As a regular blogger on SEOmoz, I’m very interested in what drives traffic to our posts. Of course, there’s the usual realm of referrers and keywords, but lately I’ve been curious about how social signals (including Google’s new +1) correlate with traffic. In other words, how much more traffic will a post get because it gets more Tweets, Likes, or +1s?
So, I set out to do an informal correlation study, looking at how Tweets, Likes, +1, and our own internal social metrics – Thumbs Up and Thumbs Down – impact Unique Pageviews (UPVs) over two sets of data. The first set is the Top 50 posts (by UPVs) for the first half of 2011. The second set is all main-blog posts after the launch of Google+.
(1) Top 50 Posts of 2011
The first study was pretty straightforward. I looked at the traffic for all main-blog posts (including promoted YOUmoz posts) from January 1st to June 30th of 2011 and pulled the Top 50 by Unique Pageviews. For each post, I gathered data on Thumbs Up, Thumbs Down, Tweets, Likes and +1s, and calculated their correlations with UPVs. The graph below shows the correlations:
Just a quick refresher – the correlation coefficient (r) varies from -1 to 1, with 0 indicating no relationship and 1 being a perfect positive correlation (when one variable goes up, the other variable goes up). Correlation does not imply causation, but I’ll get into the details of that below, because it’s very interesting for this data set. On a technical note, these are Spearman correlations – the social signal data isn’t normally distributed. All values with asterisks (*) were statistically significant (p<0.01). Finally, I’d like to give a shout-out to our resident stats guru, Dr. Matt Peters, for working through the math with me.
We wouldn’t normally expect one signal to drive traffic, but thumbs up from the community and Google +1s had a solid impact. Twitter’s relationship with Unique Pageviews seemed surprisingly low, and thumbs down didn’t seem to encourage or discourage views, but neither of those measures were statistically reliable (p>0.10).
(2) All Posts Since Google+
The +1 data in the first study is surprising, since Google+ didn’t launch until June and the button wasn’t implemented for most of the first half of the year. Many of these +1s arrived well after the original posts were published.
So, I ran a second study, using only blog posts published between June 18th (the launch of Google+) and August 15th. This amounted to 44 posts, not too different a sample from the first study. Although the +1 button rolled out prior to Google+, I felt the roll-out date was a good cutoff, since that’s when people really took notice.
Here are the Spearman correlations for the second study:
With the exception of Thumbs Up, every signal’s relationship with Unique Pageviews increased in the second study (and all correlations were statistically significant). It’s likely that social factors are more powerful for the recent past, and some of the posts in the first study are a couple of years old (even though the traffic stats are for this year).
Facebook Likes came out on top in this study, and Google +1s weren’t far behind. Given the kind of data we’re working with, a correlation of 0.83 is impressive. Tweets were roughly as strong as Thumbs Up in predicting traffic levels.
Did the Signals Cause Traffic?
Here’s where things get interesting. As statisticians like to say (and we frequently repeat), correlation does not imply causation. Let’s not just nod our heads and pretend we know what that means, though – let’s explore exactly what it could mean for this data set. A strong correlation between Facebook Likes and Unique Pageviews could mean any of the following:
- Facebook Likes could be driving Unique Pageviews
- Pageviews could be driving Likes (visitors click the button)
- Some Mystery Factor could be driving BOTH Likes and UPVs
Unless there’s an obvious 3rd factor in the mix, chasing after mysteries isn’t usually time well spent. The most likely alternative here is (2) – blog posts with more Unique Pageviews mean that more people click the Like button (+1 button, etc.). If this is the case, then we should see a relationship between Likes and +1s. If visitors drive Likes and +1s (and not the other way around), then Likes and +1s should be correlated (assuming some people click both).
The other piece of data we can look at it is referral traffic driven by Facebook and Google+. Although this is a little hard to pull out on the page/post level, blog posts often get direct visitors, so the referrer and the entrance source are similar. If Likes are well correlated with Facebook traffic and +1s are well correlated with Google+ traffic (admittedly, that connection is a bit more complicated), then it could point back to cause (1) – social signals drive traffic.
So, I pulled those three correlations (Spearman, again) for the post-Google+ data:
In a perfect world, causality-wise, either the green bar would be high and the blue bars low, or vice-versa. In this case, all 3 correlations were reasonably strong. Clear as mud, huh?
Social Chicken or Social Egg?
