Bummer! This is just a preview. You need to be signed in with a Pro account to view the entire video.
Ecommerce SEO: Cutting Edge Tactics That Scale34:06 with Adam Audette
Fight Panda and other modern SEO realities by using the best on-page techniques and content strategies for your ecommerce site. Adam teaches you how to sustainably improve your click-through-rates as SERPs become noisier and properly prepare for G+ and Graph Search. Then he'll round things out be giving practical advice on how to build your ecommerce team and work flows.
[MozCon Adam Audette Ecommerce SEO: Cutting Edge Tactics That Scale] [? music ?] 0:00 Hellooo! 0:08 Sorry! I'm really excited because I love SEO, and this is SEO's biggest stage, literally and figuratively. 0:09 Hi everyone! 0:17 Adam Audette, and I'm going to talk about SEO and Ecom SEO, 0:18 and I have a ton of slides and a ton of stuff. 0:20 Whew! 0:23 So let's get started. 0:24 First a little bit of history, and I'm actually not that interested in the history of SEO and traditions in SEO because 0:27 I think what we need more than that—more than the old stuff is we need new ideas. 0:34 We need people to come with new idea—fresh ideas, 0:39 which I think this is all about. 0:41 You guys are the right people for that, and—oh my god—that's a lot of Mac laptops. 0:42 Well done! 0:46 Very few PCs, except for Rand, right? 0:48 So let's go back and let's hearken to an older time when SEO was about making money online and getting rich quick 0:54 and people had their own websites or maybe they were affiliates, 1:01 and SEO was a way you could game the search engines to rank highly and get all kinds of free traffic and cash the checks. 1:05 Literally, money would come out of the monitor like this guy right here. 1:12 I mean, it was crazy you guys! 1:16 You should have been around in 1999. 1:17 It was sick! 1:19 Optimize for the search engines and cash the checks. 1:21 So SEO was kind of born in that mindset, right? 1:24 Find holes in the algorithms, manipulate them, rank highly. 1:27 It's sort of personified by this guy. 1:31 This is the first SEO right here. 1:34 He was an isolated guy. 1:36 He worked in seclusion. 1:39 He loved online role-playing games. 1:40 He loved to play online poker, 1:42 and he loved to find holes in the algorithm so he could manipulate them to his benefit and make money. 1:44 The problem is a lot of his ideas and his mindset has held over and kind of just come with us into the new age of SEO that we're in now. 1:50 We're kind of on this bridge where we're going for the next level of SEO, 2:00 but we're still kind of hanging on or this guy is kind of hanging on with the old mindset and the old ways. 2:03 Like rank number 1 because you should rank number 1 and links, links links—you need links all the time and keywords all day. 2:09 Go for the keywords and get the keywords and rank 2:15 and obsess on the algos because it's really about following the algorithm in minute detail so you can figure out how to manipulate it. 2:17 Whew! The thing is that is all crap. 2:24 And if you do a search for crap in Google with the definition—little modifier— 2:30 this is actually what comes up in the Knowledge Graph. 2:34 Just look at it for one more moment. 2:39 It is crap. 2:41 But here's the thing, and this is what you're going to hear is like okay—okay—I get it. 2:45 Yeah—not just about search engines—about users. 2:48 But it worked, and you can make a lot of money doing that. 2:50 Well, that is sad and kind of true is that it used to work. 2:52 But then things changed. 2:56 A couple of things happened. 2:58 It doesn't work that way anymore. 2:59 Brands came online, and they were like, "Whoa! What about this SEO thing?" 3:02 I want to get some of that because it's insane as a profit center. 3:05 You just do a bunch of stuff, and you get the free traffic coming in. 3:08 I want some. 3:11 It's incredible. 3:11 It's probably the most effective way of understanding intent and leveraging it for a commercial end. 3:12 So brand said, "Give me some SEO. I want it." 3:20 That was one thing that happened. 3:22 The other thing that happened is the Internet started hating SEO because—well, hating crap SEO— 3:23 because crap SEO was spoiling the Internet for everybody. 3:29 It was ruining it. 3:34 It really was. 3:34 As a SEO—and a lot of others in the room here— 3:35 I mean, we can all sympathize with that. 3:38 It's a really bad experience when crappy results—and still to this day we can find those— 3:40 end up ranking highly for stuff when you're actually trying to accomplish a task—you know—using a search engine. 3:45 By the way, "Dorothy Mantooth is a Saint." 3:51 She's a saint! 3:53 So the other thing that happened is Google said, 3:56 "Well, we better do something about this because our users are kind of unhappy when they see a lot of spam there, 3:58 and that's going to dive down our stock, so let's bring forth the Panda and torch the thin content." 4:03 And whoops! 