Posted by randfish
A concept we’ve covered regularly is what we call flywheel marketing, where the organic traffic, shares, and links you get from publishing one piece of content makes it easier for later pieces to see some success. One of the key pieces of that flywheel is the ability to get those social shares, and based on a recent study, we’re ready to admit it: We were completely wrong about that key piece.
In today’s Whiteboard Friday, Rand explains why, and that the real value may lie in engagement.
Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we’re talking about an assumption that I think many of us have made over the years. I know I have. In fact, I’ve amplified that. I might have even covered it on Whiteboard Friday. Thanks to some research that we’ve done together with BuzzSumo, as well as some research we’ve seen from our correlation study this summer, you know what? It’s looking like we were just dead wrong on this very important aspect of how SEO and social media and content marketing fit together.
You’ve probably seen me present on this either here on Whiteboard Friday or in one of my slide decks or in a blog post. It’s this idea of flywheel marketing, where you create some great content, you amplify that content via social media and your social channels, you attract visitors through that, you naturally earn links from some of those people who visit your site, and you grow your social following. Now, the next time your audience potential is bigger and your rankings potential is also bigger, because you have more links coming to your site, and that helps all the other pages on your site. You have a bigger social audience, so now there are more people to amplify to.
You know what? It actually looks like this is totally broken and wrong. The idea that you are naturally earning links from people who come via social looks to us like it was a bunk belief in its entirety. Let me show you.
First off, BuzzSumo did the vast majority of the work. I appreciate them including Moz as well. We did participate in some of our link metrics. The BuzzSumo crew did a bunch of this work. They looked at articles that received social shares, in fact a million articles that were taken from their database, and then they looked at the number of shares and the number of links those received.
The vast, vast majority received zero links. In fact, 75% plus of all articles they looked at received zero, not a single one, social shares. Same with links, by the way. I think it was 90% plus for links or maybe even more.
This is a like a power-law distribution. You’re essentially seeing that a few articles get all the shares out there. Everything else really gets nothing. If you’re not going to be in the top 10% of content that’s created, don’t even bother. You’re not going to get shares. You’re not going to get links. You’re not going to get traffic. Forget it. A lot of content marketing is probably spent in vain. Granted, maybe a lot of that is learning what actually works and experimenting, and that’s fine.
Then they looked at the correlation between links and shares.
As you can see from this crudely drawn scatter plot, no correlation whatsoever. If you were to draw the line here, it would probably be something like, “Oh look at that total crap correlation.” Here are the numbers. Facebook, 0.0221. Twitter, 0.0281. Ooh, slightly better, but still in the realm of totally insignificant. Google+ 0.0058. You’re just talking about numbers that suggest essentially that there is virtually no correlation between links and shares.
Now they did look at places where there were lots of shares and links, and those tended to be a few things. I’ll let you read the report, and you should. I think it’s one of the most important reports to come out in our industry in a while. Credit to BuzzSumo for putting it together.
We know from our research. We’ve done experiments looking at whether anchor text still moves things. We’ve done experiments looking at whether URL mentions move the needle. URL mentions don’t, by the way. Once you turn them into live links, they do. We’ve looked at whether you can actually rank content without any links at all. It turns out almost impossible, so next to impossible that we couldn’t find a single credible example of a page that ranked without any links unless it was on a site that had lots of links pointing to it.
We know we still need links to rank.
In fact, notably ranking correlations with links haven’t dropped over the last few years. Even though we all feel like the algorithm’s getting a little less link centric, and I think it is, links are still clearly very, very powerful. So we have to worry about things like outreach and link focused content and embeds and tools and badges and competitive link analysis and all the other many link building methods that the marketing industry has come up with over the years.
I have a theory about why this is.
I think Google is honest when they tell us, “We don’t look at social shares to determine rankings.” I think what Google sees is something Chartbeat showed a few years ago. This was another excellent study that I encourage you to check out. Chartbeat basically analyzed engagement on socially shared content. What they saw was a plot that looks like this. Very, very few social articles have high read time. Even the ones that have lots of social sharing have very little read time.
It turns out a ton of things that people share socially on the Web, they don’t read at all. They may click Retweet. They may even include the URL. They might share it on Facebook. But they, themselves, may never have even visited that content. Sounds crazy, but I bet you’ve done it. I bet I’ve done it. I bet I’ve been like well, you know, it was probably a good edition of Whiteboard Friday, I’ll go share it out, having not yet watched the video and seen whether I did a good job or not. That’s just the way of the Web.
I think Google cares much more about the engagement than they do about the social share counts themselves.
So you can see lots of things with social shares not performing well. But once they start to get engagement and start to earn links from that engagement, now they’re suddenly ranking.
Hopefully, with this knowledge in mind, you can go back to the drawing board a little bit if you’ve built up, like we have, this mental model of how the flywheel works. Look, I’m not saying that this works for no one. This actually works pretty well for Moz. It works pretty well for us in this industry, but I think, and clearly the data is showing, that across the vast majority of the Web it’s statistically extremely unlikely this will work for you or for everyone else.
I think we need to revisit this. We probably need to revisit our link building. We need to think about social in a different context of how and whether it’s earning people who will actually come to our site and want to link to us and people who will come to our site and want to engage, or whether it’s just a vanity metric.
All right, everyone, I look forward to your comments. We’ll see you again next week for another edition of Whiteboard Friday. Take care.
Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don’t have time to hunt down but want to read!