Part of the difficulty is that we have a bit of a chicken and an egg problem here – what came first, the visitors or the social signals? The reality is that it’s probably a little of both, and what we have over time may look something like this:
Social signals drive traffic, which drives more people to click social signals, which drives more traffic, and on and on. Social traffic also jumps the tracks – people who click on Like may also click on +1, driving more Google traffic, which drives more +1s, etc.
What Does It All Mean?
Although this was an exploratory study, I don’t want to just leave you with: “Hey, it’s complicated.” I do think that some of the correlations here are compelling, and that we can start to piece together a few conclusions:
(1) Social Signals Are Getting Stronger
Although the second study was a cleaner data-set, in the sense of the timeframe, the jump in the social signal correlations was notable. I think it’s pretty clear that social signals are gaining momentum and driving more traffic in 2011.
(2) People Use Multiple Social Signals
While there’s such a thing as overkill, people will click on both the Like button and +1 button, so don’t shy away from using both. I didn’t analyze Tweets in the follow-up, since a Re-tweet feels like a qualitatively different action (it’s more than a vote).
(3) +1s Are Working (In Our Industry)
At least for now, and at least for our audience, +1s are driving traffic, and their relationship, pound per pound, is almost on par with Facebook/Likes. If you’re not using the +1 button and you’re in a techie-oriented niche, now is the time to give it a try. The future of Google+ is anyone’s guess, but for now it’s having some positive impact.
We’re exploring whether these kinds of numbers would make for useful reports and tools down the road. If anyone has comments about what kind of advanced social stats they’d find useful or how they’d like to see these kinds of studies expanded, please let us know.
Fascinating data, Dr. Pete; I would be interested to analyze traffic paths.
I would only add that you may underestimate the impact and relevance of social proof. When users see tons of comments and/or thumbs up on a post, they are much more likely to click through, thumbs up, and share. The same is true if users arrives at the post through a social contact: they will be more likely to give it a thumbs up or share. This is especially true for a contact they respect that speaks highly of the content in question.
Simply, the psychological mechanism is quite strong that says, "other people like this post, so it probably has something to offer me."
There's been some interesting data on social proof when it comes to social media buttons (I think it was from Dan Zarrella). Basically, there's a tension between social proof and people wanting to have impact. The big "0" and very low numbers have some appeal, because people want to be first to have an impact and/or root for the underdog. Then, it's likely social proof kicks in, but that fades after a while. Once something has 100s of Likes, some people may be less likely to click, since they feel like it's already popular. It's tough to tease them apart, but I strongly suspect we'll see that there are factors that compete with social proof. More doesn't always drive more.
Every time I'm the 1st person to add a page to Stumble I always feel awesome for finding something others haven't shared in a similar manner. Its the same with clicking that Like button... if I'm #1 to click it then it feels pretty nice. Hell, being #5 or 12 or 30 isn't bad either. But once I see something has triple digit Likes, I'm less likely to actually click the button even though I do like it. But i'll probably just add it to Stumble instead. :)
Somewhat off-topic now, Dr. Pete are you going to be at SMX East?
No, afraid not. With the baby at home and my wife in a corporate job, I'm afraid my conference travel is pretty limited. Hoping to get back on the circuit in 2012 a little more.
Was hoping for an excuse to say Hi, buy you a beer, and talk shop. Next time then.
Satan...err Keenan, Dr. Pete, insightful as always! :)
For me, the reaction I have depends on the kind of proof. If, for example, I see a ton of thumbs up, I'm much more likely to read it. If I see 100+ comments, though, I am much less likely to comment.
Hi Peter,
interesting reasearch, useful because it offers us data to show when we say that Social Signals are important.
But, as a long time reader of the SEOmoz blog let me tell you one thing I always note
Thanks to the "timezone" I almost read the SEOmoz posts just few minutes after they have been published, and everytime what I see is that the Tweet numbers is usually already high, while the Likes and +1s no (the Likes start usually later... being mine the first one many times). The fact that the tweets are more is somehow logical, as SEOmoz announce the publication of the post first by Twitter, and its tweet is re-tweeted in mass.
So... my chicken & egg theory is this one:
Take into account that I'm not talking about what social signal IMO is sending more visits, but trying to explain what between Twitter, Facebook and +1/Google+ is the Social that starts making the visits coming to the SEOmoz posts.
P.S.: side related note. Recently I noted that the Likes count here is resetting to 0 when I revisit a post or refresh the page. I don't know if it is just I problem I have, or common to other users.
