Strategic insights
Brands Versus Facebook Versus EdgeRank

In the future, buying reach for posts on Facebook will be part of the game, whether we like it or not.

Several readers have asked me to write about the drop in 'reach' on Facebook. What has happened is that Facebook has turned the dials on EdgeRank to force brands to spend even more on promoted posts for anyone not directly valuable to Facebook.

The way we see this is that reach (number of people who see your post) is down, while 'people talking about' remains the same. Or in other words, the people who engage are the same as always (the valuable group to Facebook), but the listeners (the majority of the people who buy your products) are less likely to see your posts.

Let me illustrate this in a very simple way:

What we don't know (because Facebook decided to no longer provide that data), is how many of the missing people have liked your page. That is, we don't know if the lower reach are people outside our circle of influence, or inside it.

As Mikael Lemberg, Client Partner of Facebook wrote (Danish):

The overall purpose of Facebook's news feed algorithm is to ensure that the content people consume on the frontpage is as interesting and engaging as possible. Put another way: Optimally, we want people to click, like, comment, etc. on all the stories in our news feed.

The mission of the News Feed algorithm is to predict what stories we are most likely to interact with, and then ensure that they are delivered before the stories we just want to consume passively.

I understand why the people of Facebook think this way. Facebook is based on an advertising economy, and as such they make more money the more people engage. Brands, however, are not based on an advertising economy. They are based on a sale economy.

We often see this conflict in the world of publishing. Newspapers and magazines monetized by advertising are focusing on getting people to engage as much as they can with their content. The more you engage, the more views their advertising generates.

On the other hand, newspapers and magazines focusing on selling subscriptions are learning that the snack-sized, quick to consume, highly engaging content, is rarely good for selling subscriptions.

And this is the conflict that we have on Facebook. Facebook is an advertising based economy, where engagement means more than sale. But as a brand, you don't sell engagement, you sell products that people have to pay for. Just like buying a subscription for a magazine.

Think of it like this: Who buys your products? Is it only the people who engage? ...or is it a mix?

If it's only people who engage, or people who are directly influenced by the actions of the engaging people, then it's alright. Because then the people who are being filtered out are deadweight and are skewing your analytics.

But Facebook doesn't know that. Facebook has no idea who of the listeners (the ones who never like, comment, or share a post) never buys your products. In fact, when we look at what has happened in recent years, it seems clear that Facebook knows that they are filtering out a chunk of our valuable listeners.

As Lemberg wrote:

Your active consumers are your most hardcore fans. Those who already spend as much money as possible on your products, defend you in any debate and actively recommend you to friends and family.

Your passive consumers are more varied. It can be anything from your competitors' customers to friends and acquaintances of your hardcore fans. They are the ones you have the highest degree of interest in reaching, for it is they who are the majority - and those moving figures on your bottom line. They are the ones who can open doors to a whole new network of potential customers, as their social graph typically is significantly different than your hardcore fans' graphs, all of which include each other.

Therefore, it is also the passive consumers that make the most sense for you to pay to target.

So... in other words: The people that Facebook is filtering out, are also the ones who are the most interesting to reach, and the ones who make the most sense for you to pay for.

Remember that reach is reportedly down, while 'people talking about' is at the same level. So Facebook is increasingly filtering out the passive consumers.

It's not that simple...

While I really don't like what Facebook is doing, we have to look at it from other directions as well.

The first question is: Who influences the listeners? Facebook would like you to believe that it's the people who engage that cause your message to spread to your listeners. And in many cases that is true.

Think back to the past 10 products that you purchased. How many of them were influenced by a social action? Either being shared to you by a friend, something you noticed on a social channel, via word-of-mouth, emails, or some other action that involves another person. It might also just be a passive social action, like seeing a picture of someone wearing a t-shirt that you wanted to buy as well.

My guess is that these social actions have a huge influence on what you buy ...and why you buy.

But the problem is that Facebook isn't measuring that. They are only measuring a tiny fraction of how we engage socially.

Let me give you a simple example: Imagine that you notice a post from Threadless about a new t-shirt design, like this one:

You go to their site to see what it is about. You like the product so much that you decide to share it, like this:

Now I'm actively engaging with Threadless but ... wait-a-minute ... I did not comment, like, or share the post on their Facebook page. I was influenced by their post, but I engaged with their product instead.

