It is time. It’s time to listen to Danny Sullivan chat with Matt Cutts and hope that he’ll reveal something very, very minor that we can all spend the next month over-analyzing and fighting over. Because, really, what is a summer without some SEO-fueled drama? I don’t want to find out.
Ready for the action? Let’s go.
Oh boy. Danny starts singing a song to the tune of It’s a Beautiful Day In The Neighborhood from Mister Rogers and people either laugh uncomfortable or blink wondering what was going on. After his solo, Danny presents Matt with a pair of custom Google Vans to wear. Danny, of course, has a pair for himself. [It's a business expense, he says.] Now there’s a giant stuffed panda in the room. And panda foot stools. This place has gone panda-mad. Don’t believe me?
We start with the questions:
Danny: Can we talk about the panda in the room? Why are the scrapers back in the index?
Matt: We’re working on that. There’s a change coming related to scrapers outranking people for their own content that will be released soon.
Will the update that comes after this be called the “honeybadger”?
That’s a great name. The honeybadger don’t care. [Hee!]
Has the algorithm changed enough where some people have actually recovered from Panda?
If you go back to Florida (the update, not the state), it launched and had a big impact, and then they pulled that back. They push stuff out and then they find additional signals to help differentiate on that spectrum. He hears the pain from people in search, but he also hears complaints about things polluting the search results. They haven’t done any official changes (though they did a few tweaks) to pull things back but they continue to look at ways to differentiate there. They’re still looking for ways to find more of the low quality sites
What did you tweak? Is it in your book?
It’s in my head.
Well, then lets hope nothing happens to you before you get back to Google. [Hee!] Are your updates being pushed out manually?
Yes. A lot of algorithmic stuff happens as we’re crawling the Web. If you were hiding text but then remove it and Google crawls you and sees it, then you’ll see that change. More complicated stuff has to be run offline.
With the release of the varios Panda iterations, site usability is being touted as a srong ranking factor. Agree?
Panda is trying to mimic our understanding of whether a site is a good experience or not. Usability can be a key part of that. They haven’t written code to detect the usabilty of a site, but if you make a site more usable, that’s good to do anyway. It’s the same thing as site speed. Site speed only affects 1 percent of queries so its probably not that your site has been affected, but its a good thing to optimize for anyway. That’s a better, smoother experience. Pay attention because it’s a good thing to do, not because Google said so.
How’s Larry? Does he come and hang with the search folk?
He visit our office a couple weeks ago and stayed for two hours. He thinks Larry has said some good things for Google like they’re trying to move fast.
[After that there was a long discussion about Larry searching for [warm mangoes] and not being able to find what he was looking for. Matt wondered what that query even meant, apparently if you put mangoes in a box they get warm because of the gases released. Matt things Larry needs to search better, but if he’s searching like that, everyone else is too.]
Matt has some announcements:
Earlier this morning Google announced this thing called rel=author where you can do authorship markup. Over time they can learn high-quality authors so they can rank people higher. For now it only works for the same site, but it’s really promising. That was announced today.
Danny: This is Google saying it’s an author-specific thing you can do. It’s your attribute.
Matt: Rel=author has been around for years…
Danny: Okay, say I do that. I link my byline to my profile and now you understand that this is written by me, Danny. [Matt nods]. In the future, I can write on my personal blog and get credit for it. It sounds like you’re establishing personal page Rank.
Matt: That’s the hope – AuthorRank. We’ll see what the traction is and then over time we’ll try to annotate it in the search results with a picture of Danny. Or maybe a panda…
Danny: Meanwhile, at schema.org they have a whole other way of using authors, which do we use?
Matt: Start with Google’s method. You can use them both.
Danny: You rolled out the blocking feature and then you said not only will users be able to block sites, but Google will use that data, too. Also, how are you using +1s?
Matt: You can block things using a Chrome extension or directly in the Web search results. They have said that the second version of Panda does use aggregated data of stuff being blocked, but only wen there’s a high level of confidence. The +1 is really handy, as well. Analytics on +1 is much further along. They are leaning away from giving stats on blocks because there’s not much actionable information and they don’t want people to freak out.
Danny: So don’t expect, “you’ve been blocked 200 times”.
Matt: No. Because what’s someone going to do with that?
Danny: It might be cool if you could find out its on a per-page basis
Matt: All this is undoable. You can click to undo it.
Danny: You wanna talk pagination?
Matt: If that’s a page that’s being returned, that can be really useful. Whether you have pagination on older or new pages, I wouldn’t worry about that. Things work fine the vast majority of the time.
Matt on AJAX: Use things when they make sense. Don’t just throw the magic sprinkly AJAX dust because its pretty, do it when there’s a reason.
Danny: Weren’t you dubbed the Alan Greenspan of Google?
Matt: I’ve been pretty consistent about everything I’ve said. You could pretend to be Matt Cutts and people probably couldn’t tell it wasn’t me. He thinks its a little unhealthy if people say “this is right because Matt said so”. It’s unhealthy because you want a lot of people communicating. He doesn’t think it’s good to put all that influence on one personality.
Some question about Panda I missed…
Matt: Panda is a site-based signal that will affect you based on where we think you sit on the spectrum. If we think your site is really, really good, you can still rank if we think like your site.
If you read our coverage on The New Periodic Table of SEO you heard Rand Fishkin state that the raw number of Facebook shares is the single highest correlated metric with high Google rankings. Matt says that’s NOT TRUE. Google doesn’t get Facebook Shares, not at all. If someone blocks us from crawling, we can’t crawl that content. Facebook gives data to Bing, Blekko, etc, but they don’t give it to us. They can see Fan pages, but that’s it. They don’t see the sharing data.
If you read our Competitive Intelligence coverage, you also hard John Shaw say that to regain its Google Love, JC Penney bought 200+ edu links. Matt says this is also NOT TRUE. JC Penney recovered from their penalty by getting rid of their paid links.
Companies that do rank checking report – thoughts? Historically, we’re not friendly to ranking reports because it’s the wrong thing to concentrate on.
How do you handle multivariate testing on sites? We like it. If you want to see a good way to do A/B testing, use Website Optimizer.
Danny ask Matt if there’s anything he wants to ask the audience. Matt asks if anyone knows of a good Cajun restaurant, causing Danny to ask if he tried to Google it. Heh.
And with that, we’re done. We will see you nice people tomorrow. I hope. ;)