Spotted: Is Facebook Page Reach Back Up?

Has Facebook Post Reach Doubled

If you’ve been following along, by now you know that I put no stock in Facebook’s reported Reach metric. I call it “imaginary” because it can’t be proven and isn’t representative of substantive results.

Yet, during the past couple of months, it’s been the source of a major uproar. Everyone’s freaking out that a tweak to Facebook EdgeRank is resulting in a drop in Reach (engagement metrics be damned!).

So know that I really don’t care about this metric. But I realize that 99% of Facebook marketers do. So when I spotted a trend recently that Reach may actually be on the rise — not only for me, but in general — I was curious what types of responses it would generate.

Let’s take a look at what I’m seeing…

From the Timeline

Reach is most visible to admins from a Page Timeline. It’s a metric that is impossible to ignore.

From September through much of November, my typical post without promotion would reach about 1,600 people. Sometimes more, sometimes less.

Here was a very common sight in mid-November:

Facebook Page Reach

So, needless to say I was a bit surprised when I started seeing this at the end of November and into December:

Facebook Page Reach

Getting a Reach of 4,000, 5,000 or even 6,000 was suddenly quite common — all without promotion.

From PageLever

First, I wanted to show you my data from my Insights Exports. Unfortunately, Exports have been absolute garbage for me lately. I get errors when I try to download them, and even when I do get them I’ll have missing data. While I did get data that supported what I was seeing within my Timeline, there was too much missing data to use it.

So my next step was PageLever.

I pulled Post Details data from September 1 through December 14. I exported into Excel and removed any post that was promoted and grouped data by week. I then found the median (rather than the average) so that my results wouldn’t be skewed by outliers.

Here is my median Total Reach by week, starting September 1:

Facebook Page Reach

This supports everything I was finding within my Timeline. Up until the end of November, the weekly median Reach was always under 2,000 for posts that weren’t promoted. But for the first two weeks in December, the median Reach was 5,308 and 4,623.

My Reach rose more than three times from the middle of November.

Fan Growth

The first question someone may ask is whether my audience grew significantly during that time. It has not.

Here is my growth chart, thanks to PageLever:

Growth Rate Chart PageLever

As you can see, nothing noticeable — particularly in November and December. And to more than triple Reach based on Fan growth, I’d need to see triple the Fans in a month. That obviously didn’t happen.


So the next thought is this: SWEET! I must be getting some awesome engagement that is increasing Reach virally!

Nope. My engagement is actually a little pedestrian lately (can only blame myself). Here are charts for Stories, Engaged Users and Consumptions of my posts (here is an Insights Glossary, if you need it), all from PageLever. Exact same process of measurement as before, promotion removed.

Facebook Page Stories Engaged Users Consumptions

Well, that’s disappointing. I’d like to be able to say that I increased Reach based on my awesome content that was irresistible to users. But engagement clearly didn’t help boost Reach.

From AgoraPulse Barometer: It’s Not Just Me

So my next thought was that what I’m seeing may be an anomaly. I’m always slow to publish my results because I represent a very small sample size. What I want to know is whether other brands are seeing this.

Luckily, there’s the Barometer by AgoraPulse. This is a new tool and the sample size still isn’t huge, but about 750 Pages have been benchmarked. This includes Pages with audiences below 1,000 and more than 100,000 Fans. So it would give us some good insights.

Unfortunately, the Barometer measures Organic Reach instead of Total Reach, so our data isn’t going to be 100% consistent. But take a look at this chart that compares my Organic Reach to that of the other 750 Pages benchmarked on Barometer:

Facebook Organic Reach AgoraPulse Barometer

I zoomed in a bit because it’s otherwise difficult to read. Just know that there was virtually no change through December to the right of this graph.

Two things here:

  1. WOW! The average Page’s Organic Reach doubled starting November 24!
  2. Wait… My Organic Reach is unchanged!

I actually saw some of these same numbers in PageLever when I viewed Organic Reach more generally (instead of post-by-post). It doesn’t make a whole lot of sense to me.

On one hand, I’m seeing my Reach triple when looking at it on a post-by-post basis. This is supported within my Timeline, in my Export (when it works) and by PageLever.

On the other, this chart shows no change in my Organic Reach. Meanwhile, the typical Page is seeing the type of bump I was seeing with Total Reach, and at about the same point in time.

The only thing I can think of is that maybe I’m still reaching the same number of people month-over-month, but that Facebook is doing a better job of surfacing content to the people who matter on each individual post.

Maybe. But sometimes, the stats don’t make a lot of sense. Either way, I think there is enough here to say… well, there’s something strange going on.

What it All Means… Maybe

This is what I know:

  • My Total Reach tripled starting at the end of November
  • Many Pages saw a doubling of Organic Reach starting November 24

If you love Reach, that sounds pretty awesome. But as you know, I really don’t care about Reach.

If you trusted Reach, you’d expect a resultant bump in engagement. I have not seen that. While it’s dropped a little for me, it’s not consistent or long enough to be a trend.

Here are a few theories…

1. My Reach is up because I have a very low percentage of Negative Feedback.
According to Barometer, my Negative Feedback is far below the average of the other 750 benchmarked Pages. Facebook previously said that they adjusted EdgeRank to punish brands that were spammy. This could reflect that.

