A Tweet's Effect on SEO Rankings - An Unexpected Case Study(seomoz.org)
seomoz.org
A Tweet's Effect on SEO Rankings - An Unexpected Case Study
http://www.seomoz.org/blog/tweets-effect-rankings-unexpected-case-study
6 comments
I've noticed the same thing with reedit and my site. I've had a blog post on evolution rank in google for evolution related searches because it got to the top of the atheism subreddit. The organic traffic trailed off pretty quickly though.
Seems to be a fleeting effect. Impressive temporary bump though.
Time will tell if it's temporary. The thing is when content gets a lot of retweets, it means it's probably pretty link-worthy. So the tweets could have given it an initial bump, but then traditional backlinks could keep the rankings high.
I wonder what Google will do in the future to curb this from happening as they seem to crack down on everything and anything they think inflates ratings..
I think they already are doing "something". The article does note that botnet-tweeting doesn't appear to have any effect.
Whereas a tweetstorm from "real" people is probably indicative of actual (if temporary) interest, hence legitimate pagerank.
Whereas a tweetstorm from "real" people is probably indicative of actual (if temporary) interest, hence legitimate pagerank.
Interesting.. does Facebook provide the same linkjuice that Twitter seems to be providing?
Good question - I wonder what the default breakdown of public vs. private shares for FB users are and how SERPs are weighting FB shares vs. tweets/retweets. Does anyone know what percentage of content on FB gets indexed?
I brought this up with the founder of SEOMoz during the follow-up to an interview last night when I noticed one of my sites doing exactly the same thing -- http://bit.ly/f5iFvV -- and it's really got me perplexed.
What's happening, I think, is that different social sources are providing Google Juice with different decay rates. Which sources have which rates? Your guess is as good as anybody else's. Ugh.
I wonder how complicated the entire business of figuring out how to tell people about your startup or your web article is going to get? The trajectory isn't looking too good.
What's happening, I think, is that different social sources are providing Google Juice with different decay rates. Which sources have which rates? Your guess is as good as anybody else's. Ugh.
I wonder how complicated the entire business of figuring out how to tell people about your startup or your web article is going to get? The trajectory isn't looking too good.
I think it is more likely that a sudden twitter storm on a tail keyword triggers Query Deserves Freshness.
Example: there currently exists a bishop who will be the next pope. His web profile today is substantially smaller than BCC's, and is likely dominated by 200 word wiki articles, a page at his diocese, and perhaps some Time Magaxine profile from when there was speculation as to who would follow JP2.
His Holiness will eventually pass away, which is going to ignite an immediate Worldwide News Event that will cause query volumes for the top ten candidates to skyrocket. Google wants to return fresh results for those (e.g. not the pages currently ranking). Using Twitter to trigger QDF gets that hours before the link graph will reflect the new speculation articles' popularity, without requiring manual intervention.
This is, of douse, just one signal. Google News feeds, the query spikes, and sudden radical increased appearance of branded terms on trusted pages all point to the conclusion.
Example: there currently exists a bishop who will be the next pope. His web profile today is substantially smaller than BCC's, and is likely dominated by 200 word wiki articles, a page at his diocese, and perhaps some Time Magaxine profile from when there was speculation as to who would follow JP2.
His Holiness will eventually pass away, which is going to ignite an immediate Worldwide News Event that will cause query volumes for the top ten candidates to skyrocket. Google wants to return fresh results for those (e.g. not the pages currently ranking). Using Twitter to trigger QDF gets that hours before the link graph will reflect the new speculation articles' popularity, without requiring manual intervention.
This is, of douse, just one signal. Google News feeds, the query spikes, and sudden radical increased appearance of branded terms on trusted pages all point to the conclusion.
Yeah I thought about QDF, and I sincerely hope you are correct. I could swear that things like links from HN are showing the same decay thing, though, just at a different speed. There have been a couple of sites that I started from scratch and only posted over here -- they showed the same kind of peak/falloff.
Of course, it's not an either-or situation. Both could be true.
Of course, it's not an either-or situation. Both could be true.
So far only a week has passed and we seem to have settled into the #12 spot. But it will be really interesting to see if the ranking sticks around. Sadly with this case study being dropped into our laps we weren't set up to run it as a test and only have "after effect" data to go off of.
I have done this before, to maintain and improve the ranking get a few backlinks that match the phrase you have already now on Google. I have launched about 15 websites this way and keep good rankings. Google sees the initial feedback from the social sites, lists your site, but it will keep falling if you don't get some matching backlinks pronto. If you type into Google cheap social marketing I have the first site, I used these techniques to get listed and rank the site. However, what I am offering is overly complex and I don't believe people understand the value of knowing Google's trigger points, when they notice the social attention, when other sites start taking the twitter/etc posts and adding them to "social conversation" websites, etc. All of this feeds back into your organic ranking as long as it is relevant.
Given how unspecific the search term is, I can only assume that's because 94.58% of that traffic wasn't actually organic traffic, but was SEOmoz staff + people they talked to clicking on it.