Google's Matt Cutts: Definitive Criteria For Paid Links

[engine]

This seven minute forty second video from Matt ought to help those that have doubts over whether it's a paid link or not. Money changing hands is obviously a paid link.

Some of the other examples Matt gives are less obvious, but could fall foul of a "paid link."

[netmeg]

What if I link to a site for no payment, and use a commercial keyword to link to that site?

Again. Think patterns.

[n0tSEO]

I agree with Edge and marketingmagic, and I totally understand ColourOfSpring's dilemma, as it was my dilemma until a short time ago.

What's most exhausting about Google is that they tell webmasters to think of their website and users and not about Google; then you do that, and they penalize you because you didn't follow Google's guidelines. You see how gnarled the entire matter gets..

In the end, if you want to do SEO for Google, you have to compromise and follow the giant's 'whims'. On the contrary, if you want full control over your website and the marketing tools you choose for your campaigns, follow your plans no matter what Google thinks. :)

[rish3]

Just do it like the big brands do.

- Throw up a press release about a "Media Partnership", "Co-Marketing" with whomever you're buying links from. Then link away.

- Create an entity/site in your niche that appears to have no direct ties to your site. Like a "Widget Conference", "Responsible Widget Foundation", or "Widgets Standards Body".
Then create a bunch of worthless white papers, "expert opinion" columns, and "interviews" that your company has graciously given to this new entity. Naturally, there's a crapload of backlinks in them.

- (About 10 more strategies like this that are nothing but paid links, except that big G is fine with them)

Once you throw a few brand-licious labels on it, they are no longer paid links :)

[ColourOfSpring]

Just do it like the big brands do.

- Throw up a press release about a "Media Partnership", "Co-Marketing" with whomever you're buying links from. Then link away.

- Create an entity/site in your niche that appears to have no direct ties to your site. Like a "Widget Conference", "Responsible Widget Foundation", or "Widgets Standards Body".
Then create a bunch of worthless white papers, "expert opinion" columns, and "interviews" that your company has graciously given to this new entity. Naturally, there's a crapload of backlinks in them.

rish3, this is precisely the problem - Google can only measure brochures (websites), not products or services. Yet for commercial searches, people want the best products and services. Big difference.

[EditorialGuy]

Google can only measure brochures (websites), not products or services. Yet for commercial searches, people want the best products and services. Big difference.

Or the best prices, or the most convenient shopping experience, or the greatest peace of mind.

But yes, Google Search is about what's on the site or the page, not what's on the warehouse shelves or the overnight FedEx flight. In theory, at least, this should present an opportunity for a shopping engine that's more about businesses than Web sites--or for Google, if it ever decides to create a clearer demarcation between "shopping" and "information" in Google Search. (Then again, Google may figure that simply placing more emphasis on ads in shopping results is the best solution. I'll confess that I've used Google's paid Product Search results myself when searching for prices and sources of name-brand, mass-market products.)

[mack]

I think we all need to remember that our organic search traffic is no more than a bi-product of Google's business model. It could be argued that Google is making other online advert placements less attractive, but that's for another topic.

One thing I often think about is where do Google get their data regarding sites that sell links? Well with a huge index of the web it won't be to difficult to spot patterns. That coupled with all the "disavow" files being uploaded should keep the web spam team busy for some time to come.

Mack.

[Robert Charlton]

That coupled with all the "disavow" files being uploaded should keep the web spam team busy for some time to come.

Mack, that was my initial take on it too... that Google was going to mine the data. About a year ago, though, we published this thread, about input that member pontifex had gotten from rom Uli Lutz, of the Google Search Quality team in Germany....

Best practices for using the Google disavow tool, confirmed
Feb 3, 2013
http://www.webmasterworld.com/google/4542078.htm [webmasterworld.com]

pontifex asked some very specific question about the effect of disavow on other sites, and got one particular answer that sort of changed my mind about what would be done with the data...

"Do not worry about damaging other people, that does not happen"

I haven't seen this corroborated, but I also haven't seen it denied. I can imagine that the data might be used as a confirmation of the data which Google already has. I don't know how to "read" Uli Lutz, though, but this is as close to an official statement as I've seen. Google needs to keep the secret sauce in the bottle, but is generally very straightforward in areas where they do make public statements.

Again, my take on this is that these are the criteria that Google is setting out for clarity, but, unless they can convert these into heuristics, these criteria are not themselves what provide the clues.

Regarding where Google does get data about link sellers, a simple search should indicate how blatantly these are being advertised. Money and greed tend to shape the patterns of the link marketplace, though, and the patterns, as netmeg suggested, are usually done to death.

[Robert Charlton]

PS: It's worth noting that at a session on Penguin at SMX West today, Eric Enge, when asked if Google used disavow lists to identify bad sites, essentially confirmed the above... ie, that Google has said that it doesn't currently use disavow lists to penalize sites and has no plans to do so.

He had no idea if this might change, but seemed to doubt that it would.

[mack]

Thanks for addressing this. I can see it making sense not to hit sites directly based on disavow. The simple act of discounting links from a page may have a long term algo effect. Not as a direct result of disavow, but it would certainly raise some automatic flags when a page has several sites requesting their out bound links be disavowed.

Mack.

[MikeNoLastName]

BUT, does a large proportion of or all 'nofollow' links on YOUR site make you look bad? Why not just nofollow everything offsite?

[martinibuster]

MikeNoLastName, you have a knack for bringing up interesting considerations! :)

One could no-follow all outgoing links. However outgoing links provide a signal of relevance for your site. Outgoing links on a page send a signal about what the page sending the link is about. So if the outbound link is discounted then you may lose that relevance signal. In my opinion you will lose that relevance signal because that's the purpose of it, as in UGC blog comments you may not stand behind.

What about a site like Wikipedia, how can they lose that relevance signal and still own the SERPs? I believe Wikipedia's inbound links make up the difference from that lost signal. So if your site has enough inbound links to make up the difference in losing a signal, then I suppose there shouldn't be a problem. The outbound link signal might be so small that it hardly makes a difference.

But you know, sometimes every little bit counts. We tend to think of the top ten as the result of a single algorithm that lines up the top ten answers. But that's not really the case. There are different algos in play when you make a query. Not all ten results are there as the result of a single algorithm. One of those top ten spots might give an edge to a site with more outbound signals. Or not.