It can be seen as a reaction to Google’s success in clamping down on dodgy linkbuilding – it’s now more difficult for spammers to game the system in their own favour so they have to attack the competition instead.
They’ve been kind enough to share the data relating to the attack, which was presumably undertaken by one of its competitors.
I should point out here that Jellyfish doesn’t know who was behind it and isn’t trying to point the finger of blame at anyone in particular.
So, how did it go down?
Evidence of a negative SEO attack
This graph shows how Jellyfish’s search rankings were trending in 2014 in relation to several comparative SEO agencies.
The decline towards the end of the timeline occurred when Google implemented the Penguin 3.0 update which sought to penalise low quality backlinks.
The agency’s senior SEO manager, Jonathan Verrall, said that he typically checks for changes in the site’s link profile every week, so they were able to quickly diagnose the problem.
Closer analysis of the company’s backlink profile shows that there was a sudden spike of links in October and then again in November.
Data pulled from Cognitive SEO shows that from the beginning of August through to October there was hardly any suspect link activity, but in October there was a sudden spike with more than 1,700 new links that were seen as unnatural or suspect.
Obviously this kind of action is going to ring alarm bells with Google and is likely to lead to a ranking penalty.
Where were the links hosted?
Analysis of the suspect links showed that they had been posted as comments on thousands of websites.
This is a tactic known as ‘comment spamming’, whereby the guilty party uses spamming software to quickly post thousands of links in the comments section on blogging sites.
Econsultancy is often the target of these kind of comment spamming campaigns, though our spam filter usually keeps them at bay.
That said, before we upgraded our filter last year these comments would often slip through and in my early days at Econsultancy I was naïve enough to wonder why high profile agencies would resort to such flagrant and obviously spammy tactics.
What was the target?
Cognitive SEO’s backlink tool showed that the attack was primarily aimed at Jellyfish’s SEO training page.
All the links used exact match anchor text for ‘SEO training’ which, coupled with the fact that the landing page had been optimised to within an inch of its life (quite legitimately, and as one would expect from an SEO agency), meant that Google rightly thought something was afoot.
Consequently the agency was penalised and began to lose visibility in search rankings.
Verrall used several different site and web crawling tools to make sure he had identified all the dodgy backlinks, then set about the long-winded process of asking sites to remove them.
He said that the amount of effort involved with getting links removed depends on the type of website. Webmasters at link farms, such as link directories, article directories and blog networks, tend to ask for removal payments.
But thankfully Verrall says that these types of sites are generally a thing of the past. However…
…if you are looking to get links changed or removed from established websites who treasure their readership, they tend to be very accommodating and will change backlinks quite happily.
We get link removal requests fairly regularly at Econsultancy and in general we ignore them as it’s often clear that the person making the request is the guilty party (I’m not suggesting that’s the case with Jellyfish).
Our old content director Chris Lake wrote an interesting article discussing the various link removal requests we receive and explaining why they’re such a pain in the behind.
It took Jellyfish three days to get through the first round of emails to webmasters, which were followed up with a second round of emails to those that didn’t respond.
If any sites still failed to respond to the request or refused to remove the link then Jellyfish was forced to upload them to the Google Disavow tool.
However, disavowing links can bring its own problems, as Google likes to see that you’ve made a conscious effort to remove the offending links.
According to Verrall:
It’s also good to keep a log of the removal progress within your disavow file by using the comment functionality.
We typically keep track of the removal process in a cloud based spreadsheet which we reference within the disavow file just in case a member of the Web spam team manually reviews our efforts.
The road to recovery
All the dodgy links have now been removed or disavowed, but the recovery process isn’t a quick fix.
Though the attack was swiftly identified and fixed, Verrall said he will need to wait until the next Penguin update to see if the site has fully recovered.