Google Indexes Twitter Links
How to get indexed by Google? Just use Twitter. This is no joke. I ran a test on Twitter to find out whether Google indexes Twitter links and surprise, surprise it does in certain cases.
Of course Google can’t ignore Twitter, the fastest and most up to date social networking and link sharing site out there.
The problem is Twitter apparently doesn’t want to get links shared on it to get indexed by Google and prevents it by adding the nofollow attribute.
Yet on the other hand Google and Twitter renewed their cooperation after a few years of abstinence in 2015 and Google even added tweets for some accounts on top of search results.
Twitter uses nofollow on outgoing links
Google, along with other search engines introduced the nofollow attribute allegedly to combat spam and later to scare webmasters into submission (either they add nofollow to ads or they get penalized).
Now that’s a problem. They can’t ignore their own rules, can they? Twitter uses nofollow on all links doesn’t it? No, the mobile version which gets indexed as well does not use nofollow. See my profile in the mobile version. Ironically the Google cache version of it does.
Whatever the reasons may be: Google actually indexes Twitter links.
Whether it’s a bug or feature, I don’t know. I can imagine Twitter closing this “loophole” now that “evil SEOs” like me write about it. I’d certainly like to see how Google will justify indexation of Twitter links then. We all know that Google is black hat.
They show different content to users than to spiders, especially in countries like China… The Google engineers will surely find a new way of gaming Twitter. As of now webmasters, SEOs and Google shareholders can rejoice: Google officially is tapping into the wisdom of crowds displayed on Twitter.
The Twitter indexing test setup
How did I find out about Google indexing Twitter links? I set up a page on my onreact.com domain solely for the purpose of testing Twitter link indexation. It contained only text. It was a dead end page: No incoming or outgoing links. With one exception: I tweeted the link once and asked my followers to ignore it. Also I favorited it on Twitter.com
I checked a few hours later, a day later, two days later and nothing happened. So I forgot about the test. Now after 3 months I rediscovered my bookmarks and took a quick look in the Google index and what did I find? The test page.
Now at first I suspected some third party Twitter tool to display my link and thus Google indexing. No, as of now there is no other instance of the text on that page than my Twitter stream and the page on onreact.com itself:
Please note that the mobile version was not in the index in spite of my mobile page being even cached by Google.
How I messed up the test
The most likely scenario is that Google indexes Twitter links directly from Twitter. Sadly I messed up when setting up the test. While adding a server based analytics script to the page to see Google spidering the page I made a mistake.
I used the wrong path to the analytics tool. So we don’t know when the Google bot actually hit the page. You might want to repeat the test with a better setup to make sure the result is valid.
It seems that there is a substantial delay to indexing links via Twitter.
Maybe sneaky Google doesn’t want to get caught in the act? Remember that Twitter is a Google competitor with its real time search capabilities. I’m quite sure Google is keen on catching up here, using Twitter data directly via the API for instance.
Btw.: Neither Yahoo nor Bing have indexed the page. Thus Yahoo also can’t find any links leading to it.
Last updated: August 24th, 2015. Added the new Google-Twitter deal.
Please note that “nofollow” only means no PR will pass, never means it will not being indexed.
Eric: Of course, a page might get linked from elsewhere and thus get indexed.
Still nofollow means literally do not follow this link.
So if a nofollow link is the only link to a page it shouldn’t get indexed.
In this case there was no other link unless you count the mobile version which Google does not index itself though.
Great little experiment. I did a similar experiment, not specifically with Twitter but with NoFollow links from blog comments. You can read the complete post and comment if you wish at http://www.seo-writer.com/blog/2009/08/31/look-who-follows-nofollow-links/ , but the crux of it is that for indexing purposes, the search engines all seem to follow NoFollow links, but it’s hard to tell how much “link juice” results.
Anyway, I just tweeted this post, so that should be worth something. :-)
Funny , this very thing is getting out pretty quick. They were just talking about this over at SEOMOZ.