This is incredibly important to remember when discussing social engagement. The theory that social engagement is a significant and extremely important driver of sales ... is true. We now live in the connected world, with so many different ways to communicate with each other.

But what Facebook is measuring is not social engagement. They are measuring the isolated social actions towards specific posts, which represent a tiny fraction of the overall social effect.

For the threadless post, only 30 people have liked the post on their Facebook page. But more than 270 people have liked it on their product page. So in reality, more than 300 people have socially engaged with this product on Facebook.

So the question is: How many of those 270+ people who also engaged with this product are identified as passive consumers by Facebook because they didn't engage with the post itself?

You see the problem here? We are labeling a group of people based on incomplete data. Facebook favors the 30 people who engaged with the post itself, but what about the remaining 270+ people? What about the thousands of people who didn't share it, but were still influenced by it? People like me, who are influenced by Threadless, but don't engage with their Facebook page.

Social engagement is incredibly important, but only when you look at it beyond the activity on a single post on a single channel.

This leads us to the next question. Does FB brand activity correlate with the overall social engagement? ...in other words, if a post is popular on your Facebook page, is it then also popular in relation to the overall Facebook activity for that product?

In my experience the answer is no. Here is a simple example from Threadless. I looked at the latest 6 products they had posted about on their Facebook brand page, and compared the likes, comments and shares to the number of likes on their product page (in the webshop).

Here is the result:

For the truly engaging posts, there is a correlation between the level of engagement of the FB page compared to the engagement outside the FB page. In this case, with high engagement levels for a Pirate Guide t-shirt.

But for everything else, there is no correlation at all.

Look at the Jack-o-full t-shirt. On the Facebook brand page it is the least engaged post of all. But on the product page, more than 580 people have liked it on Facebook - making it the second most popular product overall. And, the second most popular post on Facebook, is one of the least engaging products in their webshop.

Note: I asked Threadless if they could show me their sales data for these 6 products, which could have been really interesting to determine the real value related to Facebook engagement. But they haven't responded yet. I'll update the article if I hear from them.

Another thing we have to ask is if engagement is always linked to a sale? Meaning, are highly engaging posts always good for brands?

I think the easiest way to answer that is to look at YouTube. When I visited YouTube today, and looked at the "most liked" videos, the results were rather depressing.

The top 'most liked' videos were not the Redbull:Felix Baumgartner's amazing jump. Instead it was videos about Justin Bieber and how to troll a gamer.

In terms of engagement these videos are doing great. They have received a ton of views, a ton of comments, shares, embeds, and all kinds of other activity. That means more advertising income for YouTube.

But for brands... not one of these is something worth watching. Not one leads to a sale.

If you are monetized by advertising, like YouTube and Facebook, these videos work because they produce an astonishing amount of social engagement. But if you are monetized by sales, they don't.

So the problems with Facebook's EdgeRank are:

1. They are filtering out passive consumers and lowering brand reach, knowing full well that they are valuable to us. Why else would they say that it makes perfect sense for us to buy them back with promoted posts?
2. The social engagement that Facebook uses for their EdgeRank only looks at a small fraction of the actual social engagement for your products. And there is generally no correlation between what Facebook thinks is important, and what is actually important.
3. Facebook has defined views, likes, comments, and shares as the most valuable elements of all. They can be valuable, but there are many other important forms of engagement... including passive engagement by listeners who are influenced by what you say and do.

It's great if you are monetized by advertising but for brands it is a big problem.

I don't want Facebook to filter out my valuable listeners. I don't want them to lower the EdgeRank for posts that are some of my most important ones, and I don't want to be forced to focus on creating snack-sized content that creates engagement but few sales.

I want you to influence people. I want you to build a lasting connection. I want people to have withdrawal symptoms when you haven't posted for a while. I want you to have true fans, who not only buy your products, but also listen to everything you do.

Take the Dalai Lama. He has 4.5 million fans on Facebook of which only 2% engage with his posts. Does that mean that he is not important to the other 98%? Of course not. People don't have to like, comment or share a post to feel inspired or connected to him as a person ...or to you as a brand.

If he wrote a new book, would only 2% buy it? Of course not! He would probably sell to most of his followers. That is, unless EdgeRank has filtered him out from their streams because Facebook doesn't understand the power of an inspired listener.

What you need to do

Facebook is not going to change. As a social platform monetized by advertising, they will always favor engagement and activity. That's their business model.

