There’s a lot of emphasis placed on Google’s SEO changes, particularly for those who work in SEO and marketing, or who need to keep up with what is a fairly fluid environment. The search engine giant does certainly like to keep everyone on their toes, avoiding manipulation of its rankings as far as it can. Many in the industry try to predict what Google is about to do next and sometimes these predictions are spot on – however this isn’t always the case. Below are some of the SEO changes that a number of industry insiders openly predicted Google might make, but which haven’t appeared on the agenda – at least not yet.
Obviously links really matter for SEO but what we haven’t yet seen is any kind of bias towards on-topic links. These would be links that come to a website from another site that is ‘on-topic.’ So for example, a link to a recipe website from a cooking blog – if there is a link from a cooking blog to a recipe website and a link from a fashion blog to a recipe website surely the first should be the stronger link? Well, apparently not. That’s not to say that Google doesn’t have the technology to make on-topic links shine brighter but for some reason – and many years of research have proven this – it just doesn’t do it. This could perhaps be because Google feels that bias in this way narrows the view of what is popular and important online and with the broader view the results are better. So there is very little – if any - leaning towards on-topic links and this could be one of those SEO changes that we don’t ever get to see.
Although there are now many more equally important signals, anchor text inside a link remains key – much more effective than generic anchor text. So, for example, if the target phrase is ‘cooking supplies’ linking with the text ‘cooking supplies’ rather than ‘good place to find’ from the phrase ‘this is a good place to find cooking supplies’ will attain a lot more in the way of ranking power. Of course the anchor text is where a lot of the web spam appears and, consequently, where there is the opportunity for abuse and manipulation, which is probably why many thought Google would take action here. As yet, nothing has been done but some still believe this is only a matter of time.
When rel="canonical" was first launched, Google indicated that the phrase would be taken as a hint rather than a directive, but most of the research on the topic indicates that this just isn’t the case. In fact, Google seems to be doing quite the opposite of what it initially said and taking rel="canonical" as gospel every time. So, if rel="canonical" is accidentally added to every page of a trusted site and pointed at the homepage, Google consequently won’t index anything other than the homepage. Consequently, this is a change that many have wanted to see but that Google hasn’t really admitted it needs to make. The lesson here is if you’re implementing rel="canonical" then make sure that you’re doing it properly or you might end up with a lot of de-indexed pages by accident.
When social got big there was a school of thought that those social accounts that were trusted, very visible or important would have a lot more influence with the shares than has turned out to be the case. Given how the arrival of social was trumpeted as likely to be a major factor in search engine influence, the lack of effect on rankings might seem a little puzzling. In fact, Google almost may as well have stated that social shares are nice n’all but they were never going to have any influence on its rankings. The exception here is that very direct partnership between Google and Twitter a couple of years ago when those tweets and indexation became very important for Google rankings. And of course there is now Google+, especially the personalised results, but that’s not quite the same thing.
Perhaps what Google is doing here is cutting through the fluff of social shares and indicating it doesn’t feel that everything that is getting shares from trusted or important sites as interesting or valid as the shares might make it seem. Instead, this seems to indicate that Google thinks that if content really is important enough for it to be included in the rankings then it will at some point generate a link, citation or a mention and that this is what can be used for Google’s interpretation and in the ranking algorithm. Here, Google seems to be relying on the fact that social isn’t a reliable indicator of popularity but rather a starting point that will eventually lead to the genuinely important and popular websites, products and content being reflected in the link graph. It does make some sense to wait, particularly given the potential for manipulation with social shares, but many still find this frustrating.
The quest to clear up web spam has been ongoing for many years and Google has not made it quite the process that many thought it would. This is particularly with respect to cleaning up porn, pills and gambling and the uber-spammy keyword niches that it was thought Google would try to take a lot more action on than they currently have. Having said that, it does seem as if Google has improved at least marginally with respect to the links it considers and in judging the authority of a page that might be holding on to a domain without good internal links on some of the more trusted sites.
These are just a few of the areas where changes were expected from Google that never actually occurred. Perhaps they are still to come, or perhaps they were bypassed in the first place for a reason – as ever the search engine giant likes to keep its cards close to its chest.
Author: Steve Pailthorpe - Follow us on Google+