One of the things I love about search engine optimisation is that it keeps changing. No one knows for certain Google’s algorithm, so the SEO community constantly conducts experiments. These SEO experiments are then published, shared, commented on and recreated to help us gain even more clarity.
Nothing is set in stone. You can only form hypotheses based on existing data at the time. As more data pours in, you change your hypotheses.
Mikhail Tuknov of Infatex.com wrote a great post on Search Engine Journal regarding SEO experiments. The link’s at the end of the page for those who have a few hours.
For those who want just a quick overview, I’ve provided a quick summary the experiments he listed below. You might want to head over to the original article for more details.
SEO Experiments You Should Know
Does the Google Disavow Tool really work?
Yes, it does. A case study shows how an SEO company used the disavow tool to lift a manual unnatural links penalty. [NA]
Do nofollow links pass PageRank?
While Matt Cutts clearly states that the nofollow tag does not pass value, another experiment indicates otherwise i.e. some value is passed. A bug or just misdirection from Google? [Oct 2010]
Do Twitter and Google+ influence your rankings?
They do, with results in Google+ showing up in the SERPs very quickly. From this SEO test, it shows that companies should not ignore their social signals. [Jul 2011]
Google can replace your title tags if it finds it ‘non-informative’.
This usually happens if it’s filled with keywords and doesn’t describe the contents of the page. [Feb 2012]
How useful are the hreflang and canonical tags?
The hreflang and canonical tags seem to be very useful in cases where you are conducting international SEO ie one website with many target countries and languages. [Mar 2012]
Is it possible to recover from a Google penalty.
Yes, it is, but it’s not easy and takes time. You will have to wait until the next refresh of the algorithm comes along. More interestingly, we now know that the over-optimisation filter is separate from Penguin. [May 2012]
Can blogging increase your search traffic?
Yes, it can. A real life case study showed that the ROI they got from blogging was $200 per hour. [Jun 2012]
Does Negative SEO work?
It seems like it, to a certain extent at least. A case study showed a site dropping from 3rd to 14th place after it was subjected to negative SEO. [Jun 2012]
Will capital letters effect your rankings?
Initial findings showed that only capital letters in your URL seem to affect your rankings. However, Google has updated its algorithm and there are no longer any differences between uppercase and lowercase letters in the URL. [Jul 2012]
How important is a single Google +1?
Not very. A few +1s do not seem to have a noticeable result. However, if you +1 a page, it will make it more visible to your friends in the SERPs. [Jul 2012]
Can Google read naked URLs even though they are not hyperlinked?
Yes, it can. It did not appear to pass PageRank, but with the rise of co-citation and co-occurrence, that might no longer be true.[Jul 2012]
Can someone hijack your content?
Yes, they can. If the competitor has a higher PageRank, they can copy your content, post on their own blog and Google will think that theirs is the original. You might even get penalized for duplicate content. [Nov 2012]
Can you view outbound links for sites you do not own?
It seems that there is a way for you to view links from other websites in your Google Webmaster Tools. But with backlink checkers like OSE, I don’t think it makes much of a difference now anyway. [Nov 12]
Is the number of existing outbound links on a page a major consideration when chasing an authoritative link?
It’s still worth your time chasing after links on an authoritative domain even though there are already many outbound links on it. [Dec 2012]
Can you rank a new site quickly even after the introduction of Panda and Penguin?
Yes, if the proper SEO techniques are used, it might be possible to rank a brand new website quickly for less competitive keywords. [Jan 2013]
Does contextual text surrounding a link affect how much value it passes?
Yes, it does. Surprisingly, this experiment seems to indicate that ALT image links have the most value. I’d wait for more data before changing all my ALT texts though. [Apr 2013]
Here’s the link to Mikhail Tukno’s excellent article on SEO experiments. Warning – set aside a few hours to truly appreciate it.