Perfecting On-Page Optimization for Ecommerce Websites

Back in 2009 (was it really that long ago?!) Rand wrote a post titled Perfecting Keyword Targeting and On-Page Optimization, which is one of the most popular blog posts on SEOmoz. It is still referenced as much today as it was back in 2009. The core principles haven't changed that much, but there are some new additions to an SEO's toolkit when it comes to on-page optimization. Today I want to focus on what these new additions are in relation to eCommerce websites.

Elements of the page you should work on

I made the following mockup to try and visualise clearly all the elements of an eCommerce product page that are important for on-page optimization.

Let's get into more detail on each of these elements and see what we can do to take advantage and optimise for them, starting with the new additions since Rand's post in 2009. I've related the numbers in the mockups to the sections below; some sections do not have numbers because they are not visible on the page, for example META description.

 Customer Reviews

If you run an eCommerce website and are not collecting customer reviews, you are seriously missing out. Not only is this great feedback that you need to have to improve your business, but it is also an amazing source of unique content. Better yet, it is very scalable across large websites, which means you can get lots of content onto lots of pages.

Quick tips for collecting and using customer reviews:

• Build or buy a system to automatically email customers a few weeks after purchasing and ask for a review
• When getting off the ground and trying to get volume, offer incentives such as a discount on their next purchase in exchange for a review
• Don't worry about publishing negative reviews, customers aren't silly and can tell when reviews are a bit too positive

Also, if you are worried about things like this having a negative effect on conversion rates:

See if you can customise your review system to not show this message on products that do not have reviews. Set a threshold so that when a couple of reviews are received, reviews are shown on the product page.

Added benefit: microdata

You also need to make sure you are marking up these reviews with relevant microdata. This will give Google more context about your content, as well as giving you the chance to improve click-through-rates from search results like what we see in this example:

The use of review microformats is increasing all the time so there is an argument that you are not standing out anymore if all the other results have the same type of markup. You could even argue that to stand out you should take them away :)  

 Product Videos

I'll admit that this is a tough one to execute, but it is one that I feel is very worthwhile for eCommerce sites. There are many websites already adding videos to their product pages, but they are not always doing it in the most optimal way. A great example of the right way to do this is Zappos who now have over 50,000 product videos.

There are a few benefits to having videos on a product page. One of which is helping make your product pages more link worthy and rich in content. Good quality videos demonstrating use cases of products could also help conversion rates (particularly for high-end, technical products) but I can't provide evidence for that unfortunately.

Another added benefit as you'll see from the screenshot above is how your search results for product pages can stand out from competitors. I've seen loads of eCommerce stores who have videos on the page but are not embedding or marking them up in the correct way.

By far the best system I know to embed and optimise your videos properly is Wistia, which SEOmoz use for Whiteboard Fridays. These guys have a great system and are always improving how things work and adding new features. We've used them on a test site or two at Distilled and got video snippets showing very quickly.

I could talk more about using videos to aid SEO but Phil did a great post that covers pretty much everything you need to know here. He also did a presentation on video SEO and you can see the slides over on Slideshare.

Rel="next", Rel="prev" and view all

One of the problems that always crops up on large eCommerce sites is how to efficiently deal with pagination. You can have product categories that contain thousands of products that span many pages. You want to make sure that all of these products are indexed and regularly crawled, but at the same time you don't care too much about the paginated pages ranking or having too much link equity.

Since Rand's post of 2009, we've been given an additional way of handling pagination. Namely the rel="next", rel="prev" and "view all" attributes. This markup can help Google better understand pagination and pass link equity to key pages. Google gave some good instructions on how to implement these attributes here and here which you can take a look at.

There are a few other ways to handle pagination, which Adam Audette explains very well in this post on Search Engine Land.

Microdata markup and Schema.org

Another new tool that is available to us now is the use of microdata and the support of the Schema.org vocabulary by the major search engines. That announcement back in June 2011 was quite exciting but didn't really live up to expectations and Google seemed pretty slow in showing this support in their search results. However this seems to have changed and we are seeing more and more examples of Google using this data now.

Bringing this back to eCommerce, there are a few types of markup you can use on a product page which you can see documentation on here. This page also contains details of review markup that I talked about above. Not all of the properties on this page will be applicable to you, but here are some tips on how to use this:

• Only choose the properties that are relevant to the product attributes you have
• Take development time to integrate these properties into templated elements of your page, so that when you add new products, they are automatically marked up
• Add notes to your analytics package when you put these changes live so you can monitor any improvements

 Q&A Content

Another big opportunity for eCommerce websites is the integration of question and answer content focused on products. As mentioned above, eCommerce websites have always had the problem of getting unique content onto product pages on scale. Question and answer content can help solve this problem and gives you great scope to get user generated content onto lots of your product pages.

There are a few benefits to integrating this type of system:

• Scalable, user-generated content published onto product pages
• Improving ranking for long-tail terms and question driven keywords if the content is crawlable
• Possible improvement in conversion rate if customer concerns are addressed in the answers
• Possibility of encouraging brand evangelists and even bringing in some gamification principles to help motivate users

Here is a live example from Jessops:

I personally feel like there is an opportunity for Quora here if they wanted to explore this space. Many retailers will be looking for this type of system and Quora may be able to offer something that helps them reach the critical mass of content they'd need.

 Social sharing buttons

I'm a little skeptical about whether social sharing buttons on product pages are a good idea. The goal of a product page is to get someone to buy, not to get them to tweet or like the page. Sure these social signals can help, but personally I'd rather not distract people from buying my product. For me, social sharing should be encouraged at different points in the buying process:

• After the point of purchase on a thank you / confirmation type page
• Email follow up and correspondence - follow us on Twitter, like us on Facebook etc
• After a review has been published - give the reviewer the option to share their review

There is an alternative use of social buttons, which I haven't seen or been able to test on a client site yet. But I wanted to share it anyway. It builds upon the code that Tom Anthony talked about here which allows you to detect if a user is logged into Twitter, Facebook or Google+ whilst they are viewing your website.

If you can use the code that Tom created to detect if a user is logged into Facebook for example, you could show that user a custom message. This could be anything you want but it could be something as simple as encouraging them to like your page in exchange for a discount. This not only gets you the like but also increases the chances of the user converting after giving them a discount.

Tom quickly tested this theory on a test site which you can see a screenshot from here:

You can put whatever message you want in here, this is to demonstrate what could be done if you think a little out of the box and not just put social share buttons on a page because that is what everyone else does.

Page Speed

Again, this is something that has become more of a focus since Rand's blog post. Speed has always been important but SEOs sat up and took a lot more notice when Google confirmed it was a factor in the algorithm, albeit a small one.

For me, an eCommerce site should care about site speed because of its effect on conversion rate rather than rankings. A user is not going to hang around waiting for your product pages to load and there have been some good studies that show the positive effect a fast loading page has on conversion rates.

Bottom line is that you should care about site speed for your users rather than SEO. Here is a good guide for improving site speed written by Craig at Distilled.

Open graph tags

Another new addition that you can add to your eCommerce pages is the open graph tags. These tags allow you to be much more specific with how your content is shared on Facebook. As Facebook is such a huge platform with a lot of potential for traffic, you need to make sure that you are doing all you can to optimise for it and specify how your content should be shared.

They are also pretty easy for you or a developer to setup and put live. The tags sit inside your header so you will need a flexible CMS or a good developer to make these additions for you. On an eCommerce site with lots of products you'll probably need a developer to setup the tags so they scale across all of your products and use the correct elements of the page.

Here are some more articles that help with the use and optimisation of the open graph tags:

• Facebook open graph protocol in Wordpress
• Web developers cheatsheet for open graph tags
• How to optimise Facebook open graph tags

 Search options

Ideally, a user should never need to use a search box on your website because they will be able to find their way around using your navigation. But there are going to be times when this doesn't happen and there are users who will just prefer to search. I think that a search box on an eCommerce website is essential and you should use the data that it gives you to improve your website and customer experience.

