Can Competitors Harm Your Rankings By Building Spam Links To Your Site?
A true story and case study of multiple long-term negative SEO attacks…
Now, there are many who believe you can use those penalties against a competitor…
A New Sabotage Method is Born (sort of)
Google’s updates, penalties and filters, particularly the Penguin algorithm update, have some webmasters convinced that someone could build lots of spammy links to a competitor’s website to make it look like they have violated Google’s guidelines, and cause the site’s rankings to drop off the search engine results completely. Is it possible? Maybe, but effective “Google Bombing” is not very likely as Google’s Matt Cutts explains in this video:
Despite Google’s insistence that there is no serious issue, fear mongers seeking attention, companies who sell sabotage services, and webmasters who want to put the blame on something other than their own actions still believe – or want YOU to believe – that it is easy to destroy a competitor’s rankings just by pointing lots of bad backlinks at the site. This is despite the lack of definitive proof that link based negative SEO works, or even any examples of NSEO having any effect at all.
A few reasons why “NSEO” is not likely to work against an otherwise legitimate site:
- It would take a lot of work for an attack to be successful. People who would do such a thing tend to not work hard and would find that it is just not worth the effort.
- A site with a strong, natural link profile would be able to carry a relatively small number of bad links without a problem.
- Google is (theoretically) smart enough to know the difference between a “blast” of spam and a site’s normal link profile.
These reasons could mean that a new site with not much of an established link profile (as in, not a big name like Moz.com), could in fact take some damage if someone pointed a bunch of links spam directories, irrelevant sites, link farms and other questionable sources to the victim’s site. Still, it isn’t likely to work unless maybe the site already has links that ibring it to the edge of a penalty.
How do I know?
Starting in February or March of 2013 (edit: turns out it went back to more like October 2012!), someone really started going out of their way to cause problems for KerCommunications.com and it had some interesting effects, none of which were a massive drop in rankings. Here’s an overview of how it all went…
Our Link Profile
Prior to the attack on this site, it had incoming links from maybe 300 or so different domains. Most of these were legitimate, though to be honest, there were a few “questionable” links. A few not-so-great directory choices, and a handful (less than maybe 10) blog comments that were signed with something other than a name or company name. Nothing worthy of a penalty, and I would say that none of those had much of a positive effect on the site’s rankings. So the site had a smallish number of incoming links for an SEO company, but mostly clean. Those who believe there is an epidemic of linkspam sabotage would probably see the site as an easy target.
Before this accidental experiment, Ker Communications ranked in the top 5 for several localized search terms like “Pittsburgh SEO”, and quite a few non-local searches like “guaranteed search engine optimization”. That one is kind of funny because we don’t offer that and the old blog post that ranks for that phrase is very anti-guarantee. Most of our traffic does not actually come from those local keywords – more (and better) search traffic comes from more “real language” searches than simple, two keyword searches (more on that here).
Why Did This Happen? Who Would Do Such a Thing?!
Maybe it was one or more competitors. In the Fall of 2012 into 2013, the site began to rank higher for a lot more of those competitive local keywords, often sitting at number one. It is possible that some of the local shady and/or incompetent SEO providers decided to try to sand bag the site. Since they couldn’t continue ranking with their own spammy ways, they could try to bring Ker Communications down to their level. There were also several attempts to login using my name and the names of some people who work for or with Ker Communications, which would suggest it was a bit more personal than a random hacker. I’d like to think that most of the competition in Pittsburgh is above that, but it is very possible that it was a local SEO.
I think it is more likely that I brought this on myself, in some ways. I have been active in various forums and discussions on sites like SERoundtable where people complain about how they feel that Google is out to get them, wants to destroy small businesses and a slew of other whines about how it isn’t easy to make a crappy site rank higher than it deserves. When I see someone spouting pure BS where other people can read it and make bad decisions based on misinformation, I like to try to set them straight and present some rational thought when I can. There were a handful of people who had been ranting about how Google has intentionally made NSEO as easy as buying a few Fiverr.com gigs. When I pointed out how ridiculous that was for a number of reasons, some even threatened to prove it works by targeting my site. So it could have been that, too, though I had not poked any of those hornet nests with a stick in a few months when this “attack” started to happen. But then, people who rant irrationally can do some pretty ridiculous things – like this guy, “Alan” (who deleted his account and is now “Guest”) who made a bunch of slanderous claims about me, threatened to go on a link spamming spree through Fiverr and thought that everyone who disagreed with his rants was really just me in disguise. I probably should not be surprised when people who don’t know how to behave like a decent person in forums or comments, also do not know that trying to sabotage a business to prove a point is just wrong.
