that a post reaches anywhere between 15 – 20% of a pages total fan base organically. Independent studies put the average organic reach much lower, at anywhere between 3 – 7.5%.  The average percentage of fan base reached in the sample analyzed here was 10%, but the most interesting information popped up when the data was separated into fan base brackets.

Screen Shot 2013-04-01 at 2.52.32 PM
As a page’s fan base goes from under 500K to over 1 million, the percentage of the fan base reached organically gets cut nearly in half! And pages with over 10MM fans only reach on average 4% of their fans, that’s a third of the organic reach of pages with less than 500K.

We’re not suggesting you should stop trying to grow your page’s fan base, as 4% of a page with 10MM fans is still a much larger audience than 12% of a page with 500K. However, the moral of the story is: be mindful of the quality of fans your page is acquiring, as there does seem to be a reach cost as your fan base size grows.

Optimize Your Tweets

Retweets are arguably the most powerful metric on Twitter. Compelling your followers to Retweet (RT) your brand’s message is essential if you want to make full use of the platform. So, how do you increase these shares?

In this blog post, we examine how the average RT/Tweet rate varies based on daily Tweet frequency, post-time, and post-type. Using data from about 7,000 brand Tweets* we looked at how frequency, time of day and post-type effect the average RT rate (%) of a Tweet.

*Methods: Brand Tweets were sampled from ~30 accounts associated with 13 big brands; Twitter accounts used included regional and global audiences; all data is less than 6 months old (11/5/12). RT rates were normalized against the total number of followers on the day of the Tweet.

 Tweet Frequency: 4-5 Tweets/Day

How many times a day should brands be Tweeting? Is it better to under-Tweet rather than over-Tweet, or vice-versa? In order to address these questions, we categorized Tweets based on the total number of Tweets made by the brand in that day (Tweets/Day) and then plotted the average RT rate (%) to see if there was a trend, and indeed there was!

Tweeting four times per day is the optimal frequency in terms of highest RT rate. Tweeting once per day returns the lowest amount of ReTweets. The average RT rate on accounts that Tweeted 5-6 times a day were about 60% higher than those that only Tweeted 2-3 times a day, so erring on the side of over-tweeting is better than under-tweeting.

**Data not shown

The trend seen here complements a recent socialmediatoday post, which also found the optimal tweet frequency peaks at ~4-5 tweets per day.

Tweet Time: Between 7-8am for your target follower base. If you do not know where your followers are, posting at 4, 7, and 11am EST is a good jumping off point.

When is the best time of day to Tweet? Using the same data set as above, we looked at the average RT rate based on the time of day the tweet was posted (EST). The results are shown below:

The data above is plotted as Eastern Standard Time (EST). The largest peaks occur between the hours of 7 and 8 am in three heavily populated time-zones: London, England, and the East and West Coasts of the US.

The largest peak occurred at 7am EST, and was 75-85% higher than either of the other two peak times.




This is likely due to a combination of factors such as:

  • A larger proportion of the follower base is in this geographic region
  • This time slot benefits from being between the other two peak times
  • The followers in this area are more engaged

Based on these results, it is advisable to coordinate your post time with the geographic region of your target audience. Depending on your location, simply posting between 7-8am in your time zone may put your Tweet in one of the ‘lull’ periods between peak morning hours, so be aware of the time change between your area and EST.

Tweet-Type (Link vs. non-link)

Will adding a link to you Tweet significantly affect the number of Re-tweets it receives? Probably not…according to the data there’s almost no difference between the two types in terms of Re-Tweets.

Tweets that contained links got only 3% more RTs on average than just regular ‘Status Update’ Tweets. While adding a link may benefit the site the link targets, it’s will probably be detrimental to the share rate of the Tweet. Users typically perform only 1 action on a post, and if a link is present it competes with the RT functionality for engagement. So use links carefully in your Tweets, as they may be costing you RTs.

