Warning, this paragraph is an intro/rant, click here to skip straight to the bad link metrics. This post is also a work in progress. Trust is hard for me – at least in my professional life. I probably receive 20 Skype connection requests, 100 emails and 30 phone calls from “SEO’s” in an average week. I purposely put the term “SEO’s” within inverted comma’s because while they might feel like they are an expert in the field of search engine optimisation, the truth is that they are most certainly not. Not only are they not experts, but neither are 95% of other “SEO’s”. My reason for this judgement? They seem to have no idea what makes an unnatural link. They offer services without understanding the consequences. What makes this worse is the businesses they are offering their services to obviously don’t understand either. My intentions for this post were not to rant so I will leave my judgement there.
So what can be done about it? Simple. We have to educate! If you are reading this post then chances are you will want to hear what I think classes as a bad link. It could be that I even pointed you to this post to save myself the 300,000th conversation about unnatural links this week. If you aren’t sure what a bad link is, just keep an open mind. I’m afraid the truth won’t make your job any easier though, quite the contrary actually. There are no quick fixes in SEO (not sustainable at least, there are plenty of Black methods that may-well get you slightly quicker results but chances are they wont last long unless you scrutinize every last detail – and even then, you can get hurt).
Shut up and tell me the bloody metrics!
Request granted! But before I start it is important to answer one question; “OK so what percentage of each of these types of links can I have?”. In reality, this question might seem like it has a simple answer, and it does, but not the one you might be thinking. Nobody can just provide you with a generic % figure of how much of your link profile should be shared among these metrics. What you should do is compare it to your keyword competitors and try and better them.
So here goes;
1. Has the site linking to your site been deindexed?
This one should really go without saying but it is amazing how many SEO’s don’t check. If the site has been deindexed from Google then something is wrong. It has probably been penalised.
2. Has the page linking to your site been deindexed?
Google penalties don’t always apply to an entire domain. Pages can also be penalised. If the page linking to your site is not indexed in Google then this is not a good link to have.
3. Go deeper than point 1 and 2.
How many links pointing to the domain/page that is linking to you are from sites/pages that have been deindexed? That’s right, you really should go this deep! Links have a pass through effect. The effect (both good and bad) doesn’t stop at the page the link points to, it then passes through to the page/domain that the recipient of the link points to. Visually, this would look like a tier system with filters. An example of this is as follows:
Site1.com – this has been deindexed from Google. It has a link that points to site2.com
Site2.com – this is indexed but has a link from site1.com. It also has a link that points to site3.com
Site3.com – this is your site. You have indirectly received a bad link, passed through site2.com
The above example is showing a 3 tier system but in reality this could pass through many tiers, only Google knows for certain. Obviously the more tiers, the less impact it could have. I would suspect that lots of websites have links from other websites that have links from deindexed domains. Once you add a 4th tier into this, you probably cover a huge chunk of the world wide web.
4. Fake PageRank? Use your head.
Fake PageRank should be pretty easy to spot. Use your head a little here. If a website has a so-called PR of 8 but has a low PA/DA, chances are it has a fake PR. Use every metric you have at your disposal. Check AHRefs, MajesticSEO, SEMRush and any other provider of link data, if there is a large discrepancy then you know the PR is faked. There are also a few tools knocking about that can do the same job but some are a little inconsistent.
5. Domain Age
Yep. It does still matter. You should ask yourself not only how old the domain is but how many owners it has passed through. A brand new site still has a question mark over it. I’m not saying that a new site is always a bad link – far from it. What I am saying is that you should judge each new website by its own merits.
6. Sitewide links
Having some sitewide links is natural. Having a large proportion of your incoming links coming from from every page on a website is not natural. It looks fake and it looks like you are trying to manipulate Google’s algorithm. In addition to this, also take into consideration point 3 again. Go deep. How many links pointing to the site you are trying to get a link from are sitewide?
7. Footer links
Not to be confused with the above. Technically, if a link is in a footer then there is a good chance it will be sitewide too – but not always vice-versa. You have to make sure that if you do have footer links pointing to your site, they don’t take over your link profile. Go deep again here. Look at the 2nd tier (as point 3 suggests).
8. Server saturation
If a large proportion of your links come from the same server then it looks unnatural. It looks like a link farm. Remember to apply point 3 again here – not only does this apply to your website but also check the site you want a link from. If that site has too higher proportion of its incoming links from the same server then you might not want that link.
9. Excessive forum/signature links
MORE TO COME…