Facebook’s recent rollout of Promoted Posts and Pages Insights has stirred up a storm of controversy among members of the Facebook community, including the little corner of it occupied by metal bands, their fans, independent labels, and blogs like this one. The development that generated the most intense feelings of betrayal was the revelation that Facebook shares Page posts with only a small fraction of fans who have “liked” the page, and that Facebook will deliver page posts more widely only if you pay them to do it.
Much has been written about these changes — and much of it has been wrong.
Last week I wrote an article on NCS about these changes, and I got some things wrong, too. I spent hours this weekend surfing the web, trying to separate fact from fiction, instead of listening to metal. That was a really piss-poor tradeoff. I’ve learned some things, but even after some fairly intensive research, I still haven’t found the answers to some important questions. That in itself is worrisome, not so much because I’m a mediocre researcher (though I am), but because Facebook hasn’t been entirely open and above-board in what it’s been doing.
But, I do seem to have learned more than a lot of people who’ve unwittingly been spreading misinformation on Facebook band Pages during the last week. So, what I plan to do in this post is separate fact from fiction (expressly citing the sources of my information) and clearly identify the questions that I still haven’t been able to answer definitively.
I’ll also report on the results of our own experiment with Facebook’s new Promoted Posts feature — the one that lets you access more of the people who have already liked your Page by paying for it.
This is a long post, even by NCS standards, so here’s an “Executive Summary”: I’m still fuckin’ pissed off, and you should be, too.
Here’s a somewhat more complete “Executive Summary”:
– On average, a Page’s posts are seen by only 16% of the people who like the Page;
– Facebook uses a computer algorithm called EdgeRank that has an impact on who sees any given Page post, and who doesn’t — and they’ve been using that algorithm for at least a year;
– It’s not clear whether Facebook actually prevents users from seeing all Page posts, though there are good reasons to believe it does;
– Despite what you may have read, there are no free, easy fixes by which a Page administrator can ensure that all fans of the Page who are online will see the Page’s posts in their news feeds;
– Despite what you may have read, there are no easy or effective ways that fans of a Page can ensure they will see all the Page’s posts;
– Facebook users will soon be seeing more Promoted Posts (ones that Pages pay Facebook to distribute more widely and effectively) in their news feeds — and that will make it even less likely they will see un-Promoted posts from the Pages they like.
– The rich will get richer, the poor will get poorer, and everyone will die.
Now for the details:
THE RECENT DEVELOPMENTS
The change that generated the most heat over the last week was the revelation that when the administrator of a Page adds a Page post, Facebook doesn’t automatically share them with everyone who has liked the Page. In fact, Facebook itself reported at a marketing conference in January 2012 that, on average, only 16% of the people who have liked a Page see any given Page post.
Part of the reason is that Facebook uses an algorithm (a computerized formula) to determine which Page posts will appear on any individual’s news feed and the order in which they will appear. The algorithm is called EdgeRank. But it turns out that this isn’t a recent change.
I haven’t yet found any authoritative source which pinpoints the date on which Facebook began using EdgeRank, but it’s clear that Facebook has been using it since at least sometime in 2010. The reason why so many people (including me) are only now discovering this is because of the new Pages Insights feature that Facebook rolled out near the end of May.
This feature includes lots of bells and whistles, but the simplest one is a line at the bottom of each Page post that’s visible to the Page administrator (in the case of NCS, that would be me), but not to fans of the Page or anyone else who sees a Page post.
It shows two statistics: (1) the number of unique people who saw the post — which Facebook calls “people reached”, and (2) the percentage of people who like your Page who saw the post. It also shows the percentage of your “likes” who were reached through “promotion” (more about that later).
Now, for the first time, page administrators like me can figure out quite easily that only a small percentage of our fans are actually seeing what we post on Facebook. Apparently, we’ve been whistling in the wind for more than a year and just didn’t realize it. We can also now tell which kinds of posts seem to be reaching the most people, which I guess is a good thing, though it doesn’t make me feel much better, and you’ll see why.
Looking at all the posts on the NCS Facebook page over the last 7 days, I can see (with a couple of exceptions I’ll come back to) that the percentage of our “likes” who saw the posts ranged from 13% to 34%, with the average being 23.5%.
That was surprising and discouraging. What made it maddening was Facebook’s rollout of the Promoted Posts feature, which Facebook said would enable Pages to reach more of the people who like them by paying Facebook for the privilege. And THAT raised all sorts of questions.
What it implies is that Facebook has had something to do with why the vast majority of a Page’s fans never see most Page posts. For example, if a large percentage of a Page’s fans were missing posts simply because they weren’t online, there would be nothing Facebook could do about that, no matter how much you paid them. But Facebook can certainly fix an attention deficit that Facebook itself helped create.