Here are some tips for using a search box:

• Make sure you are tracking searches using your CMS or this feature of Google Analytics
• Monitor how many people who search and then leave the site straight away - try to lower this number
• Check your search results actually return good results
• Make sure your search function still works when you try singular and plural keywords - particularly with an eCommerce site this is important
• Pull in special offers and discounts related to the searched for keyword
• Pull in product images next to search results, I like how Apple do this:

 Clear call to action

Essential for any eCommerce website. Your ultimate goal is to sell a product so you need to make the call to action as clear as possible. Make sure you are running experiments on your product pages to test and improve conversion rates. Many eCommerce stores focus a bit too much on getting more traffic via SEO and PPC, whilst a quicker way to get more revenue is to get more out of the traffic you already have by improving conversion rates.

Even if you are not actively doing conversion rate optimisation, you should at least be measuring as much data as you can from your site, in particular your product pages which are ultimately the most important pages for an eCommerce website.

Tools you can use to measure and improve calls to action:

• Crazy Egg
• ClickTale
• Google Analytics event tracking

Just get one or two of these tools setup and start gathering the data, once you start gathering the data, you are in a much better position to start caring about it and setting targets against it.

 Trust signals

You are asking people to enter their credit card details on your website. They need to be able to trust that you are a genuine company and that their personal details are secure. You can do this on the product page and enforce it again throughout the checkout process. These are the types of trust signals you should be trying to incorporate into your product pages:

Also make sure these link to secure certificates where possible so that users can go and verify what you are saying. Be sure to check regularly that these links still work - the last thing you want is this link being broken or expired!

 Breadcrumbs

These are underestimated in my opinion, both in terms of customer experience and with SEO. They can be a great way of helping the customer navigate around your website and really help your internal linking.

On an eCommerce site, breadcrumbs can be a bit complicated because there are often multiple ways of getting to the same product page. So the potential breadcrumb trail on a product page could look different depending on which categories and sub-categories you navigate through. For me, the benefits of doing anything too fancy are not big enough to warrant the time. So I'd recommend using one breadcrumb trail and sticking to it. If you are concerned about user experience, you could make the users breadcrumb trail cookie based. But this isn't always worth the development time so you should assess how valuable it is for your customer experience.

 Images

Crisp, clean, high quality images are necessary for any eCommerce website. The users engage with what they can see and will often be put off if the images are very bad. Here is a great post from Kissmetrics that gives some examples of how to optimise images for conversion.

Something I'd highly recommend for an eCommerce website is showing use cases of the product within the images and not just the product itself against a plan background. As much as I like IKEA, I don't like the plainness of their images sometimes:

I'd much prefer to see products like this shown how I may use them if I buy them and in the setting of a living room for example.

From a pure SEO perspective, you'll want to make sure you are doing basic image optimization to capture traffic from Google image search where possible. Here are a few tips for this:

• If possible, use descriptive filename e.g. wooden-oak-table-12345.png instead of 12345.png
• Add ALT text to all product images - it is quite easy to make ALT text the same as the product name automatically in the CMS
• Create and submit an image sitemap to Google Webmaster Tools

 META Title

I shouldn't have to go into much detail here as to the importance of this. Something to bear in mind for eCommerce websites is that you are generating META titles for potentially thousands of product pages. It just isn't feasible to customise each and every one of these, so you should have these auto-generated by your developers based on a template that you give them. For product pages, this is probably just going to be the product name followed by a small call to action or USP. For example including something like "Free Delivery" could work well for improving click-throughs from search. The key really is to try and avoid masses of duplicate META data.

Top tip - an eCommerce website is usually driven by some kind of database which will have various attributes (fields) for each product. A good developer will be able to use these fields to populate other parts of the page dynamically, for example a META title or description. Bear this in mind when writing your META data templates and use these fields if they are available to you.

META Description

Whilst the META description has minimal effect on rankings, you should be optimising this for improving click-throughs from search results. Ecommerce sites are in the perfect position to include lots of information, calls to actions and USPs into the META description. As mentioned above, the META description could be auto-generated based on a template that you provide to a developer. This could include database fields such as categories and sub-categories.

 Product description

In a post-Panda world, it is very important to make your product descriptions unique. Taking descriptions straight from manufacturers or product feeds does not differentiate you at all from the hundreds of other retailers who sell the same product. Spend the time and resource making these unique and engaging and make sure you include the USPs of your offering - such as free delivery or lowest prices.

 Page URL

Again, this is pretty basic SEO but there is one key thing to remember with eCommerce sites. You should not include categories or sub-categories in product URLs, especially if there is more than one way to find a product, for example if it is in more than one category. This can lead to duplicate product pages. You can fix this with rel="canonical" tags but it isn't really ideal.

Best practice is to just use product name and a code as the URL, for example - www.example.com/product-name-12345. The reason for the addition of a number in the URL is to cover yourself against similar product names - not usually a problem but worth trying to prevent.

 H1 tags

It is debatable how much H1 tags matter anymore and some studies from SEOmoz have shown that they do not have a lot of impact on rankings. However I feel that for the time it takes to optimise this, it is worth doing and certainly isn't going to hurt you. It is also good to have clean markup of the page so that if for some reason someone browses a page with CSS turned off, the page still has a logical structure.

For an eCommerce product page, I'd recommend coding your page template so that the product name automatically becomes the default H1 tag for a page. This should help to eliminate duplicate H1 tags across the website and will automatically optimise each page you publish.

 Phone number

If you can provide a phone number, do it. Not only to help in terms of customer support, but also as another trust signal. If we think back to what Panda was trying to achieve, one of the questions was "would you trust this website with your credit card?" and one factor that certainly helps inspire trust is a phone number.

A pro tip here for eCommerce websites - if you have a customer support team. Keep track of your abandoned baskets in the checkout process and if you have captured the customer's phone number, take some time to get your support team to phone and see if they can see what went wrong. This not only gives you a chance to get the sale, but you can also get feedback on your checkout process and see what barriers to conversion there may be.

 Company details

Particularly relevant for companies who target local markets, giving Google more signals of your location can help rankings for those types of keywords. You can also use a few bits of Schema.org markup to give some extra context to the content. It is also another trust signal for Google and users to look at.

Conclusion

Well that is about it, I hope that has given you enough to work on to try and improve your eCommerce product pages. To wrap up, here are some more great articles on eCommerce SEO, many of which are from this curated list of eCommerce resources by Everett Sizemore:

• Building Deep Links into Product Pages
• 65 Ways to Improve Online Sales
• 24 Ecommerce Development Tips

As always, I'd love to hear your comments and feedback or ping me on Twitter to ask more questions.

Negative SEO: Myths, Realities, and Precautions - Whiteboard Friday

Howdy, SEOmoz fans. Welcome to another edition of Whiteboard Friday. This week we're talking about a very concerning and controversial topic - negative SEO. Now, negative SEO has a number of meanings. I want to walk through them and get to some points. If you've been paying attention to the Twitter-sphere or the SEO blogosphere over the past week, two weeks, there's been a lot of discussion around negative SEO, particularly backlink pointing to bring down sites. I will get to that, but first I want to start with some of the classic ways that negative SEO could potentially hurt you.

The idea behind negative SEO is that rather than doing good, positive things that will promote signals in the search engines that bump up your rankings, there are ways to do bad, terrible, negative things. Now, obviously you could do these on your own sites, but hopefully you're smart enough not to do that. There may be things that other site owners, webmasters, marketers, or black hat SEO's, mostly we're talking about black hat SEO's, spammers, and even people doing very illegal things to bring down your website in the rankings or to even take your website offline.

There are classic types of things, like malware, hacks, and injections. So this is the first one I'm going to talk about. Basically, what we're saying here is that you've got your site, it has some pages on here, and hackers may find security vulnerabilities in your site, in your FTP logins. It may be a WordPress install. Earlier this year I had a hacker essentially come in and inject spam and malware onto my personal blog at RandFishkin.com/blog. The idea is that they all inject spam, links to spam sometimes, sometimes very subtly. They will make changes to your site. One of the classic examples of this is someone going and editing your robots.txt file to block Google bot or to restrict all IPs from a certain range, or those kinds of things. Obviously, that's going to take your site out of the search engines. Or inject viruses or malware that will install itself on computers that visit you.