4:09 I've got to go back. 4:10 And let's bring for the Penguin and torch paid and manipulative links, 4:11 and let's do a whole bunch of other algorithms and a whole bunch of other products and a whole bunch of other things to change that. 4:15 And they have changed that. 4:21 And the level—the playing field now—it's much, much higher. 4:22 This guy is like, "Don't! Okay!" 4:26 Well, I've got to fire up another site because that one just got torched, and oh, what's the algorithm doing now? 4:29 I've got a whole bunch of different sites, but I'm getting penalized all the time. 4:34 It's not sustainable. 4:38 Brands coming online. 4:40 So brands want sustainability, and they want efficient marketing channel—efficient marketing channels. 4:43 Yes, but for SEO to be valuable, it has to be sustainable. 4:47 It has to be something that you can invest in for the long-term and you can build over time. 4:51 Otherwise, it's kind of this whac-a-mole where you're constantly kind of moving and jiving and shaking. 4:54 It's just doesn't work. 4:59 So if we think about the old mindsets, back then it was about chasing the algorithms. 5:01 I want to be clear. 5:08 As an SEO geek, I love knowing what the algorithms are doing. 5:09 So I really want to understand that, but I don't that to be the focus. 5:13 Instead of the chasing the algos, I want to chase the people that are using the search engines that have the algos 5:18 to get rid of people like from manipulating and spamming their search engine. 5:22 It's really though about chasing the people and thinking about users, 5:27 and that's who we're going after. 5:30 We're not going after search engines. 5:32 They're a means to an end. 5:34 The old mindset was go for the keywords that have the most traffic. 5:35 It's all about looking for keywords with volume, and that's what really matters. 5:41 Now it's about looking for keywords that have the most relevance, even if they have lower volume. 5:45 You might want to go for a different set of keywords that are 5:50 more relevant and more tightly matched to the target demographic you're going after. 5:53 The old way was obsessing on links. 5:58 The new way is thinking about creating experiences online worth sharing. 6:00 Create great content, you create great digital experiences—people share it 6:05 and they talk about it and they link to it and great stuff happens as an outcome. 6:11 But links aren't the driving thought at the beginning. 6:15 I mean maybe they are if you're an SEO, 6:18 but it's not I want to build 500 links. 6:20 It's, hey, what kind of great experience can we create? 6:22 What kind of great content can we create in order to drive those links and that awareness and that visibility? 6:25 Here's an example of that. 6:31 Tony Hsieh, CEO of Zappos, had a billion dollar idea, 6:32 and it was basically just create amazing customer service. 6:35 He didn't even need a link development strategy 6:38 because he had that great idea. 6:41 That totally dovetails exactly with what Joanna was saying is that it's about connecting with people and delivering that wow factor— 6:42 something that Moz does so well. 6:49 Back then it was more of everything, right? 6:52 I want more links, and I want more pages in the index—more, more, more, more, more. 6:54 We used to take a very authoritative domain and just try to blow out as many pages as we could because it worked so well. 6:59 Maybe they were site search pages or maybe they were dynamically created pages where we could do just enough to make them rank, 7:07 and it was just all about blowing it all out in the index to try to get maximum value, maximum punch. 7:15 Today, less is more. 7:20 The only thing we want in the index are the money pages that are producing revenue— 7:22 that are delivering a satisfying experience for users. 7:25 Everything else that isn't working, we want out—ruthlessly out, using robots.txt in a way that I've never used before. 7:29 No follow, etc., etc. 7:36 Less is more in today's SEO. 7:38 The old way was focusing on rankings. 7:40 We still need to be aware of ranking, especially from a competitive kind of analysis standpoint. 7:43 But instead of just looking at rankings and using that and reporting, 7:48 we really need to look at bottom-line metrics. 7:51 That's where the focus should be—to look at the channel, 7:53 looking at non-brand revenue from organic, looking at conversion rates, looking at new customer acquisition, 7:57 or whatever KPIs that you come up with. 8:02 Rankings are an incidental and sometimes noisy and lousy metric that just don't work as well as some of these more bottom-line metrics. 8:05 The other thing that's happening, and this is something that I really pound into people both at my company and just other SEOs 8:15 that I talk to is that we as an industry need to do a lot better job of standing behind our work. 8:20 Standing behind our work means putting projections and real dollar amounts on the recommendations that we put out. 8:26 So we say, okay, we want to make this change—this sweeping change to project pages. 