Although for many months the topic of Google penalties has been discussed, all the time new persons appear who ask – how can I know which links have been considered unnatural? Unfortunately Google is no help, because, in spite of several promises of Google employees, we still do not get their examples in the notification in GWT.
People who are completely lost can use tools, such as Link Detox (here you can get a free trial) which performs automatic backlinks analysis. Unfortunately, relying only on the automatic tool may not be helpful, since the analysis should be quite detailed, and 100 % trust in the data from any tool can make you get rid of many valuable links, which were wrongly qualified as toxic. Still it is a kind of beginning.
How to recognize toxic links? Let me start with Link Detox examples, and later I will discuss common elements for websites that have been penalized that I had the opportunity to analyse. First of all, you should upload the file with links downloaded from GWT, so that the analysis be complex – in addition Link Detox collects data from several popular tools. In a free trial, it will not display information about the removed links but it will make analysis of those that are still displayed. Scanning may take even a few hours, so it is worth activating it and do something else to return to the tool later. The result of the analysis looks as follows:
After clicking the chosen type of links (e.g. Toxic), their full list is displayed along with the following information:
- address of the website with the link;
- anchor text;
- the rule on the basis of which the link was recognized toxic;
- several indicators determining the value of the referring domain;
- e-mail address to the webmaster;
which can be extended, among others, by:
- target website address;
- link status (nofollow, dofollow);
- link location (e.g. footer, header, comment etc.)
It is only a part of all the available information. What should raise our interest in the first place are rules (the column “Rule”). Toxicity of links is determined only by 2 rules:
- TOX1 – no domain indexation in Google;
- TOX2 – template of the website can be found on the list of hazardous (viruses etc.)
Unfortunately, it is impossible to learn on what basis and where the list of suspicious templates is, however, in the case of 1 of the rules attention should be paid to whether the tool reached accidentally the automatically generated sub-domain (e.g. asdfg.addictedtoseo.com), which returns content identical to the correct domain, but will not be indexed. In other cases, usually the rule operates correctly and enables to find low quality microsites that are banned and/or redirect to the new domains. It is also worth analyzing the links marked as suspicious, where you can find links such as:
- links from weak domains;
- networks of links – recognized on the basis of a large number of backlinks from the same referring subnets, from the same IPs belonging to the same owner, with the same Google Analytics code, Google Adsense code and on the basis of many other factors;
- site-wide links from footers;
- links from entries in directories and from articles on low quality domains;
- links from old domains whose even main websites do not have PR.
A full list of rules for suspicious links can be found here. As you can see we should not rely solely on the automatic tool, however, particular rules immediately give hints regarding what to pay attention to. Even atother links (the rule “Healthy”) it is worth analyzing them, because in turn among them we can find low quality links. At the same time, the tool may be used for competition backlinks analysis and to find interesting sources of links deemed to be “healthy”.
If you prefer to take a risk and rely solely on this tool, you may immediately generate disavow file. You can also mark that in the case of detection of many links on the same domain, the tool, instead of writing subsequently all the links, should use the domain: domain.com command. However, in the first place, you should remove possibly many suspicious links, and report only the ones that couldn’t be removed.
Well, OK, we already know how to use Link Detox so that it’s not time-consuming. And what happens if we want to have a more detailed analysis? On the basis of the previous cases, harmful links include:
- profile links – in particular if they are established in large quantities and are displayed mostly on foreign websites, including forums;
- links from entries in websites directories – in particular the ones added in large quantities;
- links from articles – also those added in large quantities, usually in the form of synonymous texts;
- links from comments on blogs and forums – again the ones added in large quantities by means of Scrapebox or Xrumer;
- link exchange systems – although it is the easiest one, because it is enough to remove the website, or the request to the owner of the system to block the domain and remove links referring to it;
- links from low quality microsites – expecially if theye are displayed in a very schematic manner and have over-optimized anchor texts. However, in this case it is worth trying to change the way of displaying them instead of immediate removal.
The problem is connection of mass, poor quality websites and over-optimized anchor texts, and therefore – schematicity of activities. So far, manual filters were observed mainly on the websites with more than 500 referring domains, although the majority was closer to 1000 and even several thousands.
I encourage you to share information about links which were faulty in your cases. Are your observations consistent with my list of “potentially harmful links?”