Google Caught Buying Links: Bruce Clay’s Analysis
|Estimated reading time: |
Video view time:
• Google’s Chrome browser division was outed for buying links, an SEO ranking tactic typically eliciting ranking penalties by Google.
• The penalty Google levied seems lenient in light of penalties frequently imposed on other sites in similar circumstances.
• This event highlights the ease with which sites can be attacked by those with ill intent. Bruce proposes a tag to allow site owners to disavow inbound links.
Hate to say I told you so… scratch that. This time it feels good.
Last month we let you in on the new direction of SEM Synergy: short-form video with an experimental format (Q&As, site reviews, couch-side kick backs). Today we’re happy to present to you the first video episode of the SEM Synergy Web series.
Topic du jour: Google instituting a ranking penalty on its Chrome browser after paid links were uncovered.
Find the transcript for the video below, and check out the Bruce Clay, Inc. YouTube channel for a look at our still evolving platform for the video series. If you like it, come back tomorrow as we dissect the process of producing and publishing video content. As we learn, we plan to share our experiences with trying to create a sticky and interesting show. To that end, I do my best Larry King impression as we dive into a hot and timely issue.
Transcript: Is Google’s Penalty for Chrome’ Paid Links Fair?
Virginia: Good evening. Paid links. Google’s public enemy number one. An SEO tactic warranting serious penalties. So what happens when the search giant gets caught red handed? Were the self-imposed consequences fair? What will the long term fallout be from this reversal of fortunes? Joining me in the SEM Synergy studio today are search industry veteran Bruce Clay, senior SEO analyst Maryann Robbins and software development engineer Michael Terry.
Bruce, what has been the most troubling aspect of this event coming from a law abiding, white hat SEO perspective.
Bruce: I think the biggest problem that everybody faces is that people who do not have quality content, that have not developed a website appropriate to ranking can go pay money and rank. You have a tendency to get at the top. I think from Google’s vantage point, it’s the same as a white hat’s vantage point, if they don’t do something about it pretty soon, whoever has the most money to spend on links is going to own the ranking.
Virginia: So is that what really bothered you about this situation?
Bruce: Yes and no. We’ve had people come to us who have seen penalties as a result of it. That doesn’t bother me that there’s a penalty. I think I’m bothered more that they’re so difficult to detect and so many people are still getting away with it.
Virginia: Maryann, you’ve worked with clients who have suffered from paid links penalties before?
Maryann: Yeah, I had a major client that was affected by paid links by another division in their company. And they were not responsible for the purchased links; it was another division that they had no control over. And they were, as a site, penalized for that link and it was huge massive result in lack of rankings for that division that was not responsible for it. And it took a lot of work to get rid of that penalty, to get back into Google’s good graces. And the other division had to remove their links for them to regain the rankings for that major keyword that they were not buying links for.
So, with the situation with Google in this case, they have at least penalized themselves a little bit for having purchased links. But they didn’t give themselves, maybe, as much of a death blow as some other sites have gotten when they have been caught. At least in that situation they are sending a signal that if you are caught purchasing links you will be penalized. But they didn’t go as far as they do when they catch other sites purchasing links.
Bruce: And even their prior history with themselves – Google Japan was buying links – mostly because in Japanese it’s difficult to optimize on page. And Matt Cutts actually at one point cut them off from a PageRank 8 to a PageRank 4 sitewide. Well, when that happened clearly their rankings were impacted. Somehow that didn’t seem to happen to Google.com. It was very, very localized to a particular keyword to a particular area, where it didn’t impact overall, globally, Google’s rankings. It was spotty.
Virginia: The penalty didn’t seem to be in parallel with previous penalties?
Bruce: Neither with their own penalty of a Google property or the way they have penalized clients.
Virginia: Michael, it seems to reveal vulnerabilities in the PageRank system.
Michael: This is making it a little higher profile, the concept that if a bunch of links coming from low quality sites point to another site, it can be penalized. And in the circles that I run in, this might not have been as well known, and there are a lot of hackers and security testers, people who would like to expose flaws in the system. And what’s to stop people from creating giant networks of cheap, low-quality sites that just scrape from other sites, and then pointing links to people that you want to hurt, either for profit or for fun. A lot of people like to do these things just for fun.
I really think that there wouldn’t be a great way for Google to detect these things and I feel like there should be a feature. Bruce has brought this up before and I think it’d be pretty easy to put something in Webmaster Tools or what have you that allows you to disavow links you don’t want pointing to your site and hurting you.
Virginia: You’ve talked to people at Google about a nofollow from the other side.