Great experiment and thank you for sharing it here.
When I view Google’s cache of the Twitter mobile page it shows me the content from the non-mobile page.
Maybe Google has (manually, without the canonical meta tag) determined that the non-mobile page is the “canonical” version and only indexes that one — though it could be following the links on the mobile page.
“Nofollow” actually has no real meaning. The badly-defined specification is here:
http://microformats.org/wiki/rel-nofollow
Google just makes up a lot of random stuff on top of that whenever they want (“PR sculpting”, secretly revoking “PR sculpting”, etc.).
We will watch out for this. This will make a big change to our rankings (MAYBE).
This means Online Businesses should engage to twitter to compete in Search Engines
This SEo stuff with Twitter, Facebook and such other sites is driving me nuts. I’ve read so many contradictory opinions but this is the first I’ve read with actual proof and screenshots so THANK YOU for clarifying!
Cool test! However, it is a bit of a head scratcher. Google clearly states that nofollows will not even be followed (http://www.google.com/support/webmasters/bin/answer.py?hl=en&answer=96569). So why would you get found by the google bot if the nofollow from twitter is your one and only link!? I’d kind of like to do this experiment myself. There are more ways than just links that Google can find your site. For example: sitemaps & through the google toolbar. I’d want to know that the results were 100% legit.
Well whatdya know? Forget submitting to SE’s. Twitter, Facebook or Ezine article them!:)
now that is giving me an idea how my site can be indexed on google..thanks!
Cool..Its just now that I saw the use of twitter in SEO..thanks!
No wonder why there are a lot of twitter users..i have a lot of people following me with a bunch of products in their profile pictures..
I think that Google indexes all no follow links. There have been a lot of tests about this on the net. SOme bloggers even practiced that they got link juice and search enigne ranking by commenting on no follow blogs.
Great post. I was not sure about it myself. Was googling to find out more info about this as I read it on an seo forum when I came across your site. I will give it a go and see how quickly my blog gets indexed.
Twitter is most usable social networking site in now times and most important thing is that,google indexing the tweets from twitter user and it can help to improvise ranking and also traffic.
There’s a difference between meta noindex and nofollow and rel=”nofollow”.
rel=”nofollow”, which is what we’re talking about here is designed to do exactly what Eric C. mentions in the first comment – not pass pagerank or link juice.
Where has this idea come from that it’s to prevent page indexing?
Frankie: It’s not meant to prevent indexing but needs another link to index that page.
In my test there was only one link and this link used nofollow.
From SEOroundtable back in 2006 http://www.seroundtable.com/archives/006926.html:
“Technically, rel=”nofollow” does not mean the search engines won’t spider the page. They will follow the link, spider the page and count the link as a backlink. What rel=”nofollow” means is “don’t trust the link”, i.e. don’t pass PageRank/TrustRank, etc.
The robots meta tag “nofollow” is different, and really does mean “don’t follow links from this page”, and has nothing to do with backlinks or PageRank.”
Frankie. I know the difference between meta tags link attributes ;-)
Also make sure to check out the oficial Google stance on nofollow links.
http://www.google.com/support/webmasters/bin/answer.py?hl=en&answer=96569
It says:
“How does Google handle nofollowed links?
In general, we don’t follow them. This means that Google does not transfer PageRank or anchor text across these links. Essentially, using nofollow causes us to drop the target links from our overall graph of the web. However, the target pages may still appear in our index if other sites link to them without using nofollow, or if the URLs are submitted to Google in a Sitemap.”
Last but not least spidering is NOT indexing but as google says it doesn’t even spider pages via such links.
Twitter, my arch nemesis. I know it well… Indexing twitter links won’t make me more likely to use it though.
Yes I agree that google most likely indexes no follow links, at least a good percentage of the time, simply because there are too many other issues involved with site relevance as a whole….
But aren’t the links on twitter nofollow??