Google issued a blog post yesterday describing the latest changes to its algorithm. In it, it makes the following comments about changes to how it assesses the value of links, specifically anchor text:
- Tweaks to handling of anchor text. [launch codename "PC"] This month we turned off a classifier related to anchor text (the visible text appearing in links). Our experimental data suggested that other methods of anchor processing had greater success, so turning off this component made our scoring cleaner and more robust.
- Better interpretation and use of anchor text. We’ve improved systems we use to interpret and use anchor text, and determine how relevant a given anchor might be for a given query and website.
Like many other sites, we have recently seen some flux in our rankings on a number of keywords. In this blog post, we’ll examine the recent changes we have seen for one of our pages, and what we think this tells us about how Google is really assessing links.
The Keywords: ‘Copywriting Jobs’ & ‘Copywriter Jobs’
Our jobs page is one of our biggest traffic drivers. For the past few months it has hovered around position 3-5 for the phrases ‘copywriting jobs‘ and ‘copywriter jobs‘. Yesterday, we noticed that we had jumped to position 1 for ‘copywriting jobs’, yet we remain at position 4 for ‘copywriter jobs’, so we decided to do a bit of research in order to find out why…
For both of these keywords, the top 5 positions are populated by the same 5 domains, namely Copify, Reed, Guardian Jobs, Brand Republic and Creativepool. A bit of research indicates that none of our competing sites are actively engaged in external linkbuilding, they are ranking purely based on the authority of their respective domains.
All 5 pages are similar in having a brief page summary and then a dynamic list of jobs, which changes on a regular basis. All target the term ‘Copywriting Jobs’ in their titles, and with the exception of Reed have this as a H1 tag.
This makes our research into the difference between our own rankings on these 2 terms all the more interesting…
Our linkbuilding to this page has consisted of a combination of blogging, press releases, articles, recruitment sites and social media, which has also gained a large amount of links from aggregators such as Topsy and Paper.li. Over the past 18 months we have acquired links from a total of 29 different domains, broken down as follows:
We have made a concerted effort to mix up our anchor text in a bid to build a profile that is as natural as possible:
32% of our links are anchored with ‘copywriting jobs’, 25% with ‘copywriter jobs’.
So why is ‘Copywriting Jobs’ #1 and ‘Copywriter Jobs’ #4?
If we look at the distribution of linking sites at keyword level, we can see a much more diverse source of links for copywriting jobs:
When compared to copywriter jobs:
What is telling is that 100% of the links anchored with ‘copywriting jobs’ are followed, as opposed to just 15% of those with ‘copywriter jobs’.
What all of the above seems to suggest is that the ‘classifier’ Google refers to is the nofollow attribute, which would explain why the 100% followed ‘copywriting jobs’ links are passing value, while the largely nofollow ‘copywriter jobs’ are not.
Google’s claim that they have “improved systems we use to interpret and use anchor text, and determine how relevant a given anchor might be for a given query and website” could be seen as a factor in the site ranking for ‘copywriting jobs’, but is undermined by the fact that our page does not rank as highly on ‘copywriter jobs’. We believe that Google is claiming that its algorithm is a lot more advanced semantically than it actually is, and is maybe trying to throw SEOs off the scent.
In conclusion, whilst it may be sensible to diversify your anchor text and follow/nofollow attributes, the old fashioned approach of exact match, followed links from a variety of sites is clearly still the way to go.
Seen your site go up or down in the SERPs?
Tell us about your experience, what do you think is really happening with these changes?