Starting in March of 2013, a huge number of spam links started showing up (EDIT: It turns out the earliest link spam attack started in September 2012, but was relatively small and unnoticed). At first, there was a surge of about 30-40 links per day which were mostly comment spam, pingbacks – the low quality type of things you can get on Fiverr.com. Unexpected links from scrapers, spam blogs, or even low quality directories are not out of the ordinary, but suddenly seeing comment spam with keyword anchored links when there was none of that before got my attention.
After a week or two, those links stopped appearing in ahrefs.com and majesticseo.com, so it looked like a failure. About a month later, similar links started showing up, this time with a wider variety of spam including more comments and trackbacks, spun articles on spam blogs, scraped copies of the site using appspot, links on social bookmarking and wiki sites nobody ever heard of, and probably some things that are so shady that the link tracking sites don’t even see them. This time, the numbers were more like 1000 new links per day. Many of these were site-wide, like in a recent comments sidebar, so the number of actual sites was more like 100 to 200 per day.
It appears that the perpetrator went to SEMrush.com or a similar site to find out what keywords for which we ranked well, and tried to spam the hell out of those in an attempt to get a penalty or unleash the Penguin. When this all started, those keywords were things like “Pittsburgh SEO” “Pittsburgh SEO consultants” “guaranteed search engine optimization” “SEO company Pittsburgh”. After a week or two, links using slight variations of those keywords started to appear, too. One strange thing to note is that there was also a sharp increase of links using just the plain URL, the company name, and sometimes my name. Just a guess, but it looks as if this was the byproduct of a low-end link spammer who was hired to do the dirty work. But they decided to “help” their client by trying to diversify the anchors, rather than using nothing but the keywords that were specified.
I guess when you hire someone to do online business sabotage, you still get what you pay for and getting it wrong what a $5 link builder will do.
Another larger wave of link spam started a few weeks later, this time targeting not only the home page, but pages on the site that previously had no external incoming links. There are some neglected blog category pages, and general info pages on this site that didn’t have or even really deserve to have any links. Perhaps the perpetrator thought that those were secret pages and thought that it would cause problems if they appeared in the search results.
The malignant growth on the link profile continued to swell with new types of link spam every week until around the end of May 2013. Then, the new links virtually stopped appearing. Since many of the sites involved were abandoned, or the webmasters figured out how to remove the pingbacks and comments; the number of incoming links to KerCommunications.com began to shrink to something closer to normal. Some spam links continue to appear in the link trackers, but those seem to be older links that are just now being crawled and discovered.
EDIT: After publishing this article, the link spam attacks got worse, and more hacking attempts and DOS attacks on the host servers started happening more often.
Does Negative SEO work? Not the way you might think it would!
Thousands of spam links later, the big question is, “what effect did all of this have on rankings?”
As of today, none. Well, nothing noticeable anyway. No “unnatural links” warning in Webmaster Tools, rankings are nearly the same as they were back in February, traffic has also grown at around the same rate it was growing before.
There was a brief period of “flux” around the time of the Penguin 2.0 update, but that also coincided with some changes made to the site, and the loss of a few good legitimate links, so I can’t attribute the ranking turbulence to the link surge, Penguin, the loss of good links, or even the changes that were made to the site since they all happened around the same time. A list of the spam links was submitted through Google’s Disavow Links Tool just a day or two before Penguin 2.0 hit, which also could have contributed the fluctuations while Google devalued those and reevaluated the site’s real links.
Why didn’t it work?
I don’t have any definite answers, since I don’t work for Google or profess to know exactly why they do what they do, but I have some good ideas. Apparently, tricking Google into thinking a site is bad is not any easier than tricking it into thinking a site should rank number one.
