You & A With Matt Cutts – SMX Advanced 2012by Michelle Lowery on 06/05/2012 • 6 Comments | Internet Marketing Conferences
Hey look, it’s Matt Cutts! Um, that stuff I said before about Google? Yeah, I was just kidding. Kind of. But not really. As a matter of fact, even if Google itself doesn’t have a sense of humor, Matt certainly does. [Okay, maybe not really Matt, but still. It’s funny.] This is the session a lot of you have been waiting for, and SMX was kind enough not to make you wait until the second day of the conference.
While Danny Sullivan moderates, members of the audience have the opportunity to put questions directly to Matt, who is not only the head of Google’s web spam team, but a “distinguished engineer.” No, really! You know you want to add “distinguished” to your job title now. But before you do that, listen up because I’m pretty sure a lot of people are going to ask the questions you’d ask if you were here, and Matt does not mince words.
Here we go!
Okay, Matt’s running a little late, so we’re starting with a video. And it’s the same video I just mentioned in the intro! Now, I have to tell you guys, I prepare these post intros days before the conference, so I totally linked to that video before they showed it here at SMX! Actually, I’m just psychic. Actually, it’s just a super fun video, so of course they’d show it before a session with Matt.
Danny tells us we’ll have 25% more Matt time than ever before. And then Matt’s throwing stuffed penguins into the audience, and he and Danny are showing us early prototypes for Google glasses. I just can’t do the visual justice. It’s mayhem in here! [Seriously, if you couldn’t make it this time, you really need to try to get out to SMX Advanced next year. Fun times!]
Now we get down to business.
Is Penguin a penalty?
At Google, we think of it more like trying to find the right calibration level. On one end, you’ll have fantastic, quality stuff, and on the other end you have spam, and then you have a bunch of stuff in the middle. Penguin was designed to tackle all that stuff in the middle. After Panda addressed the spam, there was a lot of other stuff we wanted to address, so that’s where Penguin came from.
It does demote web results, but it’s an algorithmic change, not a penalty. It’s yet another signal among over 200 signals we look at.
Are you saying that from this point forward, a penalty is someone at Google looking at a site and identifying it as bad?
That’s pretty much the nomenclature we use. If you look at our videos, we’re more likely to say “manual action” as opposed to penalty, to differentiate it from algorithmic changes. The reason we do that is we want to be as transparent as possible. We can’t possibly reveal the exact code for each change, but we’ve gotten pretty good at revealing the manual changes. If you haven’t registered in Webmaster Tools, you should, because whenever we make a change, you’ll hear about it in the Webmaster Tools console.
That’s why if you were hit by Penguin, a reconsideration request isn’t going to help you?
That’s right. People who think it should rank higher after Penguin can let us know and we can look at it, and in a couple of instances, it actually helped us make a couple of tweaks to the algorithm.
Speaking of cases being approved, there was the WPMU site out of Australia who said they haven’t done anything wrong, referring to footer spam.
They didn’t rank as high after Penguin, they made their case, and I thought it was a good case. We were able to remove about 500,000 instances of links, and that helped them.
Why aren’t blog networks effective anymore?
We’ve been taking stronger action. It’s interesting to see the reaction to that in regards to negative SEO. People have asked about negatvie SEO for a while. There’s the case of sex.com, where someone actually faxed in an impersonation of the rightful owner, who had to go to court to get it back. So we’re trying to make it more difficult for a competitor to make things difficult for another competitor. We build in a lot of protections. Some people have indicated a hope that we could disavow links. Even though we put in a lot of protections, there have been enough people to ask about that to make us consider whether it might make sense.
Getting back to WPMU…
There are people who continue to sell links, although they don’t do any good, and that’s part of how SEO has a bad reputation. If we de-index a site, it’s a clear indication that selling links isn’t doing anyone any good. Here’s the litmus test, why we think it’s important to differentiate between penalties and manual corrections: It’s not practical to have 500+ changes every year, and for every one of those, you get a notification that you’ve been affected by it. Early this year, we’ve started to send more messages, something like 700,000. People thought it was unnatural link warnings, but they weren’t. The vast majority of them were very clear-cut black hat, and spam. We told people about things like hidden text, and earlier this year we said let’s also tell the people who have a black hat violation so we’re completely transparent with the manual actions we take.
About paid links…anything going on there? Anything recently?