Bruce: Yes, and in fact I think it was at SMX Advanced, Matt Cutts on the podium, while he was up on a panel, actually did a show of hands of people in the audience who would like to be able to disavow inbound links. The problem, clearly, is that it is possible for a competitor to go buy links that come to my site and for Google to not know who bought them, and therefore they could penalize me for that paid link. It’s not fair for me to get a penalty when I have no way of turning them off and I have to demonstrate to Google that I’ve fixed the problem and stopped being evil when I was never evil in the first place. So, having an effective “I can nofollow it from the receiving end of the link and have it not hurt me,” something like that would be a very, very good thing. It has been discussed and Matt has indicated it is somewhere on the list.
Virginia: Sounds like for today, we can rest assured that Google is penalizing itself and has sent the message that paid links won’t be tolerated. What it means in the future for further features and technologies remains to be seen. Thank you, guys, for spending time with us today.
Do you have any thoughts on Google’s paid links policy? Was it an appropriate deflation or is there just no real justice in a world where Google is judge, jury and executioner? Leave a comment or video response. And look to the Bruce Clay blog for recommendations for recovering from a paid links penalty. This has been SEM Synergy with Bruce Clay, Inc. See you next time.
23 Replies to “Google Caught Buying Links: Bruce Clay’s Analysis”
“Virginia: Instead they’ll have to settle for devaluing low-quality, irrelevant links – which isn’t really settling if you ask me, as it improves the end user experience, and that really what it’s about afterall.”
Virginia, I think you have hit it square on the head with this comment. From an automated standpoint, web crawlers can’t truly identify “paid links”. Instead, they need to check for relevance and quality.
For a search engine to be able to identify “paid links”, they have to be “subjective”, they have to consciously decide, “This is a paid link.” It’s this lack of objectivity that creates a problem like this:
“Bruce: And even their prior history with themselves – Google Japan was buying links – mostly because in Japanese it’s difficult to optimize on page. And Matt Cutts actually at one point cut them off from a PageRank 8 to a PageRank 4 sitewide. Well, when that happened clearly their rankings were impacted. Somehow that didn’t seem to happen to Google.com. It was very, very localized to a particular keyword to a particular area, where it didn’t impact overall, globally, Google’s rankings. It was spotty.”
“Maryann: …And they were, as a site, penalized for that link and it was huge massive result in lack of rankings for that division that was not responsible for it.”
“Bruce: …And Matt Cutts actually at one point cut them off from a PageRank 8 to a PageRank 4 sitewide. Well, when that happened clearly their rankings were impacted. Somehow that didn’t seem to happen to Google.com. It was very, very localized to a particular keyword to a particular area, where it didn’t impact overall, globally, Google’s rankings. It was spotty.”
Bruce, what do you think about this thought: Instead of penalizing an entire site, why not penalize the page the paid link is actually pointing to? If Google can identify paid links and penalize sites, they certainly know which pages those links are pointing to.
“Bruce: …It’s not fair for me to get a penalty when I have no way of turning them off… So, having an effective “I can nofollow it from the receiving end of the link and have it not hurt me,” something like that would be a very, very good thing.”
I agree, it’s not fair. Especially when it appears the penalties are often subjective. I also think it would be nice to be able to nofollow incoming links.
I just think this entire story is the funniest i have ever seen, google penalizing itself…. you couldn’t write this stuff!!!
They have to be seen to be doing this, but at the same time most sites get away with paid links, so its all a bit nonsense…think i will have to blog about it!!
Love the video by the way thou!!
To me that xfactor page is a useless link, and in my opinion borders on a link farm. Because its many links all with no common niche, not to mention its only using the domain name as the link so there is no real SEO value there either, the juice is watered down by having several hundred other links on the page. The way links work for you is to only have 1 or 2 on a page, more then that and it takes away from the over all value of even having a link, unless all links o the page point to your site.
That xfactor page is similar to how some people set up listing directory pages, but the directory pages that offer real value to users offer tons of unique content and information not a small snipet on a page with thousands of other small snipets, but rather having an entire page all to that one listing.
When I am building links I always imagine writing similar to a wiki, and the links within the content lead to more useful information, by doing that the search engines thank you by ranking you higher, the user thanks you by staying on the page and hopefully converting, and every is happy.
To answer your last question “how is it different”. Well technically it isnt different, if you pay for a link you pay for a link, if someone agrees to give you a link back for retweeting their tweets, that is still a form of barter, there is no way to get around the fact that people will pay for links, what I think google is a doing a decent job at is showing the most relevant content not only using link text, but using on page content and many other factors.
So lets say you have a page that is not optimized at all and you build hundreds of links to that page, chances are good google is going to look at all those links and the irrelevancy of your content and punish you by not ranking you. The reason they targeted the Link farm or irrelevant links was due to them offering no real value. whereas the links written into the content leading to more relevant unique content were skyrocketed because of the value they add.