8:31 Then we need to figure out what kind of lift we can expect to see and what the conversion rate is and what the average order value is 8:36 and then get a number like—hey, this change that we recommend that's going to take a bunch of resources is going to 8:42 make us 164-ca-ching-ca-ching-cash-cash money in new monthly revenue. 8:46 That's the kind of thing that CMOs love and that businesses love—is pleased because it does a couple of things. 8:51 One is it rationalizes the resources, which is really, really important in every company. 9:01 Nobody has resources for everything, so we have to pick and choose carefully. 9:06 But it also forces us as SEOs to stand behind our recommendations. 9:09 We've got to put our butt on the line so-to-speak, 9:15 and there's something else that I like to say is that recommendations that we have— 9:18 if they're not implemented, they're just completely useless. 9:22 They don't mean anything. 9:25 It doesn't matter how good they look or how smart they were or whatever. 9:26 They're just going to sit on a desk. 9:29 That doesn't do anything for the company—no value. 9:30 Putting a business case on it and putting your butt on the line with that allows you to stand behind what you're recommending, 9:34 and you make sure it's something that's valuable. 9:40 Seriously, dude! 9:43 You better not come and tell me to change all my damn URLs unless you have a dollar amount that you're going to attach to that. 9:44 Hopefully, nobody is recommending to change URLs on your site anyway unless there's a really good reason. 9:50 All right. 9:56 Let's talk about some content strategy—how it relates to SEO. 9:58 I talked about how keywords—they only get you about half-way there. 10:04 I mean keywords are one thing. 10:08 Who is using the keywords is who we really care about. 10:09 It's the people. 10:12 It's the people that are searching. 10:13 Those are the people we want to target. 10:14 That's what we need to understand, 10:15 which means we've got to get kind of old school. 10:19 We've got to do some categorization exercises and figure out different tasks that people are going to accomplish, 10:21 and then build out personas so that we can understand who that target demographic is, who that persona is. 10:26 Then we can blow out something like this, 10:33 and we can say, "Okay, for this particular persona, here's the sample search queries that we should target." 10:35 "Here's what she wants to accomplish." 10:41 "Here's what we can do to sell her, and here's—how about some value propositions and some up-sales, too." 10:43 "Here's how we're going to do it, and here's the conversion that we're going after." 10:49 It's taking keyword research, and it's putting it within the context of a very specific demographic that you're targeting. 10:52 This can be really effective, and it also feeds into the way you create content and the way you rank and display in search. 10:59 So if you think about it—I keep going twice and I'm ruining the effect! 11:05 I've got to calm down. 11:10 I'm so excited! 11:11 We think about content strategy—SEO content strategy—as starting at the search box. 11:13 That person that we already know we want to target comes to Google or Bing or whatever, 11:18 and they search or index. 11:23 Those queries that they use are very specific to what they're trying to accomplish. 11:26 How we rank and display is very important because that's where we want to— 11:31 it's not just about putting keywords in title tags to rank well. 11:35 It's about using terms in title tags to be compelling to trigger a high CTR. 11:39 We want to trigger that click-through. 11:46 We want to stand out on the page because it's very crowded, 11:48 and I'll talk more about that in a little bit. 11:50 That's where using terms in the title tag are really important. 11:52 The snippet text—Google will often change the snippet text, 11:55 but we should put time into how we think about those schema for rich snippets, authorship, etc. 11:57 Then once we have them, if we're lucky enough to get them from the SERP, 12:03 then they land on our site and we have a piece of content or we have an experience that we created with them in mind very intentionally 12:06 to drive them towards one thing—a conversion—a sale on ecom. 12:13 So we're already departing from this guy—this old SEO, who created so much of what we have. 12:18 But one thing that he did that was really great is he was really technical. 12:25 Today, technical SEO is probably more valuable and more important than it was before—not less. 12:29 That will only happen more and more because what we're doing with SEO is we're targeting users that we want— 12:35 a targeted demographic through search engines, which is a technology. 12:40 So we need to be technologists, too. 12:43 We need to be very savvy when it comes to technical stuff. 12:45 Here's—I'll talk about that. 12:49 But really quickly, let's talk about impact and reliability. 