We’re always working on improving our tools and how we can spot people buying links and blog networks. People don’t realize, when you buy links, you might think you’re very careful, that you have no footprints, but you may be getting into business with someone who’s not as careful. People need to realize as we build new tools, it becomes a higher-risk endeavor.
Last year JC Penney got hit, and this year a major agency got hit.
Yeah, if everyone else is jumping off a bridge…no, don’t do it! If you ask any SEO, is SEO more dynamic and maybe a little harder than five years ago, they’ll say it’s more challenging. You can expect that to continue. I feel pretty good about the way all the search engines are getting more serious about it. We see good trends. Some things that work short-term aren’t going to work long-term.
Affiliate links—I think you said we don’t need to worry about that. Do people need to make sure they’re nofollowing?
We handle the vast majority of network links appropriately, but I would say if you’re worried about it, go ahead and nofollow them if you’re making money from them.
[Unfortunately, I missed Danny’s question here. Maybe someone can share it in the comments?]
Douglas Adams wrote, space is big. You have no idea how vastly, hugely, mindbogglingly big space is. The Web is like that. If you look at the Library of Congress, they have roughly 34 millions books, something like 235 terabytes of data. That’s not that much data. YouTube gets videos uploaded every minute. There’s more data on the Web than in any other source in the world. If you look at the percentage of nofollowed links on the Web, it’s a single-digit percentage. I wouldn’t write the epitaph for links just yet.
Matt likes 30-day challenges, like I’m going to go off the Internet, I’m only going to use Bing for 30 days…
I try to only do good habits!
I’d like to see you and members of your team find a small, non-profit and spend 30 days building links to that site. I’d like you to get in the trenches and see what it’s like. And they want it so badly, they’re going to agree. I would like to ask the SEOs to do a 30-day challenge where you put them into your shoes, too. So think about how you want to do that.
That’s totally fair. It’s one of the reasons I started my blog. A lot of people on my team do run their own Web sites. I’ll think about that.
What’s up with your war on SEOs, man?
There’s no war on SEOs!
Or is it just a war on spam?
Yes. We’ve been consistent over the last seven years. Make your content compelling. Give people a reason to come back to your site. If you look at black hat sites, you’ll see stuff like, how do you fake sincerity? How do you fake being awesome? Don’t fake it! Just be genuine! We do what we do so people can compete on a level playing field. Our philosophy has been relatively consistent.
Does Google look at bounce rate?
People keep asking, do you use GA? Google doesn’t use GA in its rankings. That’s the sort of thing…I like that, personally. Think about bounce rate. It’s going to be noisy. Every time you have a redirect, you lose about 1% of your traffic. That’s a good reason to minimize redirects. Bounce rate also doesn’t measure if you just got the answer you need. If you go to Google and type in “sunset,” you get the answer immediately, so you don’t need to stay. I understand the interest in people saying Google’s going to use it. We try not to take stuff off the table, I just see a ton of pitfalls in trying to use bounce rate.
What percentage of the organic queries on Google are now secure?
Not provided. If you look back with the perspective now, you can see Google is trying to make it so you can search more personal stuff. Over time we’d like to be able to index more data. That’s why we were adding all the SSL and all that. Whenever we launched SSL, we were launching it only in the US, so as I said, it was only 1 percent of sites that saw it. If you have a large percentage of signed-in users, you’re more likely to see a higher percentage of not provided. If you look at the trend, you’ll even see some browsers like Firefox saying they want more secure data. You want to know that no one can see your queries in Google. I think SSL is a good way forward to protect users’ queries. I realize that puts marketers at a disadvantage. We’re not happy about that, but we have to put the users first.
Now you can go back 90 days, and get 2,000 queries. The only thing you’re missing is the exact query when someone landed on a page. Moving toward SSL is the right direction.
We now have a Python script where you can run that every day, and keep that data for as long as you want. You can download it, at least.
What is going on with not being able to block sites anymore?
You can block in the Chrome extension, and in the search results. People who complained about some very high-profile sites, the Panda update we did incorporate that data.
Blogs powered by WordPress have that footer link. Why weren’t they affected by the most recent update?
We’re trying to address Web spam violations. When you look at the volume of WordPress links, it’s a pretty high volume of quality links.
Why isn’t AdWords secure?
I wouldn’t mind going back to revisit that decision, but the reason is if it were, I’d expect every advertiser to go in and make everything exact match. We’d have to deal with exponentially large amounts of matching data.