Aman, the sound quality is poor and we’ll work to improve that going forward. Regarding my demeanor, I haven’t seen Larry King smile much… I guess no one got the bit. :/ Feedback noted.
Speaking to the larger discussion, I think we’ve gotten to the heart of the issue here as I see it. Paid links are links obtained through exchange of money or monetary value. That a link is paid for doesn’t automatically mean it’s invaluable to users. In Google’s ideal world they’d eliminate paid links altogether on principle because it manipulates their PageRank system. Instead they’ll have to settle for devaluing low-quality, irrelevant links – which isn’t really settling if you ask me, as it improves the end user experience, and that really what it’s about afterall. :)
Google’s policy seems to be that any link that is paid for in the form of an advertisement, must be no follow, or have some similar attribute so it doesn’t pass on link juice. However, what about product review blogs, or ‘our sponsors’ pages on sites. Generally these are people providing either cash, or a free sample of their product, and getting links through these pages.
Just check out the X Factor’s suppliers page:
A whole list of do-follow links, because they’re suppliers, sponsors, friends of the show, whatever. At the end of the day, they have that link because they provided free products. How is this different to phoning a webmaster and offering him $200 for a link?
Yes it is certainly true that matt has spoken out against any link other then, something like a simple link to content. The problem with just having people link to your content is they usually only use a domain name to link, which gives the search engine no valuation of what the link goes to. I have clients who had 500 linking domains and could not be found for the purpose of their business, because only the domain name was used for all linking. So google should not worry about people paying for links as long as the links are relevant, and the content is good.
So even if you have the most relevant and unique content, if you dont have links using a keyword you will likely not come up high in the SERP. I am sure google is well aware of this also, so what is a website owner supposed to do since people can spam black hat links all over the place and rank? Well I choose to use white hat and use relevant, quality content that helps the search engines understand what the website is all about. But that doesn’t change the fact about what the algorithm determines is a paid link, which is links not surrounded by content. Until the search engines can differentiate between quality content and crappy content using some form of artificial intelligence, there is no way for the search engines to sort this problem out. I am sure google is working hard to create that calculation. The proof is in the fact that google needed to manually spank itself, rather then having the algorithm handle it.
Sidebar: Matt Cutts has repeatedly cited companies as spammers that purchase paid links disguised as content. These are links within the content that exist as the result of a monetary transaction or trade. So a paid link can and often does exist within content, and a paid link can be monetary or barter. We know of penalties where companies traded product discounts or cash in exchange for links. I really think that in an ideal world a link is earned (testimonial) but in too many cases the links come to those who can pay for them. If Google does not fight these paid links, however disguised, then they will concede the first page to those with money instead of those that are deserving. Saying it is not a paid link because it is in content is misleading. And saying that it is paid only if it is for money is misleading.
Virgnia, you need to learn to smile more. Yah you were way too serious in the video. Also is it just me or I can hear background noise throughtout the video.
Sorry for the name change it seemed so informal.
Well spammy links is quite a broad term, you can have spammy links that are just plain text, for instance a lot of irrelevant text links, or a link farm. Spammy content is when you write one article and post it to sometimes many hundreds of websites, this is what I would characterize as spammy content. Then there is a good way of buying links, which is write some relevant and unique content about the product or services and pay someone to publish it on a single relevant niche website, which has the ability to generate its own backlinks, or link bait. And the article can have 1 or 2 relevant links within the content pointing back to the more information on the website. Now there is a white hat and a black hat, I prefer the white hat. This is similar to what google did, except they did not offer quality content, but paying to post good quality content should never be punished. Think of like this, if your trying to help google, by creating quality content then its ok, if your trying to trick google then its a bad thing. My goal with SEO is to make my clients websites the most authoritative for whatever the niche is, and then promote it.
At this point, I think we basically agree, Chris. We definitely see eye to eye on the above examples of spam. And certainly the goal of SEO is to obtain prime search engine rankings via the promotion of excellent, relevant content. :)
Absolutely not that would be a major fubar on their part if they were to let that info out, they would prefer to let everyone think that a paid link is when you pay for a link, this is why they manually punished themselves. So how did I come to this conclusion? 2 ways first I monitored what I considered to be link farms, I found that the majority of the link farms themselves and the websites “consistently associated across the link farm network” consistently, were punished. While at the same time building links inside relevant unique content, posting it to niche relevant sites with good link flow and watching the SERP skyrocket, even from virgin domains.
The second way was I recently found a genius who reverse engineered the main principles of all search engines, basically wrote an algorithm into a program that shows in reverse what the search engines view is and where the penalties are being incurred, what the market focus is for each page, plus tons more incredible information.