Gianluca Fiorelli your post deserves a article for it self really fantastic one to read. And I am also agree with the point there always would be far difference between tweets and likes and as you said tweets are always higher than likes.I never understand why maximum peoples consider FaceBook as a big source of traffic than Twitter as Twitter always give big numbers of followers and tweets too. I don't know about others but I personally believe that twitter is more effective as compare to others.Even we can take the example of SEOMOZ which have 81,424 Followers and only 15,808 Fans really a huge difference.
Sir! Absolutely right about the twitter count… As per my experience the counter on the social buttons are so random and one cannot trust them right away… the best way to count the social share count is again shared count…
Nice added Moosa Hemani really the best practise to assume the performance of social networks is just depends on the shared counts they are providing us.
Our Tweet counts are a little weird, because we get a bunch of auto-Tweets almost as soon as a post goes up (I'd guess 200-250). Problem is, those tweets have relatively little impact, I suspect. It's a bit hard to tease apart. It doesn't impact the correlations, since every post basically gets those Tweets, but it does make interpreting Tweet data harder.
Tweeted, liked, +1ed and thumbed up.
And I shall do the same this instant!
I think this is my love for data that make me smell this post at the 1st place ;)
Ok, as far as I am concern, I read multiple blog post on regular basis and the common action (with respect to social sharing) is if you have a sharing button or your blog like Twitter, Facebook and +1 I will more likely to like and +1 it and if I REALLY like it I will go a step further and tweet it.
+1 attitude is more like Quora (like you mentioned in the post that it’s working for our industry) a while back same was the case with Quora… so for +1 it works for tech industry and there is a strong possibility that for other industries it might not work as it works for technology niche.
I strongly agree with the point that likes bring traffic and traffic product more likes which again produce unique traffic, but usually I witness that the traffic i see on the first day will not be the same as on the 2nd day (probably lower traffic)…and in next coming days the traffic ratio will be lower than the 2nd day (more like the half life of the element).
Over all an amazing data to share!
Moosa i do agree and i have noticed too that decrease in traffic with passage of time... This kind of behavior by search engines is, with especially blog post. I guess this is because search engines particularity Google has set a real time factor in it's alogritham.
Very interesting and thought-provoking post Dr Pete.
I am just not sure if I can agree that your data are evidence that +1's are driving traffic. Are you in danger of not heeding your own starn warning that correlation does not mean causation?
In layman's terms, your data shows (quite sensibly) that the more traffic you have, the more social signals you get. In a simplistic example, if a post has 1000 visitors, you would expect roughly 10 x the number of social signals than if a post has 100 visitors
The very strong positive correlation between pageviews and social signals gives strong support to this idea, but not to your further conclusion.
Don't get me wrong, I am not at all denying that your idea is very plausible, I just can't seem to see that the data prove it.
I hope I was pretty clear on that point in the second half of this post. It's likely that social signals drive traffic AND traffic drives social signals. In fact, I'd say it's almost certain. The trick, going forward, will be trying to separate out those two influences.
This just opens up more questions for me:
Is the Facebook "like" traffic mostly from Facebook or does some come from Bing? Are +1s driving more traffic because Google is using that metric in their search results or is the traffic from Google+?
Frustratingly, our stats are still a bit dumb.
In this study we are crunching data without two important factors, time and identity. We need better stats, that identify users throughout their interaction with a page, and track that interaction over time.
Stats will come alive, and we can much better understand the lifecycle of a piece of content. The stats will produce some nice visuals that will look similar to cellular growth, a bit like this: https://j.mp/clickstreams Any one want to invest in my new Stats start up?*
*Yes, there are some technical hurdles, but the trick is to see what everyone else sees, and think what no one else has thought.
Man, I would love to see a graph on a popular post showing Pageviews over time (hourly? daily?) and show when TWEETS happen (and traffic from twitter) and LIKES happen (and traffic from FB) and +1's happen (and traffic from goog)
My friends at thesocialnetworkcompany.co.uk already have the network visualisation bit taped, it just needs a bit more information and re-purposing...
In the end of the day a key factor is the niche of the website, Most traditional websites and entertainment sites I have worked on you see a huge percentage of the traffic comes from Facebook mainly, a small percent may come from Twitter but Google+ has not really hit the main stream yet.
Sure enough with SEO and internet marketing websites you can really get a good amount of traffic from all sources as SEO's are savvy people who test all platforms =)
Overall thanks for sharing this data with us it is really insightfull.
No argument there - not surprisingly, we're early adopters when it comes to Google+. I think it still remains to be seen whether the +1 button will have real staying power, and there's some evidence that the early adopters are losing interest. If the general public adopts Google+/+1, then things will change quite a bit.
