Reset password:

Strategic insights
Beyond Facebook Analytics, The Guide

When you look at Facebook Insights, you need to find that story that created the numbers. You need to know why something happened.



Written by on November 8, 2011

Shared By Plus Subscriber

Thomas Baekdal

This is Baekdal Plus content. It is shared with you for free by a member. Please reshare it.

A couple a weeks ago, Facebook came out with a new "Insights" dashboard. It was a big improvement from what we had before, mostly because the metrics are all based on people instead of interactions.

Why is this important? Well, the old system might show that you had 200 comments within the past week. But it didn't tell you anything about how many *people* there are behind the numbers. You didn't know if it was 200 people who each posted one comment, or 10 people who each posted 20 comments.

But with the new analytics, you do know. It will no longer tell that you had 200 comments, instead it will tell you how many *unique people* interacted with you. That is an extremely important change. It is the same with all the other metrics in the new system: reach, likes, engagement, talked about (sharing), etc. They are now all based on unique people, and far more useful.

But I will not show you how the new Facebook Insights work. There are already thousands of articles about that around the web. What I will do is take you beyond the numbers because all analytics are useless unless you apply them to a specific goal or activity.

Note: If you want to know what each element is about, I suggest you take a look at the video from John Haydon below, or check out this excellent PDF guide from Facebook.

Facebook Insights, beyond the numbers

The truth about all types of analytics is that the real story is somewhere between the lines. The numbers you see are just the result of that story. When you look at Facebook Insights, you need to find that story that created the numbers. You need to know why something happened.

Let give you an example. Take a look at the graph below.

What you see here is a page where the owner posted something on October 8 that caused quite a significant drop in engagement.

So what you need to know is "why didn't people react to that?" What was it that went wrong? You see the result in Facebook Insights, but it doesn't tell you the story of why it happened.

To learn that you need to go back to your wall and compare the wall post from October 8 with the posts before it. And in this case the reason was due to the fact that it was a different type of product - an unfinished one. But in your case, the reason could be many other things.

Then you see the effect of the Facebook Edgerank. The owner of the page stopped updating for a while, and the effect of that is that after about a week, traffic dropped significantly. In this case, to almost zero on October 20.

This is why social media professionals keep telling you to post continually.

But now we see the next interesting story. Because on October 21, the page owner posted a new update and suddenly the activity was back to "normal".

What this indicates is not that the October 21 post was anything special (it wasn't) It just indicates that if you go away for a while, you don't lose your audience. You lose the activity and the daily exposure after about 7 days of inactivity, but not your audience.

This is a very important thing to know about social media. In traditional media, lack of exposure also means lack of audience. But in social media lack of exposure only means lack of exposure - your audience is still following you.

Let me give you another example of that. I run a Twitter profile called @baekdal24hours. It's a little something special I do every year during the 24 Hours of Le Mans. I don't have that many followers and it's nothing special, but what is special is that I only update it during one weekend, once every year.

Traditional thinking dictated that this was nuts. You cannot stop updating for a full year and then expect to come back. But social channels don't link audience with activity.

My Le Mans profile is on standby, and my 333 followers are just patiently waiting for me to bring it back to life in 2012. I'm not getting any exposure or effect from it, but I'm also not losing my audience.

If you tried to do this in print or even a blog, it would be a disaster. After a year, people would have completely forgotten about you, and your audience would have dropped to zero. Since I started this stream back in 2009, my number of followers has actually increased ever year. In the social world, you have to decide to unfollow someone.

In the traditional world there is a saying that all exposure is good exposure, because without exposure you lose your audience. But in social media, bad exposure means people will "unlike" your page. In the social world, it is better to do nothing than to do something that people do not like.

Now, there is a catch. Unlike Twitter, where you can just go away for as long as you like, Facebook does actually punish you for lack of activity. It takes about 7 days for you to lose your exposure (as you can see above), but it probably takes a month (maybe even more) before you start to lose your audience. People will still like your page, but your posts might not show up in people's streams.

The real problem with Facebook's ranking system is lack of engagement. Meaning that it is worse to post something that people do not react to, than to post nothing at all.