And its actually quite logical if you think about it, if a link is simply located on a page outside content, and that page is not relevant to either the link text, or the link destination, then that is a paid link, not meaning you bought it, but it is the assumed conclusion because why would a website just put an unrelated text link for no reason? It also gives them the ability to write a rule and implement that rule. This also applies to internal linking to a lessor degree.
The search engines have 2 core calculations, onpage penalties and off page penalties, I believe the onpage penalties for paid links are not as significant as the offpage. Ie… nav links are not going to be punished, but a non contextual or paid link will be a significant punishment. Just thinking in a logical manor.
Boston, how does your criteria for paid links differentiate between spammy links and paid links? My reading of your observations in part one would identify spam links, which I think we’d agree Google wants to exclude from its calculations, if not penalize. And I’d agree with the second point you make as an indicator of a paid link, a paid link footprint of sorts. However I wouldn’t, in my own writing or in talking to clients, define a paid link as any text link removed from context. I think it confuses paid links and spam links, whereas I think paid links are a subset of spam links.
The definition I used is the one search engines use.
I realize when people see the term “paid link” the only thing they can associate that with is monetary. But from a search engines perspective a “paid link” is a link that is not surrounded by content. Link farms are pages with text links stuffed at the bottom usually, but generally text links not surrounded by other content is what google was trying to eliminate. Search engines feel that links within the content, leading to more content on the same subject is far more valuable to users. This is why so many websites that used link farms as an easy way of getting links have been punished, I had a new client come to me because they were devalued from a pr4 to a pr0, and they wanted to know why. After just a small amount of investigation into the linking used by the previous company, it was very easy to understand why. None of their links where within the content, all were just text links from link farms, which the previous company told them was the best kind to have.
This is also why they(google) had to manually slap chrome, if they did not then the “SEO community” would have cried for ever,like a bunch of whiny babies because they do not understand what a “paid link” actually is. The links google bought where not by search engine definition a “paid link” because it was surrounded by content(arguably not good content)but still content. So in order to keep the misunderstanding of what a “paid link” is google punished itself.
You have to realize its impossible for a search engine to determine who gave money for a link, but they can determine what they feel is the most appropriate way to link, in order to offer value to website visitors.
Boston, that’s an intriguing idea. Can you cite a source from a search engine employee suggesting paid links are simply links separated from content? That would also imply that many types of navigational links would be identified as paid links.
Bruce mentioned that individual pages and queries were devalued. I think it’s important to point out that especially over the past few years ranking is especially strong on an individual page-to-page basis rather than so much site wide. The reason I suspect they did this was to reduce the influence link spam online and making it a lot more work for link builders/spammers so that they now have to make relevant contextual IBL to individual pages. This helps in increasing relevancy but only increases the magnitude of spam online.
I believe one of the major reasons that in the last few years the keyword rich urls have taken off is the result of other site wide variables being devalued. We’re starting to see the QDF (Quality Deserves Freshness) over the past few years and link velocity becoming increasingly more important. Good engagement like the BCI blog will continue to be an important variable in ranking sites into 2012. Internal link strategies and refreshing old content with links in new content much like Michael Gray can often be seen doing will also be quite important in ethical link strategies.
Adam, your theory about why rankings are more page-specific than sitewide makes sense. Thanks for sharing that. Seems Google’s approached the problem of link spam from several directions, which seems obvious when I write it out loud :P Definitely highlights the increasing importance of ethical link building.
You’re so serious Virginia! You’re like a news anchor reporting a murder. I guess that’s the kind of tone Google needs!
It is my understanding that a “paid link” is a link that is not surrounded by content, this includes internal linking, you are penalized for links not surrounded by content within your own site. Not so much about paying for a link. Which is why link farms took a serious hit, and a lot of sites that used them took a hit after the panda update, it did a great job. But the reason google penalized themselves is so they could continue to keep the confusion about what a “paid link”actually is. I really like Bruce’s concept of being able to “no follow” from the receiving side. But I don’t think they should have imposed any penalty on themselves, other then firing the company who put up such crappy content. I believe the search engines are working hard to figure a way to understand what is good content and what isn’t. This way they will be able to penalize poor content more efficiently.
Hi Boston, I can definitely identify with the desire for a nofollow from the receiving side from a webmaster’s point of view – with so much business relying on Google rankings, not having control over such potentially damaging behavior is stressful and arguably unfair. I think your definition of a paid link is off – at least here, when we talk about paid links we’re talking about links that were obtained by a trade of monetary value. The Panda update targeted low quality content who’s value in the Google algo was being inflated thanks to link spam (among other tactics); Panda wasn’t about paid links in particular. Chime in if anyone disagrees :)