Main Take-aways:

Many factors will determine the success of your brand’s Tweets. If changes to content don’t increase the average RTs per Tweet, consider varying other factors such as Tweet frequency, post time (plus location of followers), and post type. Below is a summary of the recommendations for optimizing your brand’s Tweets:

  • Tweet at least 4 times a day! The average RT rate more than doubled when the Tweet frequency increased from 2-3 Tweets per day, to 4!
  • If you are targeting Tweets to a specific region (outside of your timezone), adjust your post time so the Tweet goes out between 7-8am in the area of your audience.
    • If you’re not targeting by region, Tweet at 4, 7, and 11am EST. This will put your posts at the top of users feeds in the most densely populated regions around the world.
  • Use links with caution! Keep in mind that most users will only perform 1 action on a post. Adding a link will likely decrease the number of RTs your Tweet receives.

Facebook adds New Metrics that Distinguishes Fan from Non-Fans

Fan or non-fan? That is the question. Luckily, we are beginning to uncover an answer.

Facebook recently added some new stats to Facebook Insights that differentiate user impressions and actions by page fans or non-fans. This is especially exciting for brand marketers and social data nerds like myself, because it gives us further insight into the Facebook algorithm “Edgerank”, and allows us to measure how much of a pages total fan base is reached and engaged when a brand page makes a post.

The first thing we look at is how the average audience of a brand post is split between fans and non-fans. So, who’s seeing the page’s content – fans or non-fans? Next, we’ll look at how that ratio changes when the post is selected for a Facebook ad or sponsored story. I looked at five representative Facebook brand pages, each page’s industry and approximate fan base size is in Table 1 below. The sample data was taken from March 31 – August 15 2012, and each brand had at least 100 posts during this time period.

Below is the audience breakdown of the 5 representative Facebook brand pages.
All of the posts used in Chart 1 got only organic and viral views and did not become sponsored stories or ads. As you can see, over 90% of the average post reach in all 5 pages was from fans of the page.


This tells us that when your brand makes a post, it’s seen almost exclusively by users who are already fans of your page. In order to reach non-fans in any sort of quantity, you must use Facebook ads or sponsored stories. The graph below shows the average audience breakdown in posts that were used in ads or sponsored stories after appearing in the newsfeed.

Notice how the audience split for fans vs. non-fans changes substantially when the post is used in ads or sponsored stories in brands 1, 2, and 5; the average percentage of the post audience that are already fans of the page goes from being ~90% to under 40%. Brands 3 and 4 had a significantly smaller proportion of paid views relative to their total audience, and therefore had a higher proportion of fan views per post in their total audience.

This is interesting in that if you want to reach consumers that are not already part of your fan base, in any kind of quantity, paid advertising is necessary. For all the brands included in this study, on average less than 10% of the users who saw the post in the newsfeed were non-fans. Posts used in ads or sponsored stories with the greatest reach also tended to have the lowest percentage of page fans in their total audience. This finding is also interesting from a content creation perspective. Posts that are to be used in ads or sponsored stories should have content that would engage a wider audience, while posts that are expected to only appear in the newsfeed should be targeted more at fans.

So now that we know the breakdown of who sees post content, lets look at average engagement levels. Do fans that see your pages post engage with it at a higher rate than users that are not yet fans of the page? How is engagement effected depending on whether the impression is organic/viral, or paid?

The chart below shows the average engagement rate per post in page fans vs, non-fans. The data in Table 2 is the average post engagement rate for all posts (whether or not they were used in ads or sponsored stories). The “% Change” shows the increase in engagement of non-fans over page fans.

Four of the five brands in this sample had non-fan engagement rates that were at least twice as high as average fan engagement. Table 3 breaks out the sample from Table 2 above between posts that were only seen in the newsfeed or on the page (Non-Ad Posts) and those used in ads or sponsored stories.

Once we separate out the sample between posts used in advertising and those that were not, we see that the engagement rate for non-fans is much higher only when the post is not used in advertising or sponsored stories. It has long been known that content endorsed by a friend on Facebook is more likely to garner engagement. The comparison in Table 3 backs up this idea, as non-fans engaged with brand content more when it came from a friend (as opposed to being presented as an advertisement). Even with all the changes to the platform, it seems as though Facebook users are still much better at targeting their friends than advertisers. This reminds marketers that in order engage the greatest number of people, making sharable content is key.