If we look at the graph again, we see that there is one more story to find. On October 24, the owner posted an update that caused the stats to explode. Again, to find out why you need go back to the wall, and figure what post made this difference.

What was it that caused this to happen? What is the story behind it?

Let me show another graph from another brand, about something I see fairly frequent from brands that have been using Facebook for a while. If you look at the graph below, what you see is a brand with a regular publishing schedule and a completely flat level of weekly reach.

Each post is not really making any difference one way or another. But what you can also see is that the number of people who are talking about this brand is dropping.

This is an indication that you are boring. This brand is likely doing the same few things over and over again, and people are getting bored with it. It's not that you are not interesting " the weekly reach is steady. People are just not talking about it as much as they used to.

You are slowly turning yourself into a commodity. It is just something people can follow every day, but you are not motivating your audience to act. You are not changing anything.

If this is the kind of graph you have, then it's not really useful to look at each post to see which one performed better than others, because they are all just part of the same. You need to do something new. You need to take your audience one step forward.

We see the same thing if we go into the details. Under "reach" and "talking about this" you can find these three graphs.

And in this case, what you can see that it is the same story (obviously). You are still reaching your audience, they are not going away, they are still interested in what you do. In fact, it is entirely likely that the people like to follow you and get your daily updates. But your fans are no longer talking about you.

But because you are doing the same thing, again and again, people have nothing new and exciting to talk about. Or maybe it is because you are not involving them. Are you asking your fans what they like, or are just talking about what you like?

What you want is graph like this one. You want something that moves - something that changes.

With movement you can see when something worked and when something didn't - and you can learn from it. It's the learning process that makes you smarter. If everything is just the same, you never grow and you never learn why. You are just stuck.

If you find that your graph is flat, you need to experiment. Even a failure is better than the status quo (but don't be stupid).

Analyzing the individual posts

Let's move on to a more detailed analysis of each individual post. In Facebook Insights you can find this list of posts vs. how they performed.

They will show how many people each update reached, how many engaged with it (liked or commented), talked about it (shared), and the percentage of virality.

You are not going to get much insight from looking at a list like this. Sure you can change the sort order and look at posts with a high engagement versus those with a low one, but you need more than that.

One problem is that "reach" is pretty much irrelevant. Your reach will match a certain percentage of the number of likes you page has. It is usually around 20-30%. For popular Facebook pages that number might be higher, for less popular pages it might be less.

The only time that reach can really change is if a post goes viral, but the result is not caused by your "reach" it is caused by the number of people who talked about it. So the magic lies in looking at your engaged users and how many talked about each post.

You need to put your posts into a graph like the one below. On the X-axis you have the number of engaged users, and on the Y-axis you have how many are talking about it (sharing it).

Now you can learn something from it. Now you can find the story behind the result.

A post with no engagement and no "talked about" is not working. Your fans are not reacting to it. It is not resonating with your audience.

It's important to note that, when I say it's "not working" I mean in a social context. None of this tells you anything about you conversion rate. Your conversion rate is how many people who see posts, and then decide to visit your store and buy it.

That's your conversion rate. A passive person (no engagement, no talked about) can still be an active buyer. In fact, I have seen several examples of passive users being responsible for a higher share of the actual conversion than the few "engaged" ones.

So measure your conversion rate. Measure what effect your post has in relation to where people buy your products. This is not something you can do with Facebook Insights.

But a post with no engagement and no sharing tells you that it is not working in a social context.

On the other hand, a post that people are engaging with and are talking about (sharing) is great. That's what you want. Do more of that.

Now comes the tricky part, the two other boxes in the graph. This is where you can learn a lot from your audience.

One example: If you post about a "special discount" then having people comment or like the post is always nice, but it would be much nicer if they would "talk about it" - as in share it with their friends. If they do both, then that's great. But "talked about" is very important when you try to sell something. That is what makes things go viral and spreads the word to other people.

A high "talked about" score is less about you, and more spreading what you do between your fans and their friends.

Let me put this into perspective by doing the opposite. A post about a "special discount" with a high level of engaged users, but a very low "talked about" (sharing) level, might just be a waste of money and time. You are not expanding your audience. Your fans are not sharing with their friends. They are not talking about it. They are just commenting on it. You are not growing.