12:50 This is something that I like to talk about, as well, is that when we think about— 12:52 when we think about how we prioritize our recommendations in our work, 12:56 when we put all the things that we could do on a piece of paper and we can only do maybe 3 of those things, 13:00 it's really important that we understand what is going to have the most impact and what is going to be the most reliable. 13:07 If you look at this really terrible graphic because I suck at PowerPoint, 13:12 what's is showing is that it's a pyramid. 13:18 You've got off-page factors on the top, and you've got on-page at the bottom. 13:19 You've got content in between, tying it all together. 13:23 The off-page stuff tends to be very high impact, or it can be, but lower overall dependability. 13:26 So we can build links pretty dependably, but we can't dependably make things go viral. 13:32 We can't dependably build tons and tons of links. 13:37 We can't blow it out of the park every single time. 13:40 We can, but it's not always dependable. 13:42 On the bottom is low overall impact but high dependability, 13:43 and that's technical SEO. 13:48 Technical—especially on big sites and big Ecommerce, 13:50 when we look at those, there's tons of low-hanging fruit always, 13:53 and it's dependable to go in there and fix that stuff. 13:56 It's not going to be as high impact unless the site is—say dis-allowing itself entirely in robots.txt. 13:59 That stuff has happened but, generally speaking, it's not going to be as high impact. 14:04 But it's very high dependability. 14:08 That's an SEO's best friend—to know that there's money in those hills—go get it. 14:10 Here's another reason why technical is so important. 14:16 For our client set in my company—RKG—our agency, about 88 percent of our clients are in Ecom. 14:18 So a very high focus in Ecommerce. 14:23 Twenty-five percent of all our client's traffic is mobile now and organic. 14:27 It's huge! 14:32 Mobile is so important, and it's going up and only going to continue going up. 14:32 Adobe just released a study recently that showed tablets now out-pacing smartphones in terms of usage. 14:36 We know a couple of things here. 14:44 1. Tablets are used more often than smartphones to visit ecommerce sites. 14:45 That's just makes sense. 14:50 When I'm sitting in bed, I don't really want to be on my phone if I can be on my Windows tablet or my iPad. 14:52 You have more real estate. 14:56 The other thing is this is revenue per click by device for our clients, 14:58 and you can see that Windows tablet and iPad are very, very high revenue. 15:03 Windows tablet, especially, even over desktop and laptop. 15:07 More revenue per click for those devices. 15:11 What happens is when a new tablet comes out, there is a very high spike for the early adopters, 15:13 who tend to have a lot of disposable incomes using those devices and shopping. 15:18 This is really important to know because this trend is going to continue 15:22 because the evolution into mobile and into tablets is just beginning. 15:28 I mean there is so much to do here. 15:31 This is why we need to be really cognizant of technical stuff. 15:32 The other thing that happens is tablet users tend to look at more pages because they're sitting back, they're reclined, 15:35 they're not really task-focused necessarily—sometimes they are. 15:41 But they're more—you know—you're more kicking back, 15:44 you have your feet up, and you're drinking a Negroni and you're looking at your tablet. 15:45 Any Negroni fans in the house? 15:49 So here's why it's important. 15:53 We've got to be on every single device. 15:54 It doesn't matter what it is. 15:57 We've got to be device agnostic. 15:58 I've actually argued against using responsive design in many ways. 15:59 I'm starting to come around. 16:03 I think we need to be thinking about responsive design more, and this is why. 16:04 When you do it right, you end up in a best case scenario. 16:09 From the SEO standpoint, you have one URL. 16:12 So it's completely elegant and consolidated. 16:15 If you do it right, you get faster across every device—not just on mobile. 16:18 So one of the big knocks is, well, if I'm using responsive design, 16:22 I'm going to have my all my desktop HTML bloated on my mobile devices and that's going to be a really slow and bad experience. 16:25 But what happens—it doesn't have to happen that way. 16:33 There was a case study that was recently released that showed 16:35 that these guys put a ton of time and they did responsive design right, 16:39 and they ended up with faster load times across every platform and device. 16:41 So that's the promise of responsive. 16:47 This is why as SEOs we need to know about this stuff. 16:48 I count myself in there. 16:50 I need to learn about responsive design. 16:52 I don't need to know how to build it. 16:53 But I need to know how to talk to it in an educated way so we can help our clients do this. 16:55 A couple of resources there for you guys. 17:01 There is a Vary—this is a very kind of minute thing that I want to talk about with the Vary header. 17:03 So Google said, hey, for mobile there's basically 3 ways you can handle it. 17:08 One is if you're dynamically serving the contenta ?, use this Vary user-agent header. 