And that takes me back to EdgeRank. That’s the name of the algorithm that Facebook uses to determine what Page posts appear in a user’s news feed and the order in which they appear. It’s Facebook’s way of determining what’s important and relevant to your interests — as opposed to letting you sort that out for yourself. It’s what’s responsible for a big chunk of that 84% of Page fans never seeing the average Page post.
When you understand EdgeRank, it then becomes obvious that in Facebook’s world, Page posts are in competition with each other for your attention — even if neither the Pages nor you realize it, or think of it as a competition. And EdgeRank is the referee that decides who wins and loses.
Why does Facebook use EdgeRank rather than just letting all posts from Pages you like appear in your news feed in the order in which they’re posted? Because Facebook believes their service will be less meaningful to you if the posts you care about the most get lost in a massive flood of posts appearing in your news feed that you care less about. And if it becomes less meaningful to you, there’s a greater chance you’ll leave Facebook altogether.
Note that Facebook is not content to let you decide for yourself what you want to see and giving you all the tools necessary to prioritize what you see. Why? Because we are sheep and Facebook is the shepherd, and they want our wool. They use EdgeRank to infer what we want to see based on our own individual behavior and the behavior of great masses of other Facebook users. How do they do this?
Facebook is secretive about the precise details of the EdgeRank formula, but conceptually it appears to use three factors: Affinity, Edge Weight, and Recency. In a nutshell, here’s what those factors mean:
“Affinity” is a measure of a user’s interaction with a Page. If you “like” a Page post, if you comment on a Page post, if you click a link in a Page post, if you share the post with your Facebook friends, you are increasing your Affinity score for that page. On the other hand, if you merely read the Page posts, your Affinity score for that Page will be lower, and you’ll see fewer posts from that Page in your news feed.
“Edge Weight” is a factor that decides some types of Page posts are more important to you than others. As a gross generalization, posts that include videos, photos, or links have a higher Edge Weight score than posts that contain nothing but words, particularly brief status updates. But it’s just a generalization, because the Edge Weight score will also depend on each user’s own behavior, based on the types of posts each user tends to click on.
You can slice the Edge Weight factor even more finely. This article, for example, says that these are the types of posts that get the best EdgeRank score:
- Photo with a minimum of 140 characters of descriptive text
- Link with a long description to introduce it
- Status with a long (280+) character body
- Video with a long (280+) character introduction
“Recency” (also known as “time decay”) means that newer Page posts are more likely to appear in a user’s news feed than older posts.
Applying these factors in a computerized formula, EdgeRank prioritizes Page posts in your news feed. In default mode, posts will appear in a user’s news feed based on this ranking, which is called “Top Stories”. If you click on the “Sort” button on your home page, you’ll see that you also have the option of having posts appear in your feed based on which ones are Most Recent. But even here, it appears that Facebook is applying the EdgeRank algorithm to decide the order in which “most recent” posts will appear. [NOTE: I’ve seen conflicting reports about whether Facebook filters posts using EdgeRank when you select the Most Recent setting. However, with the advent of Promoted Posts, it stands to reason that Facebook will filter even the posts that appear as Most Recent in order to deliver the promotion they promise when a Page pays to promote.]
What this means is that if you simply enjoy reading Facebook posts by a particular band, but don’t frequently interact with the posts, and if the bands themselves aren’t including links, videos, or photos in their posts, you’re less likely to see those posts — they’ll be losing out in the competition with other Page posts that get a higher EdgeRank.
DOES FACEBOOK PREVENT YOU FROM SEEING PAGE POSTS?
Does Facebook do anything but prioritize the order in which posts appear in your news feed? Does it go further and actually cut off posts whose EdgeRank scores are lower than those of some number of other posts? In other words, does it actually prevent you from seeing Page posts, even if you scroll all the way through everything that appears in your news feed every day?
Amazingly, after many hours of Googling and reading, I haven’t found a definitive answer to that question. BUT, I have seen a lot of anecdotal evidence to suggest that yes, Facebook does prevent posts from appearing in a user’s news feed, even when the user has “liked” the Page that’s generating the post.
For example, I’ve seen comments from Page admins who, as individuals, have liked their own Pages — and yet do not see all of their Page’s posts. I’ve seen comments from fans of a Page (including the NCS Page) who say they have searched their news feeds carefully and simply do not see certain posts that they can see when they visit the Page directly.
Plus, I haven’t seen any clear statement by Facebook that they DON’T artificially restrict the number of posts that appear in a user’s news feed — and it would be really simple for them to say that, and say it repeatedly, if it were true, given the hostility that recent events have generated.