Unfortunately, I was actually visiting MozCation.com, which Gianluca Fiorell, one of our Pro members from Spain - he's Italian but from Spain - had set up last year to promote MozCation in Barcelona, in Spain. Unfortunately, it looked like some spammers had injected some malware on that site, and it had been on there a little while. I think he's taken care of it now, but these are the types of problems. What you'll see is a download will go into your cache, and sometimes Microsoft Security Essentials will alert you that that's happened, hopefully if you've got it installed. So this is something to watch out for. You want to close those security holes.

The other kinds of things to watch out for is spam reporting. Sometimes a lot of people, unfortunately, in the SEO-sphere still do manipulative kinds of link building. Obviously, most of the people who watch Whiteboard Friday are not in that group, but some of you probably are. Maybe you buy a few directory listings. You go on Fiverr and you buy some cheap links. You find some spam through some forums that potentially works. You're doing sorts of things that are on the grey hat/black hat borderline, in terms of link acquisition, and sometimes you will see that your competitors might spam report you. So this guy's going to go over to Google and maybe he'll leave a threat at the webmaster forums, or he'll send it through a spam report in his Google Webmaster Tools. A lot of this spam reporting, I think they said they get tens of thousands of spam reports each month, I believe it was. Actually, fewer than I'd expect, but a lot of people do report spam to Google. These might be your competitors. These might be other webmasters. They could just be random people on the Internet who are like, "Why isn't this site ranking here?. This looks terrible. I don't like this."

When this happens, Google might take a closer look at your backlinks, and obviously this might bring you down. There are arguments about the ethics inside the search engine industry. Personally, I think that removing low quality crap from the Internet is all of our jobs, and I like to be part of that. I think that it's a good thing to make the Internet a better place, and if you're not making the Internet a better place, I hope that you're not doing web marketing because it makes the rest of our industry look bad.

However, certainly reasonable minds can disagree. Aaron Wall, from SEO Book, who I highly respect, who I grew up with in this industry and think the world of, takes a complete opposite view. He thinks that because I support disclosing spam and manipulation to Google and to search engines that this makes me a bad person. That's too bad. That's frustrating, but I think reasonable people can disagree. Certainly whatever angle you are on, on this, you should at least be aware that this stuff happens and know that it's a potential risk, particularly if you're doing highly manipulative things.

The last one I want to talk about is actually the biggest one and probably the most important and the most salient and relevant to what we've been talking about today. That is pointing nasty links to your website. Now this has been something that a lot of webmasters have been discussing actively over the last couple of weeks in this sphere, essentially kicked off by a forum thread on Traffic Power Forum. I haven't previously spent a lot of time there, but it's a very active forum populated by a wide mix of white hat folks, grey hat folks, some pretty dark black hat folks, which I'll show you in a minute.

Two members there, Jammie and Pixelgrinder, hit two different websites. One is called SEOFastStart.com, that's owned by Dan Thies. Dan, of course, early keyword research guru in the SEO space, big industry mover and shaker. Spoke at a lot of the early search engine strategies conferences. I've met him a number of times, really good guy, solid guy. He complimented Matt Cutts, the Google Webspam Chief, on the search quality team. He complimented him over Twitter on knocking out some spam. Some people on the forum felt that it was, I don't know, in poor taste. Right? Essentially they felt that because he was being complimentary to Google for kicking out webspam, that he should then be the target of this negative SEO. The other site was NegativeSEO.me, which was essentially a website offering services to get someone banned from the search indices, and this a little concerning in and of itself.

Now the thing that's interesting about these sites, and Dan admitted this about SEOFastStart. Not a very big site. Right? Not a lot of great brand or link signals. Potentially some small amounts of not wholly white hat types of activities already happening around these sites. So we're not talking about (a) big brand sites, or (b) sites that have no idea about the SEO world and aren't doing anything manipulative and are clean as the driven snow. These are a little off that track. These were both hit by these guys, at least presumably, according to the forum thread, and lost a lot of their rankings.

When I say hit, what I mean is this type of thing happens. So here's your site.com up here. Right? Essentially, what's going on is you've got some nice white hat, editorially given, earned links, high quality stuff, and that's great. Then there's some kind of this dark cloud of black hattery, spammy, manipulative posts. They talked about a number of things, XRumer blasts, buying links on Fiverr, buying links from some link networks, pointing some links that they had seen get hit on other sites at this site, and essentially trigger this loss of rankings. Now, they didn't get banned from the index, but they fell from, I think Dan Thies' site in particular fell from ranking #1, for his personal name, to number30, 35, somewhere around there, and hits like that similar across both these sites.

The second example was another forum thread started by a user with the user name, Negative SEO, and that was for the domain JustGoodCars.com. Now again, Just Good Cars unfortunately looks like they were doing a little bit of things that might be construed as manipulative, even prior to this attack on them by the Negative SEO guy. Some links that were of questionable sources or how they were acquired, and then a big network of websites that were all pointing back and forth to each other from many different pages on these many different sites. This guy took it upon himself to say, well they were . . . I guess this website had been complaining in the Google webmaster forums about some other sites outranking them, so this person took it upon themselves to do some pretty nasty, evil stuff.

Now I can't support this in any way. I'm frustrated that unfortunately this is a part of our world. But you should be aware of it, because what they did was creative, almost to the point of ingenuity, but definitely dark and evil, maybe even bordering on illegal depending on the legalities. I'm not really sure. Here's what they said they did. Of course, I can't prove that they actually did these things, but here's what they said they did. So they did go do a lot of manipulative, nasty backlinking to the site from a lot of those sources we talked about. They mentioned a few XRumer blasts. They posted a lot of duplicate content. They set up fake WordPress splogs, essentially a spam blog, and then they re-posted the content that existed on JustGoodCars.com on tens of thousands of pages across the Web so that Google might say, "Oh, well why is this duplicate content?" I don't know that that's actually highly concerning in and of itself. A lot of people copy content from all over the Web for both good and bad reasons.

Then they did something that's really nasty. They went to Fiverr and they asked for people to post fake reviews to Google Reviews to make it look like Just Good Cars was manipulating Google Reviews, and actually got them thrown out of that program. According to the forum post, anyway, that's what happened. They got their stars and their Google Reviews and their ratings removed, and all that kind of stuff, which that's whew, that's really low. That sucks if that's what really happened.

It's even more terrifying, but they sent fake emails. They set up email addresses that looked like they came from Just Good Cars, and sent fake emails to websites that had posted good editorial, positive links, saying, "Hey, you should stop linking to this site. There are these problems with it. We're requesting a DMCA take down action against it. Our attorneys will be in touch if you don't remove your links." Those kinds of things. So really just, oh man, that's really evil. But stuff that we definitely need to be aware of in terms of the world of negative SEO and what this kind of stuff can happen.

Now, it's very tough to verify anonymous users on an anonymous forum posting and whether all of this stuff actually happened, but certainly the ideas behind it are very concerning. What I want to express today is that there are some things you can do on your site that will make you higher risk and lower risk to these kinds of things.

Higher risk is going to be, like some of these other sites, you've already done a little bit of manipulative linking. Right? You've already done some spammy stuff. You have manipulative on-site stuff. Meaning for example, like Just Good Cars there's kind of that footer with all these links pointing to all these other places. This was mentioned in the forum thread. So I'm not giving away new information here, but there's stuff on this site that looks like it might be not wholly kosher, not wholly white hat.

Your site has few high quality brand signals. High quality brand signals, things like lots of people searching for your domain name and brand name. Lots of mentions of you in the news and press, in outlets that are high quality. Lots of offline sorts of signals. Lots of user and usage metrics types of signals. Lots of verification kinds of things. Using high quality providers of everything from the IP address, where your website's hosted, to the domain registration link, to the services you might have installed on your site, Akamai or any of the CDN networks suggest you're very popular. Any type of signal like this that looks like a highly brand intense signal.