17:12 Just tell us so we know that the content will vary based on the device. 17:17 The problem with that is it breaks Akamai, and it may break other CDNs, as well. 17:22 So Akamai tries to cache everything. 17:27 When it sees the Vary user-agent header, it doesn't cache. 17:29 What happens is we get a huge spike in requests as Akamai starts to pull every asset and every page without caching. 17:32 So it causes a bit of a problem, but there is a fix for it. 17:40 One is you just configure Akamai to ignore the Vary user-agent—ignore it but still use it. 17:43 Secondly, there is an "is mobile" flag that Akamai uses. 17:49 It's all documented here. 17:53 There's an explanation and a fix. 17:55 It's a good thing to know. 17:56 This is why technical SEO is more important today than it's ever been, and it's only going to continue being that way. 17:59 Here's another reason—500 errors and stuff like this. 18:03 We've got to be really on top of this stuff. 18:06 Every site has finite crawl resources, and when you're dealing with large ecommerce sites you've got a ton of pages that are just dupes, 18:07 overhead, junk, crap, stuff—it's not your money pages. 18:15 You don't care about it. 18:17 It's in the index. 18:18 It's all over. 18:19 Google's crawling it. 18:20 What happens is your good pages that you do care about—your money pages—end up getting skipped and don't get crawled 18:21 because you have finite resources. 18:25 This is a site—large ecom just launched a web performance upgrade May 22. 18:27 You can see the decrease in time loading a page and then the occurrent increases in pages crawled per day 18:33 and kilobytes downloaded per day. 18:40 So good stuff! 18:42 I mean this really is important and works really well. 18:43 You should fix server errors when they spike and when they're large and when they're—you know— 18:45 you shouldn't fix every server error ever because that's not smart and not a good use of time, 18:50 but you should fix them when they spike. 18:54 Speed does rule. 18:57 Jonathan Coleman has put together some fantastic presentations on this. 18:58 There's 2 right there. 19:03 You should read those. 19:04 He does a great job of detailing out some of the more important factors to look at. 19:04 Use parameter handling in Google and Bing. 19:08 Google respects that pretty well. 19:11 Bing not as well, but they're worth doing anyway. 19:13 Tell Google—give them hints. 19:16 Give Bing hints on what to crawl, what not to crawl, etc. 19:17 Monitor indexation and XML. 19:21 I think it's really sad and a professional travesty when I see a 19:23 site colon search and a number like for an indexation rate in a report somewhere because it's not accurate. 19:32 It's pulling from different data centers, and it's going to be different every time you check on it. 19:33 I feel like I'm taking crazy pills when I see that. 19:36 The only accurate way to monitor indexation is by XML site maps, and the best way to do it is to blow them out into different page types. 19:40 So on an ecom site we would have category and sub-cats and products—maybe brands, etc., etc. 19:47 I like using the 304 not modified. 19:55 Google has said in the past they they may choose to skip it. 19:56 We've seen good things from using it. 19:59 So I recommend using it on large sites. 20:02 Then use log files. 20:05 It's important to look at those to see what's happening. 20:06 So you should be monitoring kind of the distribution of response codes. 20:08 Green is good. 20:11 Yellow maybe okay—maybe not—look into it. 20:12 Red is bad. 20:15 When you see this, you know you have an issue. 20:16 What's going on! 20:21 Usually, this is the kind of thing that if you're not looking at log files actively and you don't know this is happening, 20:23 it's the kind of thing that there's an outcome in a few weeks' time or even a few days' time where you've got problems with ranking 20:28 or you've got problems with traffic or problems in the index. 20:35 So it's great to be really proactive and look at this stuff as it's happening. 20:37 We have also seen this. 20:43 On the right hand—on your right hand side, there's a spike in 404s. 20:45 On the left hand side, there is also a spike in 301s for this site, 20:50 and there's a drop—the light blue graph trend below, there's a drop in 200s. 20:53 So there is a correlation between what Google is crawling. 20:58 So if Google is suddenly crawling tons and tons and tons of 404s or 500s or something like that, 21:01 there's going to be a drop in good 200 pages that they're going to be crawling. 21:05 That's why it's important to fix that stuff and be really proactive. 21:09 You can also use it to find different URL dupes and stuff. 21:11 Here's a bunch of different examples—things like error pages that are returning a 200, 21:14 different—duplication with 3 separate URLs for the 21:20 same piece of content because of capitalization are because of trailing slash or whatever, 21:23 different parameters that you don't want crawled. 21:27 When you start to comb through logs, you can see this stuff. 