That takes me back to Promoted Posts. As I explained in last week’s article, Facebook is now soliciting Page admins to promote posts by paying Facebook. Facebook promises that if you pay to promote, your posts will be seen by more of your Page’s fans — and by friends of those fans, even if those friends haven’t “liked” your page. The more you pay, the broader the reach that Facebook will give you.
I haven’t seen a description of exactly how Facebook produces this broader reach, but it stands to reason that Facebook is either adding a factor to the EdgeRank formula — “promotion” — or just arbitrarily increasing the total EdgeRank score that a post would otherwise receive. Either way, Facebook will inflate your post’s EdgeRank score and give it a boost in the competition with other posts. Or at least that’s my educated guess about what happens when you pay the piper.
Paying Facebook to promote a post is still not a guarantee that all of a Page’s fans will see the promoted post. Of course, some fans won’t be online when the post goes live, but even those who are will not all see the Post, because even a Promoted Post is still in competition with other posts in the EdgeRank formula. Maybe there’s some amount of money you can pay that will vanquish all the competition and get you close to a 100% reach. I don’t know.
What’s really annoying about this development to a lot of Page admins is that some of them have already been paying Facebook to promote their Pages through Ads (those things that you now see on the right side of every Facebook Page and profile), ie, paying Facebook to help increase the number of people who “like” their Pages, under the mistaken impression that they would be able to communicate with those people through Page posts.
And now they have to pay Facebook again in order to increase the odds that their communications will actually reach the people who have liked them.
QUICK “FIXES” (THEY WON’T WORK)
So, is there any way for a Page to increase its reach among fans who’ve liked the Page without paying Facebook to promote posts? I’ve seen a bunch of bands add posts to their Pages over the last week asking fans to go to the band’s Page, hover their cursor over the “Liked” button, and by looking at the drop-down menu that will appear, make sure that “Show In News Feed” is checked.
Hate to say it, but that won’t fix the problem — not even close. If you do that, you’ll see that “Show In News Feed” is already checked, because that appears to be the default setting whenever you “like” a Page. Moreover, just because a Page’s posts are selected to be in your news feed doesn’t mean you will see it — because Facebook is still going to apply the EdgeRank algorithm to all the posts from the Pages you like.
I’ve seen another suggested fix, too. It goes like this (I’ve modified this “fix” to make it easier to use):
1. GO TO a page you’ve “liked.”
2. Hover you mouse over the “Liked” button.
3. Once you (finally) get a drop down menu, CLICK + NEW LIST
5. Click PAGES in the menu on the left side of the box that pops up, and you should see a listing of all the Pages you like
6. SELECT all the Pages whose posts you want to see in your news feed (eg, band pages)
7. Give your list a name (eg, BANDS I LIKE)
8. PUSH DONE
I don’t know if this will work, but I doubt that it will. It makes use of an Interests List feature that Facebook rolled out in March 2012. According to Facebook’s announcement of the feature, creating the list will cause “the top stories from each interest” to appear in your news feed. It doesn’t say ALL STORIES. In addition, “top stories” is Facebook’s euphemism for the posts that win the EdgeRank competition.
LABOR-INTENSIVE “FIXES” (THEY WORK BETTER, BUT WILL STILL FAIL)
A Page admin and fans of a Page can also try to alter their behavior to increase the likelihood of fans seeing more of a Page’s posts in their news feeds, ie, doing things that will take advantage of the EdgeRank algorithm and generate higher scores.
Pages can change the nature of their posts to increase their Edge Weight and their Recency. There are web sites that teach you how to do such things (here is one, as an example). Fans can also interact more with the Pages that mean the most to them, instead of just reading posts from that Page when they see them.
But these kinds of behavior-mod “fixes” take extra time and effort, which are generally in short supply in the realm of underground metal — and I’ve seen no evidence that doing such things is going to consistently give a Page’s posts a reach of even 50% of fans or more.
I’m skeptical that there is, or ever will be, ANY workable fix to the problem of Page posts not being seen by the vast majority of a Page’s fans. Why? BECAUSE FACEBOOK WANTS PAGES TO PAY THEM TO PROMOTE PAGE POSTS!! If Page admins and their fans could easily fix the disappearing-posts problem themselves, why would any Pages pay Facebook to do it?
Your average metal band, indie label, and metal blog cannot afford to pay Facebook to promote all their posts. But some of the Pages you like will be doing that. And some of the Pages your Facebook friends like will be doing that — and you will start seeing posts in your news feed from Pages you don’t even like, because part of what Facebook is offering Page owners in the Promoted Post service is the ability to reach friends of the Page’s fans.
What will this mean? It will mean that Promoted Posts will be given priority in your news feed, and if I’m correct in surmising that Facebook puts a limit on the number of posts that will appear in your news feed, that means you will be seeing more Promoted Posts and even fewer posts from the Pages you may care most about.