Lower risk is going to be the opposite. Right? So things like a totally clean backlink profile. Never done any kind of manipulative linking, at least not intentional outbound backlink building. Don't forget, everyone's going to have some spam links. Even if you've never done any manipulative backlinking or any backlinking or marketing of any kind, you will have some bad backlinks, because the Web, just there are all sorts of weird crawlers and bots that host links all over the place. It's fine. Don't sweat those. It's the normal volume. Things like having a beautiful, elegant, high quality UX. A great UX is a fantastic defense against a lot of spam and manipulation. It's even a great tactic for folks who are trying to do SEO. It's just a great signal in general. Right? Having a great UX is going to get you more conversions and more people using your site. Anyone who is browsing your website, say, from the Google Search Quality team or the webspam team, or the Google reviewers, which Google hires, or from Bing, any of those folks who are looking at your site are going to say, "Oh this is clearly a great site. We want to have this in our index."

If you review some of these other sites, you can take them or leave them. One that does not feel very SEO. I think you all know what I mean. There's sort of that sixth sense of, boy, they're doing a lot of things on the page and off the site that don't feel like they're natural, don't feel like they're for users. Whenever you have that sixth sense around a site, that's going to put you in a higher danger category. Not doing that, having that very natural sort of site, you can target keywords, do a good job with your titles, do a good job with your content, do a good job with your internal linking, but make it feel very natural. I'll give you good examples. Amazon, very well SEO'ed, but doesn't feel SEO'ed. Zappos, doesn't feel SEO'ed. Even SEOmoz, it doesn't feel very SEO'ed, but it's doing a good job. TechCrunch, doesn't feel SEO'ed, but ranks phenomenally well.

Finally, having those strong brand signals, the branded searches, lots of people searching for your brand name specifically. Good links, good mentions, good press, good user and usage metrics, all these types of things are going to protect you from a lot of these types of spam attacks.

That being said, there's nasty stuff that other people can do. So you want to (a) keep your eyes wide open. Make sure you're registered with Google Webmaster Tools so you can get any of these warnings ahead of time. If you happen to see an influx of really nasty looking links, you might want to send a preemptive reconsideration request to Google saying, "Hey, we don't know where these came from and we have nothing to do with this. We just want you guys to know that this is not our activity. Please feel free to disregard or not count these links." 99% of the time Google is not going to say, "Oh these bad links that are pointing to you, we're going to count those as reducing your SEO and bringing you down in the rankings." They're instead going to say, "Oh well, we're going to ignore these. We're going to remove the value that these pass." They're not going to pass PageRank or anchor text value or link trust, or whatever it is. We're just going to count the good stuff.

I remember being in a session, this was years ago, probably five or six years ago, with Matt Cutts, the head of webspam for Google. He was looking at a site on his computer, and the person asked about their website from the audience, and he said, I see, I don't remember what it was, 14,000 odd links pointing to this site, but Google's actually only counting about 30 of them. That's why you're not ranking very well. Most of those links we've removed all the value that they pass. So it's not that they were having those bad links hurt the site. It's just that they're saying, "Oh these are not going to pass any more link value."

Now, what I would suggest here is, if you see stuff that looks like manipulative and negative SEO, you just be careful. We are trying to do some things here at SEOmoz to help with this. One of the things our data scientist, Dr. Matt Peters, is working with some folks here at Moz to build a large list of spam so we can do some classification, and eventually inside the Mozcape index, which will appear in Open Site Explorer, show up in your Pro-web app, show up in the Mozbar, we'll try and classify sites to say, "Hey we're pretty sure this is spam. This looks like the kind of thing where we've pattern matched and seen Google penalize or ban a lot of these sites." We're also trying to build some metrics to show what are really good, high quality, and editorially given sites. So domain authority and page authority already exist to try and do that.

Then, we're also running some experiments where I've offered up my personal blog, which is a relatively small site, probably has as few links as any of these, probably fewer than Just Good Cars, RandFishkin.com, to see if some of these nasty folks, who are hitting and taking down sites with negative SEO, would like to concentrate their focus on my sites. For two reasons, number one, we'd be very curious to see it happen, and number two, we can certainly afford the hit. We offered up SEOmoz as well. Most people seem to think that SEOmoz is not a good target. It won't actually be taken down.

We're going to run some experiments internally as well on this front and hopefully be able to disprove that negative SEO is a common thing that works very well. I'd hate to see an industry spring up like this. I think that this type of activity, particularly some of these really nasty things, are just an awful part of being around the black hat spam-sphere. I hope that it's something that we can defend against. I hope you'll join me in contributing. I look forward to your comments. If you've seen stuff like this before, please do feel free to talk about it either anonymously or openly in the comments. I will see you again next week for another edition of Whiteboard Friday.

Community: The Inbound Resource You Forgot About

Remember a time when 'community' meant the neighborhood or town you lived in? Or when a business only had to be concerned with making customers happy when they were inside their store? It's sometimes hard to remember what life was like before online communities became such commonplace. But the fact of the matter is that online communities are where we spend most of our time these days (especially if you're in search marketing... #amiright?).

We talk a lot about Inbound Marketing and how it encompasses areas such as SEO, Social Media, content, blogging, email, Public Relations and Q&A/forums (among others). Now, let me ask you this: What's the one thing that all of those have in common? (You probably know where I'm going here) That's right, Community.

Ok, I admit it, I'm surely a bit biased here. However, I really want you to think about this. At the base of everything we do as marketers, our community (whoever that may be for your particular organization) is right there, standing tall.

The Value of Community

What does "community" mean to you? Or perhaps a better question should actually be: What should "community" mean to you? I want you to think about the people who visit your site, participate in your forums, and buy your products or comment on your Facebook wall. They are the people who you want to find you in the SERPs, who you send emails to giving them discounts and they're the ones you hope will retweet and share your content. You write for them, you create products for them and you (may) want money from them. Yep, they're pretty damn important.

Your community is important in so many ways. They are your:

• Brand advocates
• Software supporters
• Members/Users
• product feedback specialists
• content generators (UGC BABY!)
• link builders
• critics (hecklers?)
• lovers
• sharers of content
• forum participators
• website uptime monitoring service ;)
• reason for eating lots of cupcakes (yea, ok. Maybe that's just me.)

Determining the actual value of your community isn't so cut and dry. Sure, we all want to put a dollar value to everything, but you can start with these steps:

1. Figure out who the community is in your organization.

Who are your community members, what do they care about, why does it matter, where do they hang out, and how are you going to interact with them? Are they the people who participate in the forum on your site? Are they the ones who read your blog? Are they the people who buy your products? Are they all of the above?

You can get this information in a number of ways:

• Look at your analytics! Where are your users coming from, and where are they going next? What keywords are they searching for to find you? Who are they based on demographics?
   
• Facebook Insights. If you're active on Facebook why not dig into your user data and see if it matches up with your analytics.
   
• Dig into your own data. Do you require sign up on your site? Do you ask for gender or any other useful information that can help you determine who your community members are?

2. Figure out what your community really cares about.

Do they care about sharing your content? Will they spend endless hours in your forums? Perhaps they just want a daily email update from you (never to hear from you otherwise). But how do you figure out what they care about?

Ask. It never hurts to ask. Add a poll to your site, send them an email, or ask them on your social channels. We do this all the time here at Moz and getting feedback from the community helps us grow!

3. Determine how much time/energy/money you're putting into your community.

Think about where you're putting your resources. Do you spend your time creating blog posts? How much time do you want to devote to social media? Is someone managing social media full time? What about SEO and content? Do you pay outside contractors to help in the forums or write content for you?

You want to know this, so you can then determine if you're spending resources in the right areas (see next step ;-).

4. Are you spending your time/energy on the things your community actually cares about?

Boom. You've figured out who the community is, what they care about and how you're currently spending your time. Now you can determine if you're utilizing your resources well.

You want to know if you are wasting time creating blog posts (that no one is reading), because they care more about writing their own content. Do you scour the internet looking for content to share on Twitter, only to realize your community doesn't really get into Twitter. Should you focus your energy on beefing up your emails because your members like to get info that way?

5. Rinse and repeat.

This isn't a one-time process. You need to constantly be thinking about how you can leverage your community in the right ways. Don't stop simply because you found something that works for now. The biggest takeaway here is also that you need to determine what works for YOU. You can read all about how others manage communities, but it's up to you to set your own course.

Here at SEOmoz, our community is always on top of mind as we develop software, create resources and share content. We believe strongly in keeping our community alive, strong and continually growing. We want to challenge you, please you, help you and whenever possible, make you laugh. So what is the value of the SEOmoz community? This is the best part, it's invaluable. Because without our amazing community, we're just another software company. But as you well know, we're more than that, and that's because of you. (Can we say "job security")?