It'll be interesting to see how social media starts to impact and placing more and more. Thanks for sharing this information!
Interesting study.
You define statistically significant and unreliable as follows:
statistically significant (p<0.01)
[not] statistically reliable (p>0.10)
What is p in your model?
What was the sample size of the analyzed data?
Was the sample size equivalent for all measured correlation pairs?
I am just trying to understand your statistical methods better, sorry for being so picky. I come from the world of quantum physics ...
There were 50 posts in the first study, and 44 in the second (that got determined by the timeframe I used). You can measure a p-value for a correlation, as with other statistics, which is essentially the probability that the result occured by chance.
With correlations, it's a little weird, because obtaining a statistically significant result doesn't mean the result is really meaningful. For example, we can be 99%+ sure (p<0.01) in the first study that r=0.42 for Likes and UPVs. However, that doesn't mean that 0.42 is a strong or meaningful correlation. It just means that we have enough data to believe that number is reliable.
OK, thanks for the clarification. Right, statistical significance (in your example p) only measures the reliability of the data, and does not quantify the observable itself.
The ability for a social media user to share or rebroadcast a post to their own followers has to be one of the best innovations on the social media world. Drawing attention to a business' post by liking, retweeting, +1ing, ect really has created tons of opportunity. One of my favorite things companies are now doing is the "retweet, like, follow, ect" contests where you retweet a post for a chance to win some sort of prize. For example I follow a twitter account for a site called "Daily Dead" that is constantly providing updates on Zombie-related media. They have retweet giveaways almost every day. Many of the things they give away are either inexpensive or given to them. Without fault, I always retweet these posts. It really is an inexpensive way to gain more followers and awareness.
you
No, you.
Fantastic article! It's really interesting to see how even social networks are shaping/changing traffic, and how it could be significantly correlated - great hypothesis testing btw (and taking into consideration the google+ network dates)! I've only just recently gotten advice on how important social media networking is to websites and how it helps grow traffic, so I do think this article is very necessary. Keep more posts coming!
[link removed]
Good analysis, shame it's not as clear cut as I'd like. +1 is clearly a good thing to add to techinical savvy pages, but for a lot of the internet, I think it's impact is neither here nor there at the moment. I personally don't think that any of our visitors would use it at the moment.
Great post! Not only do social signals drive traffic (or reverse) but they also really allow you to carefully sculpt a content strategy that your web visitors want to read! Thanks!
Great Post Peter, I think this post will gonna increase the traffic rate of our websites through the use of social media signals. As we all know that SMM helps various companies to increase the sales, market awareness & helps to sustain in the market via social media signals. In todays scenario the soicla signals is in health competition along various competitors in the market to capitalize the market growth rate , revenue generation & aware the people thorugh their products. I think this will really helps us a lot from marketing prospective.
I agree with the above post. I see same story for my clients. Every single like, tweet or +1 is like a natural link earned. There is no doubt social media is getting strong and strong.
Like, Tweet, +1 & Thumbs up this post.
I appreciate the time that you've taken to do this research! However, the scientist in me is cringing at your methodology, which is a little suspect in a couple areas.
1. In the first study, you pull data from all posts in 2011. But then you end up only using the top 50 posts with the most unique pageviews. Why do this? If you have data for all the posts, use all the posts! You're unnecessarily restricting the variance of the data, which makes it difficult to draw conclusions. The only thing that you can conclude is that social media have a moderate correlation for the top viewed posts. You can't say anything about posts in general, which seems like a more more important thing to investigate. Even a comparison of the posts with the most pageviews vs. the fewest pageviews would offer much better insight into the possible effects of social media. But ideally, use all the data you have.
2. Your conclusion that "social signals are getting stronger" is not justified. In order to make such a statement, you'd need to perform an appropriate test to compare the correlations from the two time periods to determine if the difference is significant. Without this, you risk labelling something as truly different when really it's just due to random fluctuations.
At any rate, now that I've been mean and torn apart the post, let me restate that I appreciate the time you've taken, and I think more research like this needs to be done! It just needs to be done carefully, so that we're not making erroneous conclusions :)
I think I made it clear that this was an exploratory study - I intended it only as a starting point, and would welcome others to contribute their own data and analyses.
(1) My original population of interest was actually the top posts of 2011 - I wanted to see what the driving forces were for traffic for our most popular posts. You're correct, though, that this makes it dangerous to generalize the results to all of the posts on the site. That was also one of the reasons that I did the follow-up study (where I didn't restrict the sample based on popularity).