- It didn’t look like business as usual for the site. With just a few hundred domains linking to KerCommunications.com over the past few years, rarely growing at a rate faster than maybe 5-10 new links per week (more is not always better), even a dumb algorithm could figure out that this surge of bad link building was not typical.
- Keyword choices. The perpetrators went after keywords for which the site already ranked highly. Why would we spam those if we already ranked well for them without spam? Prior to this, there were very few links using those keywords as anchor text. Why would anyone suddenly start spamming for high ranking keywords this way when it was not needed? Even if the algorithm couldn’t see that, I am certain even the sleepiest manual reviewer at Google would see it as a frame job.
- It just looks like a NSEO attack. Even if the previous two reasons did not apply, it seems likely that other sites have had almost the exact same pattern of spam links – especially if the attack came from one of the services that have been promising to destroy your competition for just a few bucks. Chances are, they automate that process and automated SEO is rarely effective.
Google really is smart enough to know the difference
While this campaign had no significant impact on rankings or traffic. It has been a nuisance.
- Link profile is much harder to analyze. With all that spam taking up space in the charts and graphs, it is more difficult to determine what links have had an effect on rankings and traffic.
- A little stress. While I have been confident that malicious link spam from third parties is nothing to be overly worried about, such a huge increase in the number of links did make me a little uneasy at times.
- I can no longer smugly point to this site as an example of how a site can do well without a large quantity of backlinks.
Maybe I am just a glass half full optimist, but I have seen some very positive effects from this.
- Link Profile is now obfuscated. Webmasters and SEOs of new sites will often look at the competition’s link profile to try to find places to get good links. Competitors who want to try to get the same links as KerCommunications.com will have a much harder time figuring out which ones are actually good. I have already seen evidence of a few newer competitors who are trying to get links from some of the shadier sites involved with the attack, and they may even be trying to buy spam comments and bulk social bookmark and wiki links.
- Some of those previously unlinked pages on the site are now getting some traffic that is actually converting. Thanks! Also getting some direct traffic from some of the spam links, but that is probably webmasters wondering why they have all those pingbacks and trackbacks.
- Learned about “the dark side”. I see everything as an opportunity to learn. From this, I learned more about some types of spam links to look for when helping clients deal with penalties from previous SEO mistakes. I also became very familiar with link monitoring tools from ahrefs.com, majesticseo.com and linkresearchtools.com.
- Made some new connections and even got some new clients. The comment spam prompted some bloggers to write or even call to ask why they kept getting these notifications. I took the time to explain what was going on, and when possible helped them clean up all the other comment and pingback spam on their sites and prevent it from happening in the future. As a result, there were a few who hired me to do some other work for them. I don’t think an increase in sales was the goal of this attack.
Are Unwanted Links Something to Worry About?
To be perfectly clear: This is not a denial that link based sabotage is possible.
Like Cutts and others have said before, it is possible, but extremely unlikely to happen in a way that would do permanent harm. This attack went on for months against this relatively small site using a variety of methods, and it failed. Doing NSEO effectively may be just as time consuming and difficult as doing legitimate SEO. If someone is out to destroy your website, do they really have the time, resources, or enough malice to make it happen? If they hate you that much, they will likely find a more effective way to mess with you than Google bowling.
I help many sites deal with ranking drops due to penalties and algorithm updates. Some of the website owners come looking for help with a ranking drop they believe is caused by a surge in maliciously built links. More often than not, the problems come from irrelevant spammy links they built in the past, or the work of an SEO company that chose to go against the guidelines without the client knowing about it. Nobody really wants to look at their own site or actions as a cause, and it is easier to blame Google’s Penguin or a competitor, but I haven’t seen a real effective case of linkspam sabotage yet. An analysis of every suspected attack I have seen has revealed a number of other issues that had nothing to do with the malicious competitors. Telling people things like that may be what provoked this attack to begin with, but it is the truth.
UPDATE November 2013: One Careless Spammer Caught!
Scroll down to the comments to see how the most recent attacker, Paul Rone-Clarke (AKA “Scritty”) known for “Ultimate Demon” link-spamming software, was caught almost immediately. Makes a nice example of how search engine sabotage can work against the perpetrator.
And a little more about it on Google+.
More to come…