21:29 Another great way to do this is to crawl using something that Screaming Frog. 21:32 Screaming Frog is a great crawler. 21:35 You can have it on your laptop, and you can literally crawl anytime. 21:36 It wasn't enough for us so we built our own crawler, and it's no where near as gynormous as LinkScape. 21:40 It's designed specifically for SEO. 21:46 What we did is we took a—basically we created this data warehousing model. 21:48 So we crawl, we store it in a warehouse, and we query it. 21:53 Because one of the problems is we end up with this. 21:56 You end up with tons and tons of files with lots of data in them. 21:58 It's hard to go through. 22:01 So what we do is we crawl, 22:02 we warehouse everything, 22:03 and I recommend this approach. 22:04 Then just query it with some canned stuff that we know we want to look for—flags or breaches or things that we know we want to see— 22:06 like find every link canonical tag that returns a status code other than 200. 22:12 Some top technical gotchas—I've got about 5 and a half minutes, so I'm going to go really quick. 22:19 One of them is inconsistent signals. 22:24 We still see this a lot where sites are using link canonical tags that aren't part of their internal link profile, 22:26 or they're using link canonical tags that are self-referencing and then they self-reference a non canonical with a parameter in it. 22:32 Those are both really common. 22:38 One of the things that I talk about a lot are canonical crawl paths and having a really consistent experience— 22:40 like from the home page straight down to the product page level where all the navigation and internal links and XML hints 22:46 and all the link canonical tags—all that stuff lines up. 22:53 When that happens, it can be very, very powerful. 22:56 Pagination is always an issue. 22:59 Rel prev next is the standard for that. 23:01 We haven't had a lot of adoption on ecom sites with view alls. 23:03 But rel prev next is working great. 23:07 One thing to keep in mind is that the rel prev annotations need to reference non-canonical versions. 23:10 You should include things like session IDs. 23:15 Then you use a self-referencing rel canonical to the actual canonical URL to clean up the dupes. 23:18 So they're 2 different things. 23:23 Rel prev next does the series and rel canonical does the duplication. 23:24 They work together, but they're separate. 23:29 For example, here's a rel prev next referencing stuff with an ssID in it. 23:31 The actual canonical is /pg3. 23:36 That has to get referenced, too. 23:38 Product variations—we see a lot of problems with these. 23:41 The old way was stuff like this where you would have different colors of products and go ahead and 23:44 just create a new URL for each one and put those in the index. 23:51 That doesn't work as well anymore. 23:53 I like REI's approach where the URL does not change based on the color selection. 23:55 It stays the same, and that's just given in the interface. 23:59 Yellow, blue, green—you have one URL. 24:01 It's very nice. 24:05 Zappos does it a little bit differently. 24:05 They create color-specific URLs. 24:08 But, then, they just reference back to a color agnostic version with rel canonical, and that works really well, too. 24:10 That way you can take them, you can share them, you can send them to your grandma and she can buy you some shoes. 24:16 Faceted nav is one of the big bug-a-boos in SEO on ecom sites because it's a great user experience, 24:22 but it's a huge problem for bots. 24:28 One thing that we like to do is we build—we build the SEO out of the way people search. 24:30 So if people are searching for different facets, 24:36 then we'll create the URLs and let them get indexed for those facets. 24:38 It doesn't have to be in folders like this is, making it look like sort of a static-looking URL. 24:42 Parameters are perfectly fine if they're clean. 24:46 But the basic idea is pick the ones that are important for your customers to find in search, 24:48 and then create URLs out of that. 24:55 Create the pages. 24:58 Make sure they're relevant—titles are unique, content—maybe a little snippet of content on there—dynamically generated, etc. 24:58 Then any overhead facets that aren't important for search— 25:04 things like price range and sizes and stuff like that. 25:08 You can add a PIN to the end with a parameter and rel canonical back to everything before the question mark. 25:11 It works quite well. 25:18 Product inventory. 25:19 We tried this thing at Zappos where we said, 25:21 "Okay, for all the products that are no longer going to be carried, we'll call them dead products." 25:23 Let's put them in this section of the site and then we'll say, 25:26 "Hey, this is no longer here but here's a bunch of other stuff you might like that's relevant." 25:29 This didn't really work out that well for us, actually. 25:34 I've heard from others that it's worked for them, 25:37 but I haven't had a good experience doing that. 25:38 Some sites like this one, they use review content. 