(2) I wouldn't have made that statement based just on comparing these two data sets. I only felt it was justified given that the data backed what we're seeing anecdotally and hearing from Google and Bing. I think it's safe to say that the influence of social signals is growing. I grant, though, that this study alone isn't proof of that. The two samples differed by many factors and can't be viewed as simply two windows in time.
Awesome Post. Thanks for including the refresher course on the Spearman correlation... it's been a while !
Thanks for the study so we now have definitive evidence that social signals drive traffic. This is great in helping to justify spending on social media marketing.
Many thanks to the folks at SEOmoz who take the time to actually do the research the answer the questions the rest of us have. I think one of the most useful conclusions is #2- People Use Multiple Search Signals. Content promotion is so important, and why would you want to limit where your content can be shared and seen by only including a few social share buttons?
I completely think that social signals play a role in driving traffic from many different corners. I think as time goes on social interaction and influence will play an even stronger role in rankings and visibilty than ever before.
I almost wish I could believe the same, however with the constant threat of black hat techniques to improve SEO - putting to much reliance on social media opens up far too many avenues of attack from malicious users. I believe that the social signals are and will remain as a minor part in the equasion. I actually think they're at the highest point they'll ever be at, now.
Social media is always been in a news from when new social networking are comig and playing a vital role for traffic driven process.Moreever,Google launche +1 which is more popular in now days to drive the right audiance towards your site.
I am fully confident on social media for traffic driving.
And at the end,Nice Article....Thumbs up for it..
Nice read Dr Pete, Actually we all are used in our daily life and also quite addicted with it also. Major thing is to how you get the traffic or good signals from it. It is vital and this all factors are impacts a lot.
As you said +1 is also now a one factor to get social signals and i agreed with you. We are getting good traffic as well form that. Some mystery factors you explain which i really like it. Thanks for connecting with us.
-Hiren
Dr. Pete
I Found Very Much Prospective Information about +1 button particularly concerned to traffic. The strong social signal by it, will surely help out for the company who is looking for Branding. I believe Google + Button is an important traffic driving factor for consistent reputation.
Appreciated and Thanks A Lot Dr. Pete now I got more knowledge about Google + Button Remarkable Effects.
I did some testing too. I wrote a poor post, something that shouldn't get any sort of traffic, but I got a few of my buddies to do a +1 and also RT it. Now that post is amount my top 5 posts with highest PVs.
As for FB like, I think Bing favors it more than Google (from Analytic Data I have seen on my own sites).
Asefati - Thanks for sharing your results. I agree with Linztm that the SERP effect of social
indicators is probably at a max now due largely to their ease of manipulation. G will probably
move on to some other novelty to remain 1/2 step behind the forces of darkness.
It's certainly tough to tell how much social signals drive traffic especially with correlation/casuation and chicken/egg problems and so forth. Of course in general you would expect posts with more views to have more social signals, just as you would expect them to have more comments.
Here's an idea for trying to see if +1 is more relevant than the other social networks... Look at the ratio of +1 to tweets or +1 to likes. Do posts with a greater proportion of +1's have more views than posts with an average proportion?
As a social media marketeer whos getting into SEO, this topic re iterates studies made about like buttons and tweet buttons months, if not around a ago when Twitter and Facebook first announced social plugins.
Perhaps it's Google Plus that has drawn the attention from the SEO community.
They are a powerful tool for driving referal traffic, and are a good signal of what content is most popular.
Great Job to those @ SEOmoz who conducted this experiment! The wrap up says it all in 3 simple notes:Social Signals Are Getting StrongerPeople Use Multiple Social Signals+1s Are WorkingSocial marketing can drive great amounts of traffic to our webpages!
Great post Dr. Pete - very interesting. Without getting mired down in the statistics, it seems kind of obvious that the more traffic you get, the more social signals you'll get, and vice versa. It makes sense that when people see a blog post that has received a lot of Likes or Thumbs Up or +'s, they're more inclined to want to read it. However, it was really interesting to see the patterns that you uncovered, especially with re: to FB and Google+. Thanks for doing this!
We all are influenced by the actual trends and therefore we like to participate and be in the game, too.
Not to forget the psychological aspect. If I stumble upon a page with already lots of likes or tweets I am automatically much more interested than there would be just a few of them.
How search engines read data of thumb ups and thumb downs?.
There's really no direct connection between thumbs and traffic, in the sense that thumbs aren't seen outside of the SEOmoz site. There are indirect influences, as someone who thumbs up a post is more likely to share it, link to it, etc. Part of what's tough with this data is that these signals are all connected.