Community Resources

Before I let you go work on determining the value of your community, I wanted to give you some great resources on building, managing and keeping a strong community:

• Community Roundtable - Be sure to follow their tweets as well as they tweet lots of great info about managing communities.
• #cmgrchat - This is a great chat that runs every Wednesday at 2pm Eastern. They have different topics each week, and I've learned a ton participating.
• My Community Manager - Holds a weekly Google+ Hangout on Friday's at 2pm Eastern where they discuss various aspects of community management. This Friday, I'll actually be talking about why SEO is important to Community Managers. :) (My worlds COLLIDE!)
• Some valuable posts:
  • How Are You Measuring Your Community? (Older, but still quite relevant)
  • Community section on Inbound.org (I post all my favorite community posts here)
  • Have any more? I'd gladly post them here!

Ok. Now it's up to you! I'd love to hear how you value the community and the steps you've taken to figure that out. Do you think I've missed anything or should add any steps? Why is community important to you?

---

By the way, I'll actually be talking about Community Management as a part of Inbound Marketing at Mozcon this year. I hope to see you there!

An Open Letter From a Frustrated Outsourced Link Building Author

Outsourcing your link building efforts can work, but as a Textbroker author I can tell you right now, most people aren't doing it right. The majority of clients who submit articles are still living in a world before the Panda update, and a striking number of them still use tactics that haven't even worked since Google's inception.

I'm going to tell you now, what you're about to hear isn't pretty. We're going to take a visit to the underbelly of the outsourced link building world. I'm going to get a little rough here, but odds are you'll learn something.

How Much Do Your Link Building Articles Cost?

About 1.2 cents per word? If you're really looking for quality, you're paying 2.2 cents per word. I'm a four star author, so the articles I write cost 2.2 cents a word. But I only see 1.4 cents for each word I write. Let's do the math, shall we?

To make $10, I need to write 714 words. I am a native English speaker who lives in the first world. I'm not willing to work for less than minimum wage. What does that mean? It means I'm not going to spend much more than an hour on a 714 word article.

I've chosen this source of income because I'm a college student with a bizarre schedule and the hours are convenient. $10 is not a lot of money to me.

If you ask me for an article about [your keyword here], and your description reads as follows:

'Please write an article about [your keyword here]'

then you should know something. I won't write you a decent article. Odds are, I'll reword something that I come across on EzineArticles. You know, that site that got slammed by the Panda update, and knocked down all of the sites that got their links there?

There is only one subject that I write decent articles about: SEO. That's because I actually know what I'm talking about, and I can recite most of the stuff from memory. I'm even up to date on the subject, because I regularly educate myself about it. Of course, those still pale in comparison to the articles on my own site, which take hours or days of research and a willingness to actually communicate with science experts.

The thing is, I could write good articles about other subjects. It's just that Textbroker's clients would need to learn how to streamline their outsourcing processes. And nobody seems to be doing that.

Allow me to offer you some advice.

You're Still Keyword Stuffing? Really?

Some of you might be shocked by this, but I am routinely being asked to stuff my articles with keywords. It takes up more work than it ever should for me to sift through the article requests and find one that isn't asking me to jam the same keyword 12 times into a 400 word article.

Then there's the client who only asks for the keywords to be repeated 2 or 3 times, but wants five or ten tangential keywords to be included in the text. 'But that's only an 8 percent keyword density!' No, it's not, because each one of your keywords is three to five words long, and pretty soon a quarter of your article consists of meaningless keyword phrases. I can't tell you how often I've written an article where I had to include at least one keyword in every sentence.

In case you think I'm joking...

(That's a 20 percent keyword density. MINIMUM. I repeat. That's at least one in every five words.)

Forget Panda. You think that's going to pass a manual review?

It gets worse. There's the client who wants me to write about 'dog leashes seattle.' Please tell me how to use that phrase in a sentence. Textbroker now allows you to give writers the option of using connecting words, so I could write something legible like 'Dog leashes in Seattle.' Inexplicably, very few clients use this option.

Presumably, the clients actually believe that including an exact keyword in their text is preferable to legible writing. Inevitably, I am occasionally forced to write things like 'When it comes to dog leashes Seattle is the place to buy them' even though it makes no sense to write this, ever, and it completely ignores proper comma use.

More often than not, I have no choice but to say something like 'Looking to buy dog leashes Seattle?'

Please. Clients. Each article should be about one keyword. If it's in the title and it's in the article, and it's what the article's about, congratulations, you've met Google's keyword density requirements.

Don't Ask Me to Do Research

It might sound like I'm whining, but you need to think logistically here. As I said before, if it takes me much more than an hour to make $10, you're paying me less than minimum wage. That means one of two things. Either you're not going to get any research, or you're going to get research from somebody who does not speak native English.

There may be exceptions, but they'll burn out fast.

At 1.4 cents per word, you're not paying me to do research. You're paying me to type. I've had enough experience that I can offer you a bonus: actual writing infused with emotion and good article structure. But research? It's not gonna happen.

'But I ask that only people who are experienced with the subject should write about it.' Good for you. That's why my SEO articles are actually good. But if you really think anybody on Textbroker is actually much of an expert on anything else, you're deluding yourself. The experts are writing elsewhere, or doing something else that makes a lot more money.

Am I saying that you should accept crap articles? Of course not. I'm saying that if you want your articles to be well researched, you're going to need to handle the research.

What would I do if I had the budget to outsource link building articles?

• I would spend a week scouring the web for facts about my subject.
• I would visit my local library and read a book about the subject.
• I would talk to industry experts about the subject, and leverage those relationships later on.
• I would subscribe to blogs about the subject, go to Reddit and visit subreddits on the topic, and join Facebook pages about the subject. I would go to StumbleUpon and use their search feature to stumble the topic. I would search for the keyword in Twitter or FollowerWonk and follow those who tweet about the subject.
• I would upload all of this information out of my brain into a list of a hundred or so of the most pertinent bullet points on the subject.
• I would write the most stellar article in the world on my subject.
• I would pick five or ten random bullet points to go into each one of my outsourced link building article requests. I would ask the author to use each one of these facts in the article as the basis for a subtitle, and I would ask the author to react to each one of these bullet points with their own personal view on things. Such an article could easily take up between 400 and 800 words without using filler words. It would also contain more genuine content than anything else I could buy, short of contacting a professional author.

In case you're curious, this is a virtually unlimited source of link building articles about your subject. When choosing from a list of 100 bullet points, there are literally more than ten trillion ways to pick ten of them, no matter what order you present them in. No need to write the same article about the same ten things over and over again.

Of course, I'd probably play it safe, and 'only' buy 100 articles on the subject, then move on to learn another 100 bullet points, write another stellar article, and buy another 100 link building articles.

Third Person? Neutral Tone? No Company Names? Are You Crazy?

Allow me to take a leap in the dark and assume that most of your link building articles are intended to become guest posts on blogs. Spend some time reading a blog. Blogs address their readers directly. Blogs aim to be entertaining, not 'neutral.' Mentioning other company names is a widespread practice on reputable blogs.

Oh. You're submitting the posts to EzineArticles and low quality directories that have no idea what an engaged audience is. You're doing this in the hopes that Google will reverse it's strategy and that social media is just a phase. You're doing this because you think that non-promotional means never having any fun.

Let me break it to you. It doesn't matter if you're just out to game Google with some manufactured links. Your article's not going to make it onto a high quality site if it isn't written in a way that gets an audience excited. There's never been any question that a link from a high quality site is better than a link from a low quality article directory, and it's clear that Google will only get better at telling the difference.

If you were trying to get your articles published in an academic journal, then it would be a different story. And you'd be going to the wrong place for your articles.

Don't ask your authors to be neutral. Give them something interesting to talk about, and let them respond to the topic like a human being. Maybe then it will be easier to convince the search engines that it was, in fact, written by a human being.

What Have We Learned?

Don't keyword stuff, do the research, and ask for articles that look like they were written by human beings. It's really all pretty basic stuff, and I know anybody with a budget is capable of doing it. But you wouldn't know it from taking a look at the article requests I sift through every day. Good luck, and godspeed.

