Negative SEO – The 800 Pound Gorilla

There has been some talk lately of whether negative SEO exists or not – something that rears it head now and again. Google say it's incredibly rare, not sure how they’d know this but I guess it depends on how it’s defined? Link spamming, scraping etc. or even hacking cracking?

Here's a little quote from Matt Cutts, he says here…

One of the big reasons we say softer things (rather than “it’s impossible”) regarding the idea of negative SEO is that we have seen people do pretty crazy things to steal/hijack domain names in the past, like the bizarre history of sex.com: http://en.wikipedia.org/wiki/Sex.com

People talk about negative SEO far more than people actually attempt it, because you’re never quite sure what effect (say) pointing some links to a site might have–it might help the site instead of hurting it–plus it’s typically a better use of your time to develop your own site.

But if there’s a site that is worried about negative SEO, the site can disavow any links they want using our the disavow tool in Google’s Webmaster Tools. You can even sort to see the most recent links if you’re worried that this is something that just started.

The primary usage of the disavow tool is so that a site can disavow bad linking that the site did itself and can’t get removed from the web, but the disavow tool also works fine to disavow links that you’re worried might be spammy.

This is over a year old and is actually an explanation of the content changing over on this page which used to read ‘There’s almost nothing a competitor can do to harm your ranking or have your site removed from our index…'

Again all this is fine and of course old news but what are we classing as negative SEO? Rand Fishkin recently wrote about a possible experiment to test the effects of ‘negative SEO' whereby he suggested…

A sacrificial website (and a generous owner who's willing to risk losing their search traffic)

A collection of spam links that's known to remove sites from the search results

Funding to purchase said links.

I personally wouldn't start with building links to a site, instead I'd start with quite the opposite, I'd manually audit the sites links and try to figure out why the site is ranking and pull the equity, I'd then look for potential issues with the site and subtly amplify them; I'd look to take advantage of any kind of duplication issues including, internal site search etc.

Back to the links, we've already mentioned our own successes with removing links simply using a gmail account and Dave took link takedowns a step further but if we really wanted to increase the likelihood of links successfully being removed we could simply register a domain that looks similar to the target domain (wikipeadia.org for example, note the added ‘a’) and replicate the site for social proof and use that as the email – to be honest from my own experience people hardly ever even reply, I assume from being inundated by removal requests, in which case you could just pose as the company and set the email to the real one.

(Any excuse for a little Python!)

import smtplib
 
user = 'user_name'
pwd = 'password'
 
to = 'target@company.com'
frm_whoever = 'imitator@company.co.uk'
 
smtpserver = smtplib.SMTP('mailserver')
smtpserver.ehlo()
smtpserver.starttls()
smtpserver.ehlo
smtpserver.login(user, pwd)
header = 'To:' + to + '\n' + 'From: ' + frm_whoever + '\n' + 'Subject:Link Removal \n'
msg = header + '\n please remove this link... \n\n'
smtpserver.sendmail(user, to, msg)
smtpserver.close()

Notice the frm_whoever variable!

This of course will have the biggest rewards but at the highest risk of being caught. Above I mentioned about webmasters being inundated by removal requests, this is actually the reply I usually get when doing a legitimate link clean-up for a client; an auto response saying exactly that; they’ve been inundated and it will take a few weeks to remove the links. Which again brings me back to the question, ‘how would Google actually know?' Sure a link blast with XRumer or whatever stands out a mile; suddenly a site gaining a few thousand links with spammy anchor text overnight – We’ve also seen this still working, but removals?

Negative SEO is often just seen as lowering a sites rankings but what about where a site simply outranks another through taking its content? You hear all the time about webmasters being outranked by scraper sites and the fact is it's very real. I recently picked up a dropped domain that had links from the BBC, Guardian etc. including a .gov domain and 301'd it into a scraper site that I had set up (as an experiment), a few weeks later I was outranking the original sources of content.

The next stage (if I wanted to) would be to keep pushing links into the site, one of Dave’s points, to look out for here was common-or-garden PR distribution services, the spin on this would be, the article text to use rich anchor text to the target site and brand links to scraper site.

The next stage after stealing the traffic would be to get the scraper site penalised and then 301’d into the target domain and then start the process all over again.

Here’s a look at the traffic…

Here is the organic breakdown in case you were wondering, yep all Google.

Speaking to Dave to see what he would do (we all know he loves a bit of blackhat), he pointed me towards bbc-now.co.uk, this site is essentially just a proxy, you can see how it duplicates the bbc.co.uk, it’s even indexed within Google.

Matt said recently at SMX that the Paid Day Loans update incorporated algorithms to prevent negative SEO attacks, so how come 1,390,000 pages are being indexed from bbc-now.co.uk?

This of course is causing massive duplication issues, this probably won’t have much effect on the BBC, if any at all, but take someone smaller, and you could really ruin someones day. You can also look for issues with IP addresses and wildcard DNS configurations, that allow you to create duplicates.

bbc-now.co.uk was on the winners list last week in search metrics – Note: we are just using this as an example of something that could cause duplication issues, we don’t know why the bbc-now domain was actually set up.

So what do you do if you are getting links thrown at you? Well I'd just disavow them and then try get to the source of who is doing it and inform Google – no easy task. You should be able to detect links coming in fairly easily using webmaster tools.

Removals on the other hand, if done with tact, you might not know until it is too late. The best defense you can have is knowing exactly what is in your link profiles and monitoring them regularly – ideally programmatically and if a link goes down, find out why and do it quickly. If you are being ‘attacked' email and phone all your linking sites and tell them what is going on and that any such requests should be ignored.

If you are being scraped and outranked by sites using your own content first contact the website and ask them to stop/take it down etc. if that doesn’t work then file a DMCA report or try removing content via Google’s ‘Removing Content From Google‘ feature.

Personally I think Google does a pretty good job, when you consider the scope of the interwebz but negative SEO as far as I’m concerned, with Panda and Penguin manipulations is an 800 pound gorilla, it’s very real and not going anywhere, anytime soon.