That sounds like a good explanation. I’d buy it if I was also seeing a resultant bump in engagement.

2. Facebook has changed the way that Reach is reported.
This is one of the reasons I cautioned you about relying on Reach. Facebook could say you reached 1,000 or 10,000 people, and you’d have no way of knowing for sure which was true. They could completely change the way they reported that number as well.

Since I’m not seeing any other changes in data, I would consider this to be quite possible.

3. It’s a Facebook bug.
Also possible. As I mentioned earlier, I’ve had serious issues accessing my Exports lately. This could be a sign that something is wrong under the hood. And if that’s the case, all bets are off with the data.

4. Facebook has increased the shelf life of a post
In addition to having the data to support that my Reach is up on a post-by-post basis, I also have a less scientific observation. It seems that I’m Reaching about the same number of people within the first couple of hours. But that number is increasing far more than it used to beyond that initial two-hour point.

It once was suggested that a post was essentially dead after two hours. I’m not seeing that right now, particularly if you rely on the totaling of Reach. This could be a sign that Facebook is doing a better job of bringing my content to the top for people who care most about it.

What Are You Seeing?

I’d love to hear what you’re seeing. Go through your Timeline. Check your Exports (if you can). Use PageLever and Barometer.

Is your Reach up? Let me know the results you’re seeing in the comments below!

  • Marie Page

    I’m with you on number 4. My data seems to suggest a long tailed effect whereby after the initial boost of people seeing the posts in the first few hours the numbers carry on creeping up in subsequent days. That may be more exaggerated on my page (http://www.facebook.,com/musicademy) because the average age of our fans is about 45 and I assume older fans frequent FB less regularly (and when they do they see content that is quite old) and the less friends they have in relation to younger people (which again would mean that content stays on the timeline longer.

    I can see arguments for your other 3 hunches too.

    Love your work here btw. So good to see someone analysing their insights data in public and communicating so clearly about it. We’re not all so confident with numbers and it’s very helpful to see your analysis and explanations.

    • Jon Loomer

      Interesting, Marie! It definitely seems odd. I won’t notice anything different the day I post. Then I’ll look back a couple of days later and see a big number. Tells me those posts are sticking around for some reason. Could be a benefit of EdgeRank, I guess.

      And thanks so much for the kind words! The more people who share this kind of data, the clearer it will become for everyone.

      • Marie Page

        Oh and I forgot to say. I’m seeing almost exactly the same curve hike as you for Reach on around 23 November onwards.

        • Jon Loomer

          Good to know, Marie! I enjoy hearing that actual, breathing people are seeing these things, too, instead of just accepting the stats!

  • Vincent Vizachero

    Are you SURE you are excluding posts that are not promoted? I’ve found that if you monkey with the Sponsored Stories post (by deleting it, as you described earlier) that Facebook reports incorrect organic reach numbers in Insights (not just in the API, delivered to PageLever and Agorapulse) but in raw downloads.

    I saw the same spike in Organic reach on Agorapulse, but it was all due to one post in late November that I promoted with a deleted Sponsored Story.

    • Jon Loomer

      Yes, I’m sure, Vincent. I haven’t performed that delete in quite some time. I actually didn’t do any promotion during the time of the spike, so it wasn’t difficult to confirm.

      Regarding your spike with AgoraPulse, that would be highly coincidental if you and the average page of now 850 pages all have the same spike for the same reason of deleting a Promoted Post on one post at the same time!

      • Vincent Vizachero

        I figured (and felt bad about suggesting it).

        However, I don’t see any change in reach for our pages. Curious.

        I’m wondering if the data are showing two coincidences?

        1) Some pages are improving reach by using more status updates (which EdgeRank is now favoring, apparently) even though these don’t seem to help engagement.

        2) Some pages in Agorapulse got a big lift in reach on Black Friday and it is skewing their barometer.

        • Marie Page

          Interesting point Vincent. As I stated, we are seeing the same increase as Jon but I have increased the number of status updates over photos.

  • Pingback: Marketing Day: December 20, 2012()

  • Alex Hollins

    I’ve noticed that even when viewing “most recent”, posts that I’ve commented in or that my friends have commented in get pushed back to the top after what seems to be a certain number of comments.

    • Jon Loomer

      Alex — Yeah, I’ve definitely noticed that in general the “Most Recent” feed is having less and less value lately. It’s simply not “most recent” at all. I’ve found there’s virtually no difference between the two feeds, so I would not be surprised if Facebook soon gets rid of it. This is unfortunate since I once felt that Most Recent had significant value as a user.

      • Alex Hollins

        a bit late, but I for one would love a most recent that has a highlight around posts I haven’t seen yet, or a, most recent unseen, button.

  • Brian D. Meeks

    I love data. I don’t really follow how well my blog is doing, or care that much, but I like reading about how people measure their blogs. I figure, I may want to start doing so, one day. I used to analyze data for GEICO, where a 15 minute call could save you 15% on your auto insurance. I appreciate a solid analysis that tries to dig below the surface.

    I don’t really have anything interesting to add. I just wanted to say I enjoyed the post.

    • Jon Loomer

      Thanks, Brian! I try to provide as much data-related info as I can. It’s something that is lacking from most Facebook related analysis. Glad another stat-head appreciates it!