Scratch that. Luck has nothing to do with it.

Why It's So Easy to Get Marketing All Wrong

I got a couple emails last week I wanted to share in anonymized format. Here's the first one:

It's me again <redacted>, just wondering I have been learning allot more about how to link build without software like senuke x and other automatic software and becoming a better manual link builder with google alerts etc.

And here's the second:

I look after around 6 clients at the moment, but my daily jobs just seem to be very repetitive e.g. finding related blogs, commenting on them, submiting sites to decent directories and guest posting, an now and again creating infographics and sharing them with blog owners and across sites such as reddit/quora etc...mostly I'm just blog commenting though.

I get A TON of emails like this. When folks are relatively new to the field of online marketing, or are moving from classic marketing into SEO, they often reach out seeking advice and help. Unfortunately, the volume's become a bit overwhelming of late, and I'm only able to respond to 50%, sometimes less (side note: I tried an experiment w/ email scalability a couple months back that failed). Thus, I wanted to write a post to express some empathy.

Yes. Marketing is really, really damned hard.

I understand the temptations to phone it in, to spam instead of creating authentic value, to outsource responsibility, to proclaim for all to hear that you HATE marketing, to give up. You're not alone. In fact, I've been just inches from all of those perspectives time and again over the last decade.

But that's also what makes great marketing so powerful. When:

• Very few people are qualified or capable to do something
• Many people believe that thing to be impossible
• Only a handful make exceptional invesments to achieve it

That, in my opinion, is when remarkable things are in your grasp.

The marketing channels we invest in - SEO, social media, content marketing, community building, virality - fit these parameters well. It's easy do the basics, tough to get the intermediate items right and mind-blowingly challenging to get that last few percent that takes us from mediocrity to extraordinary.

So many times, marketing professionals are called in to execute on Step 3 after being handed half-assed 1s and 2s. My friend Philip Vaughn told me at a lunch some months ago that "startups aren't really an engineering, product or organizational problem. They're mostly a marketing problem." But if we're handed crap to market, we can't help but do crap marketing.

At the risk of pissing a lot of people off - A large portion of SEO is just compensating for not being awesome.

' Rob Woods (@robdwoods) April 23, 2012

 

So many of the questions I see around inbound marketing boil down to the same fundamental challenge:

The way I see it, we only have two options:

A) Give in to giving up.
B) Take/earn responsibility for Step 1 and 2

Embracing option B and taking responsibility for your product -> marketing lifecycle is something very few people are qualified or capable of doing, many people believe to be impossible and only a handful ever execute exceptionally well. And it means remarkable results are in your grasp.

How We Managed to Benefit from the Panda Updates

As I am into the online marketing field, I read a lot about SEO. This is my first post about SEO, so please don't be harsh in the comments. The Panda update is what made the SEO community roar about how many websites lost ranking and so on. There is so little information about the ones that benefited of the update and we are one of the winners.

I personally think that the Panda update made the SERPs quality a lot better and to some point buried the medium to low quality websites deep into the results. Even some of the high-authority websites went down.

I will share some insights of an user generated moving reviews website MyMovingReviews.com and how we got positively impacted by the Panda update. The website features many US and Canadian moving companies and provides the opportunity for people to rank them and write moving reviews. Additionally to that, there is a blog/article section with moving tips and info.

Industry specifics that influence the analytics data

Before we begin, you should know that the specifics of the industry add some additional noise to the analytics data. These are the main trends in the moving industry:

• Weekly trends: People search a lot more about moving services in the beginning of the week in the working days. Mondays are usually the most active days. We assume that people usually search for movers at work during work hours.
• Monthly trends: People search for movers more by the end of the month and less in the middle of the month and during holidays.
• Seasonality: People search 30% more for movers in the summer months than during the rest of the year. Nobody wants to move in the winter (especially in the Northern states).

The Fist Panda update

Since the first Panda update in 2011 we started seeing some increase in rankings. Because of the specifics of the users behavior in our industry, the analytics data is looking weird but you can see the pattern.

Further benefits from the Panda update

As we saw a huge opportunity in the Panda update, we tried to adjust the website to better suit the visitors, give them alternatives once they visit the website and make visitors consume more of the moving industry related content. The goals were to increase the time on site, reduce the bounce rate and increase the pages per visit.

What we did to increase rankings/visitors

1. Reducing the bounce rate

We stared by working on the high bounce rate pages. We edited some of the content and deleted some of the pages. One of the very high bounce rate pages were the blog section posts. Since we are always committed to build only high quality content, we knew that the problem with the high bounce rate on the blog was elsewhere. We knew that visitors were able to find the information they were searching for and after that they were leaving the blog. We added a suggestion fly-box. The box appears on the right side on the page once the visitors scrolls by the end of a post and suggests another random post from the blog. This had a huge impact on the blog bounce rate by lowering it with more than 30%. From the highest bounce rate section of the website, the blog become the lowest one overnight.

2. Creating a mobile website

We have about 11 percent mobile visits (we don't consider iPads to be mobile traffic). We decided to further lower the bounce rate by creating a full-featured mobile website. This of course brings the benefits of higher conversion rates. We've been postponing the mobile website for some time now and we finally decided to finish it and launch it by December. We kept the same URLs as the desktop version and only changed the templates.

3. More content

As part of the Panda update is the amount of content on page. We didn't want to have many pages with thin content so we increased the minimum text required for a moving review to be posted. After reading about how Zappos corrected the spelling mistakes of all their reviews, we additionally wanted to avoid spelling mistakes as much as possible. We included a spell checker on the moving review form. We are also planning to correct the mistakes on all old reviews in the future.

To recap, here are the changes we did:

• Editing some of the content with the highest bounce rate.
• Adding a spell checker on the write a review page and setting a higher minimum amount of text for the reviews.
• Giving suggestions to users once they finish reading a blog post to reduce the bounce rate.
• Started a mobile website to reduce the bounce rate and time on site for mobile visitors.

The results

We had almost 50% increase in visits in the next one-two months. Please note that we introduced most of the changes in December, so we can't really measure how fast did these changes influenced the rankings because of the holidays. Not surprisingly, the largest part of the increase was from the blog as this is where we managed to reduce the bounce rate the most.

Conclusion

I can't say that all of the gained increase of visitors came because of the above changes, but given the changes and tactics we did at the time, these were the most significant ones. Targeting the visitor and thinking of how to enhance the customer experience results in more visitors. It is as simple as that. Working on the design and thinking of techniques to reduce the bounce rate will result in better rankings, especially if you are a high-traffic website.

What do you think about the bounce rate and its impact on rankings/visitors? Let us know your opinion in the comments below.

The Real Impact of the Google SmartPhone Crawler (Part 1): Situation Overview

Not everyone realizes it, but Google has been serving different search results to mobile phones than desktop computers for a long time. Beyond that, they serve different results based solely on the handset you are using to search. The differences are often subtle, or focused on the order of Universal Results that are included in the mobile result-set, but Google is algorithmically trying to prioritize content that will work well on the phone that submitted the query, and give less priority to content that might not work on the phone. If you want to compare for yourself, MobileMoxie has a mobile search simulator tool that allows you see the results of one search query across three different phones at a time (cool!)

Now Google has launched a new smartphone crawler, and this will likely push the differences between desktop search and mobile search more into the mind of the average SEO. (Think: 'Does Google consider this a smartphone or a feature phone or something else? I don't know!')

So let's see if we can demystify the impact of this new smartphone crawler using some data. This is the beginning of a three-part blog post series, all focused on the new Google smartphone bot. This first post will focus on how it works and what sites will be effected. The next post will focus on how you can optimize your mobile content for the new smartphone bot best by creating effective mobile redirects. The final post will discuss common mis-indexing problems that mobile websites have and review how to protect your mobile content from search engine indexing problems.