25:40 So they'll just—okay, we have a review on here, but we're not going to carry this product anymore." 25:44 Let's just keep that up. 25:48 That's a pretty bad user experience too, and I don't think it's that valuable, especially when you're not recommending other stuff 25:49 that's related like they are not. 25:54 So I don't like that. 25:56 Best of all way to handle expiring products is to generate the product not here message, 25:58 and then recommend tightly related products. 26:02 If you can't do that, you can 301 to a related product or you can 301 up a level. 26:04 But it's a little bit of a harsh user experience coming from search to get 301, and until that gets updated— 26:09 sometimes Google is not going to follow that the way that you would expect either if those aren't really equivalent. 26:16 It's okay to 404 the page. 26:21 It's actually okay. 26:23 In ecommerce, typically what happens is links accumulate home page, category level, links accumulate at the sub-cat level, 26:24 less so at the product level. 26:31 So, yes, equity flows down to there, but it's not so much that on a big scale you need to worry about it. 26:33 I think that we should be more open to using 404s when we are constantly trying to scrape link equity and 301 and rel canonical— 26:39 404s sometimes are okay. 26:48 All right, let's shift gears a little bit to some key challenges. 26:50 Loss of data. 26:54 Annie Cushing is going to talk a lot more about this, so I'm going to be brief. 26:55 But the iOS 6 issue where if I'm using my phone or tablet and I search in Google, 26:57 we don't get to see any of that on organic. 27:02 It comes through as direct. 27:04 So about 80+ percent of all iOS 6 traffic is hidden. 27:05 It's now coming through in paid. 27:10 You can see this blue trend line falling down. 27:12 That's a good thing. 27:14 That means Google is giving us more referral data in paid search. 27:14 Because they're not doing it in organic, I just—I can't believe it. 27:18 I mean—so now you can get not provided in paid search, 27:23 and you can get the iOS 6 in paid search. 27:26 But you can't in SEO. 27:28 Wow! 27:29 Okay, Google. 27:29 Thirteen percent of all Google searches for our very ecom-focused clients—all Google searches are now hidden because of this issue. 27:30 And over a third are not provided. 27:38 Big issues here with data. 27:42 The other problem is this right up here—PLAs. 27:44 So shopping—the shopping feed went from Google shopping to product listing ads. 27:47 Those product listing ads I'm sure you've seen. 27:51 They're incredibly effective for Google. 27:53 They're incredibly effective at taking click share. 27:54 You see the light green bar going down. 27:58 That's the old Google shopping. 28:00 The dark green or blue going up—that's the new PLA. 28:01 The adoption has been swift. 28:04 Thirty-three percent now of all spend and non-brand paid search are PLAs. 28:07 So PLAs are huge, and they are taking a ton of click share— 28:13 not only from traditional paid search ads but also some organic. 28:17 Here's another one—more good news is the image search change. 28:21 So the old—this is the old image search. 28:24 You would go into Google and you would look at an image, 28:26 and what it would do is when you brought the image up that you wanted to see, 28:28 it would load in the background and gray it out. 28:31 It would load the site that it was hosted on. 28:33 So here you can see Sears in the background. 28:35 That actually triggered an impression on the site that was grayed out behind it. 28:37 What happened is with the new one, they don't do that anymore. 28:41 There is no impression triggered. 28:45 But what we're seeing is a plummeting of traffic for Google image search. 28:47 You can segment this out pretty easily in Omniture. 28:52 It's not so easy in Google Analytics. 28:54 But that's a loss of up to 150,000 visits a month from image search because of this. 28:56 So it's really important to know about this stuff. 29:02 We've got 13 percent of all Google organic hidden with iOS 6. 29:04 We've got 35 percent across the board not provided. 29:07 It's probably higher. 29:11 We have clients where it's 89 percent of their traffic is not provided. 29:12 We've got PLAs taking organic CTR. 29:15 We've got image search traffic change. 29:18 Google will roll out new products and change stuff all the time. 29:20 We've got to know all about that. 29:23 If you don't know about this stuff, 29:24 it could look like your campaigns are really doing poorly and your SEO program is actually down, 29:26 and it might not be that down. 29:30 You need to know the realities of this and set expectation with your team. 29:32 Make sure that you're thinking about the SERP now—not just about SEO and doing it for algorithms— 29:35 doing it for people that are clicking on them. 29:41 Maximize your CTR by thinking about the way titles are formed, 29:43 using schema for rich snippets, authorship, etc. 29:46 Estimate that lost data. 29:52 So for iOS 6, here's a link where you can back-fill that iOS 6 hidden data. 29:53 Annie Cushing is going to be talking a lot more about this. 29:57 So stay tuned for that. 30:00 Social for Ecom. 30:03 First let's set the stage. 