Facebook says it made EdgeRank to increase the relevance for you and me. They want the content you see to make you feel so excited that you want to engage with it. That's not a bad idea, but as you see above, that means that Facebook is moving further away from the listener.

It's easy to understand why Facebook is doing this. It's in their interest. It's based on the fundamental principles of their business model, and it's truly what they believe social to be about.

For brands, we have to look at things from a bigger perspective. We need to realize that engagement is great, but it's only a part of what we need.

Think of it like this: Imagine if the only songs you could play in Spotify were from bands that you actively engaged with. That is, unless you commented, liked, or shared songs from an artist, their music would disappear from your playlists.

That's what Facebook EdgeRank is doing. They are trying to manage who you can connect with socially, based on your activity.

This means that we need to stop thinking about Facebook as a platform. That's not really what it is any longer. Facebook is more like a search engine. You decide what you search for, but Facebook decides (based on their algorithms) what posts you see at the top... and what posts you will never see.

This means that just as people think about SEO, we need start thinking about FEO... Facebook Engine Optimization.

As brands, we are no longer in control. People can no longer decide what's most important to them (not easily, at least) and as a brand we cannot assume that we can connect with those who like our pages.

I don't like this at all. I think it's terrible, the same way I dislike SEO. When you, as a brand, create a blog, I want you to focus on creating content that connects with and inspires your followers. But because of the way search engines work, we also think about SEO. Often that leads to a conflict in which the content is optimized more for search engines than for your readers.

Here is one example: How to Use Pinterest to Improve Your Writing ...that's a blog post designed for SEO. It contains no value for the reader.

And because of EdgeRank we are now seeing the same thing on Facebook. Here is one example from Samsung:

This is not optimized for people, this is optimized for the news feed. It's designed to encourage a quick form of social snacking that creates engagement.

C.C. Chapman tweeted: "More likes = higher FB engagement score so their future updates show up more. I'm not a fan of the tactic but it works."

Facebook posts like these are the same as the Pinterest blog post example above. It's SEO ... or more specifically FEO.

But here is the thing. There are two ways to rank high on search engines. One is to use all the SEO tricks that you know. The other is to create something so valuable and interesting that people share it without you needing to trick them into doing it. Both methods work...

It's the same now with Facebook optimization. You can use all the engagement tricks out there, like Samsung is doing, and it works. That Samsung post you see above has 115,000+ likes, 2,600+ comments and 2,000+ shares.

That's a staggering amount of exposure for something that is essentially an image campaign that took 5 seconds to make.

But there is also another way. That way is to be passionately awesome. One example is Harrison Krix over at Volpin Props. He never uses any of the FB engagement tactics that we hear about all the time. Instead he just talks about how he makes props for games and other things. His Facebook page has an extremely high EdgeRank score, simply because of his very open, authentic and passionate work.

Promoted posts?

What about promoted posts? Clearly Facebook has been tweaking EdgeRank to force brands to spend money on them (both in terms of reaching passive consumers, but also to get back the reach we have lost).

Should you do promoted posts?

Well, I think that in the future this would be integrated into any Facebook strategy.

I don't like the concept that I have to pay Facebook to get back my own fans, people who have decided to follow me as a brand. If people decide to follow a brand, how dare Facebook filter or reduce our interaction with them?

But this is the reality of the new Facebook. Promoted works across platforms, including mobile (unlike display advertising), so promoted and sponsored posts are the future for Facebook as more and more people move to mobile.

We may not like that, but as long as Facebook is monetized by advertising, this is the future for them.

I think EdgeRank will become more filtered in the future. I think we will see even more instream monetization options and in the years ahead, buying reach for posts will be part of the game.

So my best advice is to start testing it. Figure out which posts it makes sense to promote. Is it those which create engagement (and possibly optimize the EdgeRank for future posts?) ... or is it the posts that lead to a sale, the ones where you talk about a new product that people can buy today?

I don't have a universal answer, because it varies greatly from one brand to another. I think for a company like Ford, the engaging posts are far more important to promote because the decision to buy a car is something that happens over time.

If you are a fashion brand however, it might be relevant to promote the products themselves, because there is a level of impulse buying taking place when we buy new clothes.

Promoted posts are cheap. It's one of the cheapest forms of advertising we have. But like everything else on Facebook, the result is not based on how much you spend, but rather on how engaging the posts end up being.

Facebook is no longer a platform. It's a media channel. Think of it as such.

Head over to my Google+ post to comment and discuss this article.