Mongoose Metrics published a study that broke down how the top sites in the US are handling their mobile traffic. They divided the QuantCast Top Million sites (most US traffic' d sites) into 3 categories based on how they were publishing mobile pages: Server-side redirection, JavaScript redirection and what they call 'cloaking' or selective serving of HTML assets. The study went deeper, and looked at results for iPhone, Android and RIM requests, but we can just look at the summary for all smartphones, which show the following stats:

For SEO, these results are each unique and relevant; here is why:

Server-Side Redirection

This is a mobile site-architecture strategy that uses two (or more) urls; one for the desktop or primary content, and one for the mobile content; designations can be added for tablets, WAP and other devices as well. The mobile content can be on a mobile subdomain, a mobile subdirectory or a totally different domain, and those decisions can all impact the content's ability to rank. The mobile urls can be static and optimized urls or they can be temporary dynamic urls, which are usually stuffed with the exact paramaters of the mobile page request.

Until Google's new smartphone crawler, this mobile SEO strategy relied on the rankings of the desktop pages which automatically redirect to mobile content when requested by a mobile phone OR building independent SEO value for the mobile pages so that they would rank on their own merit. This strategy sometimes also includes joining the mobile pages with their desktop counterparts by using the canonical tag to help share SEO value. What is most relevant for SEO is that that both versions of the page are left crawlable by search engines. This might be important if you target lots of WAP phone searches, which are still sometimes using a separate 'mobile-only' Google index and are not affected by the new smartphone crawler. This could also be important if there is a future algorithm shift that puts a stronger emphasis on mobile file size (which could still happen because it is so important for a good mobile user experience).

SUMMARY:

RESEARCH RESULTS FOR SERVER-SIDE REDIRECTION: 52.52% of the QuantCast Top Million Websites in the US may see an immediate user experience benefit from the new smartphone crawler. Once these sites have been crawled by the new smartphone crawler, they will be serving mobile content to mobile users from search results automatically, without their server having to process each of the redirect requests.

 

JavaScript Redirection

This mobile site-architecture strategy also involves a primary and a mobile url for each page and an on-page JavaScript redirect is included from the desktop page to its mobile counterpart. The strategy actually relies on the FAILURE of the search engines to crawl and execute JavaScript in order to work properly. Frequently this strategy relies 100% on the desktop page to rank well in smartphone search results, and the mobile pages are blocked from search engine indexing in the robots.txt file, to prevent the risk of duplication or confusion in the index. In some cases, this JavaScript is detecting specific phones, and redirecting to landing pages that are just built for those specific phones, which can get very involved, but is great for user experience. Unless the new smartphone crawler begins to execute JavaScript redirects (unlikely), sites that rely on this method will not benefit from Google's new smartphone crawler.

SUMMARY:

RESEARCH RESULTS FOR JAVASCRIPT REDIRECTION: 2.15% of the top million websites are not going to benefit at all from the new smartphone crawler, but may have already had good results in mobile search rankings without indexing mobile-specific pages.

 

Cloaking or Dynamic Serving

This mobile site architecture strategy relies exclusively on one url that can display a page with different characteristics, depending on the device that requests the page. What content is served is determined by the server and something that is usually described as a 'mobilization engine' or a 'transcoder.' These are essentially databases of rules and content at various sizes or stages of degradation that can be sent, depending on the capabilities of the phone requesting the page. With this system, a desktop computer will get the full version of the site, but a mobile phone might be served a similar HTML skin, with smaller components switched in, to replace place of the larger elements that are served to the desktop computer, all on the same url.

A similar but less sophisticated version of this type of mobile publishing can be accomplished using mobile-specific style sheets and media queries to re-render or re-organize the content on the page based on the screen size of the device that it is requesting it (Responsive Design). This strategy uses only one url which might be appealing if you are trying to keep things simple for maintenance or SEO reasons. In both cases, bots will be served content based on the device or browser that they are emulating, so the smartphone crawler would likely (hopefully!) be served a smartphone-friendly version of the page unless the JavaScript is purposely made un-crawlable.

SUMMARY:

RESEARCH RESULT FOR DYNAMIC SERVING/CLOAKING: 45.33% of the QuantCast Top Million websites might be at a disadvantage in terms of bandwidth, user-experience and load-time. All of the dynamic processing is still required by their server to render the page each time there is a mobile page request. They will not likely benefit from Google's new smartphone crawler, and might now be at a disadvantage, (in terms of the load time of the dynamically generated pages) where they probably had a slight advantage previously.

 

Mobile SEO is an ever-changing field. The search engines don't all agree, and still don't seem 100% sure how to best rank and evaluate mobile search results. Your best option is always to know how things work, keep a close eye on how your sites are ranking and do your own mobile testing wherever you can. Until the smartphone crawler many people were unaware that mobile search results were treated differently from phone to phone. Now you know that testing on your own phone might not be enough; you may have to address the new Google smartphone crawler with some well-planned mobile strategy. Hopefully you can use this analysis to help determine how your site will manage with mobile serving, and still please the new smartphone crawler.

Stay tuned for the next post in this three-post series about the new Google smartphone crawler. It will cover how you can optimize your mobile content for the new smartphone bot best by creating effective mobile redirects, then the final post in the series will review common mis-indexing problems on mobile and discuss how to update your mobile server settings to prevent search engine indexing problems.

Modern Business and Content Guidance from Bell Labs

I've been reading The Idea Factory, the history of the innovation that came out of Bell Labs around WWII. Innovation, invention and ideas are things that we don't get to talk much about in SEO. It's all about keywords, links, and content, but I want you to take some time today to think about ideas, inventions and innovations. Why? Because we all want to be millionaires (oh hush, yes you do), and while that might not happen to all of us, idea generation is central to our jobs in search marketing.

Bell Labs was born out of a technology being invented (the telephone) and the problems AT&T faced building a communications infrastructure in the United States (and how to connect overseas). We are talking serious problems like wood decay, birds and wires, distance and signal strength. Things we take for granted today. 

The men and women at Bell Labs strove to find the best materials, to make products and services better. We can learn from the team at Bell Labs. Use the ideals of hard work, research and development, invention, and innovation to drive us to make our websites, products, and services better. 

That's how you win on the internet. Be better than everyone else. 

Innovation is a Lengthy Process

The term "innovation" dated back to sixteenth-century England. Originally it described the introduction into society of a novelty or new idea, usually relating to philosophy or religion ... an innovation defined the lengthy and wholesale transformation of an idea into a technological product or process meant for widespread practical use. 

This portion of the book really hit home with me. Innovation is a lengthy process of taking an idea to a product or process for widespread practical use. There are no overnight successes, just people that worked really hard behind the scenes pouring their blood, sweat and tears into an idea.  

The same should be true with your business and website. Innovation in your space is not just putting up a website and knowing the right keywords to target. It's about waking up at 5:30 in the morning to be at a networking breakfast at 8am a few times a month for a few years. Success is about determination, dedication and hard work. 

The time you take to discuss your idea, how you plan to be different with members of your target market, will pay back not only in the refining of that idea, but also in friends and allies when your business or website launch. Choosing the connected people is a strategic process; if they are there at 8am, they are dedicated too and have met others that are of the same breed. Now does that mean that everyone needs to go to 8am networking breakfasts? No. I hate them. But you get the idea.

Want your website to be a success? Then spend the time in the following areas:

• Business Idea - don't copy someone else's idea unless you can be better than them and everyone else in the space doing the same
• Website Copy - Develop the copy for your site, and if you have products what you'll let others use. Write into affiliate contracts that your product descriptions are not to be used elsewhere. Treat website content like a white paper, protect it, don't allow others to copy it. 
• Promotion - I am not talking online promotion, I'm talking the whole marketing umbrella. This is not something you buy/outsource and leave alone. It's your baby, own it! 

Make Goals, Not Hard Timelines

Inventions are a valuable part, but invention is not to be scheduled or coerced. Harold Arnold, Bell Labs

Need $10k by next year? Don't build a website, add products, pay someone in another country to write copy and do SEO and think you can do that. Good ideas don't fit in timelines. You cannot rely on keyword research tools to tell you what product area has low competition and can make you millions. 

What you can do is find an area, a business idea that you really love: something that you can get lost talking about for hours and not mind. Your first goal should be to find that idea. Once you have that (and most of you do as you have clients or work in-house for someone that has that idea already), then you need to set goals for what you want to accomplish. 