And since the only people you really reach are your core audience, they probably would have bought the product at the full price anyway.

It's like telling an Apple fan boy that he can get the new Macbook air 20% off. Sure he will jump on that, but he would have bought it at the full price anyway. You are just losing money.

On the other hand, if you are posting about things that you do - behind the scenes stuff, personal posts, fun posts etc, the number of engaged users are more important than "talked about".

Again, if people do both, then that's great. But engagement tells you something about how strong your community is and how loyal your fans are towards you.

You have a high engagement for posts about "what you do or why you do something" and a high talked about level when you talk about your products.

Post about your products

If you are selling more than one product, there is one vital thing you need to do every month: Rank your products for your product team.

You need to go into Facebook Insights, take out all the post about your product and create a small report highlighting which products people liked, and which they didn't. Here is a small example:

Information like this is of absolute critical importance for brands. It is not enough just to post things on Facebook and have a "conversation" with your fans. If you want to be successful you need to bring that information back into your organization.

Facebook, and the other social channels, tell you something about what your customers think about your products in real time. Learn from that!

It takes only 10 minutes to do this once per month. But the value of it is incredible ...especially if you learn from it over time. And it is not just about the product. It is also about how it is being presented. Maybe you will find that when you try to be fancy, people like it less than when you just show it "for real".

Looking at the momentum

Everyone who is working with Facebook knows that the real power is in the flow of posts, not with each individual update. It's like rain. Each drop makes no difference, but when it all comes down at once, you get soaked!

You need to look at the momentum you build. A while back I wrote an article about how to build a social media release plan. It is about how to build up momentum before a launch, and turn it into sales afterwards. It looks like this:

What you want is to use the first couple of months to build up momentum. You need to attract new people and you need to get them to engage with you. The point is to make people feel like they are a part of something.

But then when you launch your product, your graphs should explode. If it doesn't explode, then you have a problem. Maybe people didn't think it was that exciting after all.

But if it explodes, the most important element is "talked about" - how many people shared your product. Engagement and likes are great, but it only tells you something about what your existing fans think about your product.

Talked about is how much they share. It's the raw reaction from people telling their friends "Get this new album, it's freaking awesome!!"

Talked about is what helps you grow. It's what makes you go viral. And the number one success criteria when you launch a product is to see how many people share it.

Engagement is less important. Because if you have done it right, people will already be happy about it before you launch.

Real fans, vs. fake fans

The final thing I want you to look at is your graph illustrating likes and unlikes.

The first interesting part is just to look where people are liking your Facebook page from. Are they liking it because they found it in the newsfeed (because people talked about it), or are they liking it by clicking on a like box or button on your website or blog.

That in itself can give a lot of insight. You would probably be surprised how few people actually like a page because of a like button on a website. Those buttons are great but they are rarely the main reason for new likes. Sharing is!

But the interesting thing to look at is when you see a spike in either new likes or unlikes. What is causing it? What is the story behind it?

For instance, when you see a huge spike in "likes" after posting about a special discount "just for fans", and then a week later a huge spike in "unlikes" when you post about something else, you know that those "new fans" are really fakes fans.

They didn't come to your site because they liked you. They just wanted to exploit your special offer.

You want your "new likes" to correspond to the activities that also have a high level of engaged users and sharing. And you want people to unlike you when you are not creating value. Not the other way around.

The true story is between numbers!

I hope this guide helped you get started using Facebook Insights to your advantage. And remember to share it. As a Plus reader, you can share every article with your friends for free. You friends will have full access, they don't need to signup or register, and you will even be credited for the share.

Shared By Plus Subscriber

Thomas Baekdal

This is Baekdal Plus content. It is shared with you for free by a member. Please reshare it.

Share on

Thomas Baekdal

Thomas Baekdal

Founder of Baekdal, author, writer, strategic consultant, and new media advocate.


Check out my book: THE SHIFT - from print to digital and beyond? Free for Baekdal Plus subscribers, $8.79 on Amazon.

There is always more...