If you received a warning due to unnatural links, should you submit a reconsideration request?
Yes. Then we’ll let you know whether you did enough, or you still have work to do.
Does Google penalize sites for using rich snippets?
People used to complain about not being able to get rich snippets. Now, it’s that other people have rich snippets. I’m glad we erred on the side of inclusion. You’ll see people using rich snippets wehre they don’t really apply. So that could lead to a demotion. That’s an example of where we need to make sure it doesn’t get abused.
We were penalized because of incoming links, submitted a reconsideration request, but we remain penalized. What do we have to do?
When you’re getting a warning like that, we’re typically looking at a sub-sample, a random sample. When you submit a reconsideration request, we’re going to look at those, and if none of those have been removed, nothing’s going to change.
If you can’t get rid of the bad links pointing to a page, should we get rid of the page?
You could, if it’s not a critical page. But think about how you’re going to document the effort.
[Missed another question here. Sorry!]
We have a good list of leads ourselves. We’ve shut down tens of thousands, maybe hudnreds, involved in networks, link buying, etc. We have to differentiate between what the person is saying, and what information we have available to us. The balance is, we take the impact of the spam report, and multiply it by a factor of four. We basically say, look, they might be reporting a small incident of spam, so if you multiply it by four, we get a better idea of the impact it has.
Is Google making Google+ so SEOs simply become your pawns to take over Facebook?
Not in my opinion. It’s still early days…(the crowd erupts in laughter)…no, wait! It’s still early days! When we look at that data, it’s not the best quality signal yet. We’re going to look at search quality rankings, and other things, and the +1 data has to make its way in that world too. I think Google+ has a lot of potential. There are reasons to be excited about it, but I wouldn’t say you have to do a ton of +1s right now.
There’s a blog post going up right now, “Matt Cutts says don’t use Google+!”
Hey, why’d you guys call it Penguin, by the way?
There actually was an engineer named Panda. We let the lead engineer pick. I suggested a cute, furry animal might be good, but he chose Penguin.
If you were hit by Panda and Penguin, should you just give up?
Sometimes. Sometimes you should. It might be better to start over. But both are algorithmic, so if you change the signals, you can recover. If you produce content people like, you can come back. We’ve seen sites come back, like WPMU. But if you’ve been spamming, don’t expect that to work anymore.
What’s with paid inclusion?
You call it that, but that’s not a Web ranking. Google’s take on it is that you’re taking money and not disclosing it. I think Google’s Web rankings remain just as pure as they were ten years ago. By doing these different things, we can get access to data we didn’t have before, and it’s helpful to users. But I think if money is going to change hands, you should absolutely disclose it. We care very much about the integrity of our rankings.
So, I stalk you. I searched your name. People say some really mean things about you sometimes. How do you handle that? Do you want to slap them upside the head sometimes?
The beauty of participating in forums for the past decade is I’ve developed a thick skin. People aren’t striking out because they’re vicious. It’s because they’re defensive, or they’re hurt in some way. But you also have people who sincerely think Google is doing something wrong, so we want to listen to those people and see if there’s anything we need to change.
What’s the coolest thing you like in search right now?
There’s some fun stuff I’m looking forward to that hasn’t launched yet, but I’m excited about it.
That doesn’t help us.
If you search for Danny Sullivan, you’ll see a little knowledge graph about you, and it’s pretty cool.
What’s the coolest thing you like about Google right now?
I’m really excited about the fact that we’re pushing for more transparency. I think we’re not done, but we’re getting there. If you had told me before that after we caught a spammer, we were going to tell them that we caught them, I would have thought you were crazy. But we’re doing it, and we’re trying to make the Web a better place.
What other thing do you think is really cool right now?
FitBit. It’s like a little pedometer that tells you how many steps you take every day, and you can connect with friends. It’s actually very competitive, though. People will ask me, do you want to be FitBit friends? And I say, I don’t know, how many steps a day do you take? Oh, you’re too fit for me, forget it. But yeah, it’s a fun thing.
And that’s it! Day one of SMX is over, but there’s plenty more good stuff coming up tomorrow. So in the words of Matt Cutts, peace out, spread the love, and we’ll see you tomorrow!
Get all the SMX Advanced 2012 coverage here!
About the Author
Michelle Lowery is an ardent word nerd, but is also known to say "y'all" from time to time.