It was 6:06 am on a Sunday morning when a Skype messaged arrived that said:
“Eric HELP! It’s gone… it’s all gone! My site is gone from the index!”
He was right!
It was an emergency. His entire business had been de-indexed from the search engines due to a severe manual penalty and come Monday morning, the phones would not be ringing for the first time since he opened the business 8 years prior.
SEO miracles have been accomplished before so I dropped everything and set to help him. The penalty was due to links violating Google’s guidelines and as expected, he definitely had links that violated the guidelines.
How I Removed The Bad Backlinks In 5 Steps And Restored His Traffic
Yep, the typical profile links, there were private blog network links and other various spammy links found. Those were the first targets removed.
It’s the strong links stupid!
Anyone that’s been working with Penguin & manual action penalties will also know that Google loves to go after links that affect rankings (which means links that pass juice and can have a significant impact on the search engine rankings.). Yep, that means they target do-follow links that can be hosted on excellent sites.
(On a side note, that’s why all the services that sell you on finding your penalized links by looking at “bad neighborhoods” and “spammy looking links” are complete garbage. If you’ve ever paid for a service that runs a scan and provides you a list of bad links in the hopes of getting out a penalty, then you’ve just been screwed over.)
But getting back to the bad links… Google is going after the LINKS THAT MATTER. The ones that can take you from page 3 to page 1. The ones that REALLY make things move and when the manual spam team returns sample bad links, they often pick out the ones that have the most impact on your rankings.
Why? Because Google already ignores the links that have little-to-no impact. Those don’t make your site rank. Do you have a no-follow link on a PR0 page that has been de-indexed since 2007? Big deal.
Google doesn’t care.
But… if you just purchased 20 PR6 domains from an auction and you slapped on your keyword stuffed links on all of them… That will raise some red flags!
And rightly so.
Google has to figure out what they can trust in some way and they do it through patterns.
That’s not new (I unveiled about a year ago during the SEO Rockstars convention in California) but what IS new is that Google is screwing up and including legitimate links that do not violate the quality guidelines within their results.
The algorithm that is usually correct
Dear Google, if you’re going to create an algorithm that penalizes links, then it should NEVER penalize legitimate links. I would rather miss 10 bad links and have an algorithm that is slightly less efficient than to start penalizing legitimate links. This is equivalent to sentencing innocent men to death without proper evidence. You just don’t do it because it’s better to let 9 people walk than to kill an innocent man.
One recent example of a legitimate link getting ‘caught up’ in the mess was from a newspaper in the United Kingdom. The recipient of the link (ie: my client) had never heard of the link. He didn’t request it and he didn’t have anything to do with it’s creation.
The link was completely natural, organic and had a unique anchor text never used before. The author of the newspaper article must have Google’d the answer, found my customer’s website and linked to it. (Just how Google wants you to discover links in their Utopia of links).
Yet… Google was now claiming that this link was in violation of the webmaster guidelines AND it was responsible for crashing the site’s traffic.
Another example is a link from Moz’s post http://moz.com/blog/dear-google-links-from-youmoz-dont-violate-your-quality-guidelines where the brand name of a writer was submitted as a bad link. In that situation, the editor actually does endorse and support the link back to the website. If you can’t credit websites anymore… how is the web going to work?
And of course, the examples continue. I cannot tell you how many times I have talked to a web owner that has said: “I have never seen this link before… I have no idea why and/or how it’s there” when performing link analysis.
The Google screw up
So why is Google serving up these links?
The short answer is: They are screwing up.
The long answer is that they are hiring poorly trained manual review experts that are using a broken algorithm to track down bad links. The algorithm that tracks down bad links does so by looking at link building patterns that use link velocity, first crawl date and anchor text to determine which links are valid and which ones are not.
In English, if you build 15 do-follow links with the same anchor text on the same day, all from different IPs and with the same text, then you’re likely to trigger a penalty. (And of course, the thresholds are determined by the total links pointing to the site. So a site with 100000000 incoming links will have a much higher tolerance than a brand new site with 2000 links.)