If you have a profile on Facebook, you will be fed what Facebook thinks is relevant to you and what they are paid to show you. You will not see everything you want to see. You will not be given your own choice about what you want to see.
If you have a Page on Facebook, you will not be able to consistently reach even a third of your fans unless you pay Facebook. In other words, you can’t rely on Facebook as the sole means of promoting yourself and interacting with your fanbase. What was once (supposedly) the great leveling platform that allowed communities to form and communicate for free is gone, or at least in the process of going. Welcome to the new Facebook.
Face it, you are important to Facebook only to the extent you will ultimately help drive ad dollars or other forms of revenue in their direction. That doesn’t make them any different than almost every other for-profit corporation — but it’s a suckass change to be witnessing. In other words, understanding this fact of corporate life doesn’t make it smell any better.
Maybe a revolt among Facebook users can change all this. Maybe Facebook users will leave the herd and move over to greener pastures on some other social media platform. It has happened before (see MySpace). But really, what a pain in the ass that would be, having to start from scratch somewhere else.
Some people have made the argument that no band, label, or metal blog should be relying on Facebook to promote themselves or interact with fans. Instead, so the argument goes, they should be creating their own web pages and also using other forms of social media such as Twitter, Google+, etc., etc.
These arguments make sense, as far as they go. But they don’t change the fact that the biggest social media platform on Earth — one in which a big segment of the metal community has invested a lot of time and energy — is in the process of degrading. Telling us to grow up and get used to it may be good parental advice, but it is not a defense of what Facebook has done.
Because I’m not optimistic that the new, publicly traded Facebook will change its new ways, all I really want now is for them to be open and honest — to be transparent to their users about what they have done and what they will be doing going forward. And then the rest of us can make our choices.
Yeah, like that will happen.
RESULTS OF THE NCS EXPERIMENT
So, when I wrote the first article last week about this recent series of changes at Facebook, I added a post to the NCS Facebook page with a link to the article here at NCS — and I did two more things. First, I paid Facebook $10 to “Promote” that Facebook post. Second, I paid Facebook another $10 to “Advertise” that post.
I haven’t spent nearly as much time trying to understand what happens when you Advertise a post on Facebook as what happens when you Promote the post. I assume my Facebook post showed up in some abbreviated form on the right side of pages for NCS fans and their friends, but I really don’t know.
Facebook will only allow you to Promote a post for 3 days, and I only paid to Advertise the post for 3 days. So the 3 days are over, and what were the results?
As I’m writing this article, Facebook reports that my Facebook post reached 10,968 people and 62% of the people who like the NCS Facebook page.
Of the 10,968 people my Facebook post reached, 929 people saw the post “organically” on the NCS wall (I assume this was the “natural” result of the EdgeRank algorithm), 1,024 people saw it “virally”, i.e., because one of their friends took some action with respect to the post (liking it, commenting on it, sharing it, etc.), and 9,900 people saw the post in an ad.
The most interesting part about these results is that $20 only got the post to 62% of NCS fans, even after 3 days. Of that number, 25% were reached through promotion.
Paying for the Ad clearly caused the post to reach more people than paying for the Promotion. But do I really care about all those extra people the Ad reached? By definition, most of them are people who hadn’t previously shown any interest in NCS, and just based on the increase in “likes” for the NCS page since the Ad started, it appears the vast majority haven’t become NCS fans.
Also, not everyone who saw the FB post clicked the link in the post and came through to the NCS site to read the article. In fact, most of them didn’t. If my Google Analytics statistics are correct, 2,272 different people have viewed the article on NCS (accounting for 2,388 page views) since it went up, which is about 21% of the number of people who saw the Facebook post about the article — and many of the people who viewed the NCS article undoubtedly never saw the Facebook post. However, that’s still much higher than the average number of page views for a post on the NCS site.
I tried one other experiment. I paid Facebook $5 to promote my post about Andy Synn’s recent review of Whitechapel’s new album on the NCS site — but I didn’t buy a Facebook ad for the post. As of this writing — the maximum three days after my promotion started — my Facebook post about Andy’s review has reached 854 people, of which 418 were classified as people who saw the post because it was sponsored.
At a cost of $5, it has reached 49% of the people who like the NCS Facebook page, and Facebook tells me that 25% were reached because the post was promoted. In other words, the reach of the post is significantly greater than it would have been if I hadn’t paid for a promotion — but it’s still less than half of the people who like the NCS Page.
The practical lesson of all this is that it pays to Promote Facebook posts. For bands and labels, paying even $5 is worth it if you have a FB post that’s significant, eg, announcing a new song stream or video or album release or tour.
Okay, I’m done. All that’s left is for me to show you the sources of information I used in writing this article (and for you to express your thoughts about this article in the Comments if you’re so inclined):