30:05 I'm going to talk about social just in the context of SEO. 30:06 First of all, Facebook sends about 30x more traffic than Google+, so we know that. 30:09 Pinterest is a little bit—is pretty good. 30:14 They're up there. 30:16 But Google+ isn't that important, right? 30:17 Well, total referring sites—social sites—the ones right here are just less than 2 percent of all referring traffic—for our clients. 30:19 So it could be larger for very socially focused companies. 30:29 We're not a socially focused, but it's still not that much when you look at it. 30:32 So why is this so important? 30:38 It's important for the SEO. 30:39 When you think about Google+—this is rel publisher showing the G+ page for Urban Outfitters. 30:40 This is the thing we focus on right here. 30:45 This number—the number of followers. 30:49 We try to get that as high as possible because that is pure wallet share. 30:51 Those are people that are interested in the brand and are online and using Google+ and using search engines and are savvy. 30:55 So how do you get that up? 31:02 There's a few ways you can do it. 31:03 One is to get here—get on Google's recommended list. 31:04 When you get on that list, what we—this is what happened to Express—in June of last year they had about 200 followers. 31:06 By November, they had 300 thousand. 31:12 Then it started—it was rapid, rapid up and then it started to kind of flatten and level off. 31:13 Being there is good. 31:19 So get there if you can. 31:20 Sometimes you can work with the paid search side on Google, and you can talk to them and get an introduction over to Google+ folks, 31:21 and you can get on it. 31:27 You have to build a real G+ presence and really put good content in there and have it be valuable. 31:29 One cost that comes with this though is that brand visits start to deteriorate. 31:34 So that Google+ knowledge graph is starting to pull share from paid search ads and from organic, 31:39 and not provided will also go up because you're getting more and more people who are probably logged in searching. 31:45 Authorship, as Matthew said yesterday, this is rich, fertile ground, 31:52 and it's still early adopter. 31:56 We do not see many people leveraging authorship so get on that. 31:57 Here's an example. 32:00 "GQ" is not using it in its SERP. 32:02 "Esquire" is not. 32:03 "Men's Health" isn't. 32:04 The one is "Art of Manliness," and look how that stands out. 32:05 I think authorship—one of the problems with it though is Google can drastically change the SERP based on just very small query changes, 32:10 so this is classy bucks for men. 32:18 There's an authorship. 32:19 With suede bucks for men, it's all commercial. 32:20 You can see there's PLAs. 32:23 There's more commercial results. 32:24 There's only one "Wall Street Journal" non-commercial result. 32:26 So authorship is great for product reviews and guesting on ecom. 32:31 So ecoms always ask how should we use it. 32:34 Use it for your blog in reviews and use it for guest posting on other sites. 32:36 But Google can change the benefits based on how they weight results. 32:40 This is a really tricky one. 32:43 Say Joe is an author. 32:45 Joe has a prolific social presence. 32:47 He is writing for an ecommerce site. 32:49 He leaves, and he goes and works for a competitor. 32:51 What happens to that content? 32:53 Because of the way authorship works, it looks like Joe personally—he's got a lot of ownership over that piece of content. 32:55 The ecommerce site can remove authorship for Joe, but then they lose the benefit of that. 33:01 So what do they do? 33:06 Do they put another author in there? 33:07 Do they put a different image in there? 33:09 How do they do that? 33:10 There's a lot of legal issues around authorship for ecommerce sites that are tricky. 33:11 All right. 33:14 Three final take-aways and then I'm done. 33:15 Understand your target audience. 33:17 Understand that we're going for people. 33:19 We're not going for search engines. 33:20 We're just using search engines to get to people, so we need to stay in front of technical problems. 33:22 Do that. 33:26 Stay on the cutting edge of tech. 33:27 Understand data trends. 33:29 Know where traffic loss is happening based on product changes, based on Google changes, 33:30 and remember this guy. 33:35 But do not be him! 33:37 Thanks, guys! 33:40 [applause] 33:41 Adam, that was fantastic. 33:45 You've got about 100 things in that deck. 33:48 I've got about 100 questions for you. 33:49 Okay. 33:51 But we are out of time. >> [Adam] Yep. Thank you. 33:51 But just a reminder, your deck will be available for download on MozCon Live. 33:53 Under the agenda, you can share it with your team, and you're on Twitter. 33:58 I follow you. >> [Adam] Yes. 34:02 It's awesome. 34:04 Thanks, Cyrus. >> [Cyrus] Thank you, Adam. 34:04
You need to sign up for Treehouse in order to download course files.Sign up