Google looks for patterns of unnatural link building and when you can identify those patterns, then removing the penalty is just a matter of removing the links within that pattern.
Pretty simple… and that’s what has been working for myself and my clients for years.
So how do you do it?
Step 1 Finding unnatural link patterns
First you’ll want to go to Webmaster Tools to download your latest links.
Click on download latest links.
We’re using Webmaster tools for a few reasons:
1st. That’s the official list that Google will use themselves… and when the manual reviewers mention a link, it’s usually in this list!
2nd. These days, it tends to find more links than services such as Majestic and Ahrefs (It wasn’t like this in the past… but Webmaster tools has stepped up their game.)
Step 2) You want to find the unnatural patterns of anchor text
The beauty of this list is that all the links are in ORDER found. That means that you can use that exact order to find unnatural patterns that might have occurred.
For instance, if you received 10 links in a row with the exact same anchor text, it’s highly unlikely that it was done organically.
What are the odds that 10 different sites would link to you with the exact same anchor text?
Moreover, what are the odds of different sites using the same long tail anchor text to describe your content?
That’s why you have to go and pull the anchor text links from those URLs to figure out how they are linking to you.
Step 3) Write the anchor text
In this picture, you’ll see I have the anchor text of the link on the right hand side. What we’re looking for is unnatural patterns of repeating (or very similar) anchor text.
Step 4) Find Unnatural Anchor Text Pattern
If you want to know how to speed up the process of finding all the anchor texts, download the full checklist for 2 tricks not included in this guide.
Can you guess which keyword is penalized?
<hint> It starts with “CD” and ends with “Uplication” </hint>
What do you think so far? Join the discussion of this post on Facebook
Step 5) The Removal
And that’s really all there is to it.
(At least… for finding the unnatural link patterns)
Now the challenging part is addressing the bad links.
Many people agree that the disavow tool is borderline useless and in my experience, I have found that it essentially turns do-follow links into “no-follow”. (Source: John Mueller Webmaster Talk).
So you can add them to the disavow list however it won’t help your site pop back into Google’s good graces.
Instead, you have 2 main options: MOVE or REMOVE
By moving, I mean moving the content from that page and letting all the links 404. Sure, this means you’ll see 404 not found in your webmaster tools but not all 404’s are bad.
When the links point to a 404, they no longer count towards your website.
That means if you have 999239 spammy links pointing to yoursite.com/index50.html and then you REMOVE the index50.html file, then those links are essentially pointing to nowhere.
Problem solved! (Albeit, you have to restart link building and move the content to somewhere new!)
This works great for sub-pages that are penalized but not-so-great for the homepage.
If it’s the homepage that’s penalized (and you don’t want to change domains), then you’re going to have to start removing some links.
Yes, that means cold-calling/emailing the people that are linking to you.
That means going into all the properties that you control and removing the links.
That means a lot of tedious, heart wrenching work. (Which is why you hire minions to do it for you!)
The bit that EVERYONE misses
When you’re finally done with all this work, does your site automatically bounce back?
Unless you do this ONE last thing.
You see, when you remove a 1000 links from your website, it sends a negative signal to Google that indicates that your site is dying. Google tracks your link velocity to see if your site is gaining or losing links.
Negative link velocity
If it sees that you’re gaining links over time… then your site must be good, healthy and active. Great!
If it notices that your site is losing links over time, then it assumes that your content is no longer relevant and webmasters are removing the links. Bad!
So, if you remove 1000 links, you must compensate by receiving at least another 1000 links to maintain a neutral link velocity. (And ideally, you want more so you have a positive link velocity.)
It’s only when you rebuild the links to account for the removed links that your site will thrive again.
Positive Link Velocity
So hopefully this explains why you might have removed a penalty but never recovered. It pains me to see that most webmasters do all the hard removal work and are so close to success… yet miss that crucial last step of rebuilding the missing links.
Let me know what you think on our Facebook post!