The goals should be:

• Lofty but attainable (one viral infographic a month is not an attainable goal, an increase in sales from organic search by 20% is attainable)
• Developed around making the customer happy (the happier they are, the more you sell, simple)
• Tied to your boss's goals (even if your boss is your wife)

Once you know what you want to do, the end result, that drives how to get there. Now you have a purpose, but you can't rush it. You will fail, content will flop sometimes, there will be a keyword that you just have a hard time ranking for. Things will go wrong but things will also go amazingly right if you focus on a good goal.

Do Something

When you don't know what to do, do something!" Jim Fisk, Bell Labs

This is my new favorite motto. Do something. As Will Critchlow says all the time, fail faster. Come up with as many ideas as you can to meet your goal, no matter how far out of the park they are. Execute. Try things, be risky, and fail (sometimes). I can almost guarantee you that even if you fail; you'll learn something you can apply to your next attempt.

Studying Sharks

Jim Fisk proposed a $50,000 study of sharks to help naval warfare. It made it to the top and the lead of Bell Labs finally killed it. How cool a story would that have made? I would have linked to it, what about you?

Dare You

Don't know what to do? Here are some ideas for every SEO out there. I dare you to try one.

• Data Mining. Take a half-day, look into the data you have in your company about clients, about the industry. If you were to clean it all up, could you show trends about your industry? Get creative and find a way to share that information in a resource, a white paper, blog post or infographic. And then do a marketing campaign around it. If it works well, make it an annual update about the industry.
• Content Revisit. Find your top 10 products or services pages. Revisit the content, when was it last written? Update it! Add resources; make it the page to go to for that product or service. Think from the perspective of a potential customer, what would make that page better than all of your competitor's pages? Consider new images, a video, new resources, reviews. 
• Page 2 Revamp. Take 5 pages that rank on page 2.  Revisit the title tag and meta description of those pages. Is the title tag over optimized? Does the meta description really pull you to the page? Rewrite them to be real, accurate and descriptive!
• $5000 Challenge. Ask for a budget that is normal for a project in your company or for a client and do something that is way out in left field. You in finance? Pay off someone's car loan. Don't give away a iPad, donate 100 to a school that teaches computer science to under-privileged kids. 

So would you do with $5000 to help your site? What is your shark bait?

Shark Stock Photo from Shutterstock

#SocialSuccess - An Inbound Marketing Case Study for B2B

There has been a lot of great discussion about the term 'inbound marketing' of late and exactly what is covered by that phrase. For the purposes of this case study we are using the hubspot definition of inbound and outbound marketing. The following is a case study of how we (Salesforce.com) used inbound marketing along with social advertising and great retargeting to grow both our traffic and leads in the UK. Whether you are in B2B or B2C marketing, this case study should be relevant to you and your markets.

The new B2B Purchase Journey

The online landscape for marketers is changing at a rapid pace. People don't buy the way they used to. There is a new purchase journey with three key elements:

1. Search-initiated - Most people begin their research of a new product via search engines, 78% of Internet users conduct product research online (Source Pew Internet & American Life Project, May 2010)
    
2. Social-powered ' The growth of social networks has meant we can now tap into our own external networks for recommendations. Twitter's active user base alone generates 90 million tweets per day, with 24% of adults have posted comments or reviews online about the things they buy. We now have a lot of user-generated content to review before making a purchase decision.
    
3. Buyer controlled - People can now choose where and when to engage with your brand, plus what content they would like to consume. You have to produce marketing strategies they choose to engage with.

For B2B companies this means their sales people are being engaged a lot later on in the purchase cycle and presents marketing with an great opportunity to become an integral part of the overall sales process.

"Get Found"

Considering the above, we decided to run a pilot project in the UK around the concept of 'Get Found' (coined by Brian Halligan of Hubspot). Our aim was to get found by the people who are actively looking for help with the kinds of issues we address. We would do this by harvesting our own expertise in content that helps our prospects do their jobs better.

Since the core mediums involved in this project were search, social and content, we needed to consider how these different tactics are starting to converge and try to hit our sweet spot.

To do this we needed to answer three key questions:

• What do our prospects care about?
• How can we harvest our expertise to help?
• How can we get this content to market now?

Our Answer ' 'Content Rich Microsite'

When discussing microsites, a lot of people probably conjure up images of those used in new product launches (they have a very short life span) or those used to build elaborate link schemes. Our solution was to build content-rich microsite filled with the kind of content our target market would value. One critical aspect of the project was the location of the site. If you look at the salesforce.com structure, you will notice we already have a lot of great blogs sitting on http://blogs.salesforce.com/company/. Since I am interested in EMEA and in particular the UK for this project, I wanted the site to sit within our UK folder, so it would benefit from all the inbound links and social shares generated. To build our micro-site strategy, we had to address six key points:

1. Personas:
Who would this site be for?

For me persona development is the foundation of any good inbound marketing strategy. I am a massive fan of persona development, from the usability and design of your site, to content development; they ensure you strategy stays on target. In fact one of the best link building posts I read last year involved a type of persona development. We ran an intensive persona workshop (with the help of iqcontent.com) that included people from marketing, sales and customer feedback. We came up with 5-6 profiles of users we were trying to reach.

We mapped these against different stages of the purchase cycle and segmented by company size. All of this would help us when it came to content strategy and promotion.

2. Theme:
What would be the overarching theme that would hold all of our content together?

We used our own Radian 6 our social media monitoring tool, analytics and feedback from personas to come up with 'The Social-Powered Business'.

3. Topics:
How do we take that theme and break it down into specific topics we can generate content around?

For us, this was pretty easy; we looked at the areas of business where social media had the greatest impact (sales, customer service, collaboration and marketing). It's also important that your topics and themes are aligned to your products (we are trying to generate leads after all).

4. Process:
Exactly where would this content come from and how would it be validated?

Getting people excited about the project is key. You need to have people who will help with content development, feedback and amends. We used our own collaboration tool Chatter to build an internal social network around the project that consisted of 56 people. All content development was driven through that group.

5. Resources:

Of course we needed to source budget and a team.

6. Metrics:
How would we measure success?

This is a really important part of establishing any successful strategy. Brand awareness is never a good enough metric, traffic; leads and pipeline are what count. We built a dashboard in omniture with all key business metrics to measure our project.

The Launch ' #socialsuccess

In 12 weeks we managed to develop:

• Strategy
• Personas
• Website
• 32 pieces of content

and our #socialsuccess site was launched on January 3rd, 2012.

The following five items were important in terms of making the launch of the site a success.

1. Content Types

For launch we chose four different categories from which we could generate content:

1. Created: Original content that was created from scratch. These are obviously the most resource intensive. They included things like an eBook, infographics, articles and slideshares.
2. Curated: These are round-up style posts. Choosing a topic like social selling and pointing to the best resources from the web on this topic.
3. Collaborative: We choose some of the best thought leaders around our topics and reached out to see if they would contribute some content.
4. Legacy: One of the easiest ways companies can quickly scale their content for inbound marketing is to repurpose content they already have into different assets. For example, our Dreamforce event that runs in San Francisco has a huge amount of expert presentations that are recorded over three days and put onto Youtube. We simply took the best videos and turned them into articles.

2. Product Messaging

Remember this sort of content is not product centric. Best practice for this kind of content is to follow the 80/20 rule ' 80% non product and 20% product, for launch we stuck to 90/10. Product references were used where they made sense, but only on a limited basis.

3. Promoting the site

If you build it, they probably won't come unless you have an awesome promotion plan. Some of the things we did to promote the site were:

1. Facebook/Twitter: Of course, all our best content was shared via our own Facebook, Twitter and Google+ pages
2. We took over the home page of our corporate site (www.salesforce.com/uk) to promote this new microsite
3. Expert advocates: We collaborated with 15 experts for launch, who were kind enough to share our content with their networks.
4. Email/Newsletter: We promoted the site launch to our UK email database and also created a newsletter called #socialsuccess Insider to keep connected with users who signed up via our eBook download.
5. Guest Blogging: We did some guest blogging on relevant sites to promote #socialsuccess
6. PR: We did some PR around some of the pieces we produced
7. Employees: We galvanized our internal employees to share with their external networks

4. Outbound Marketing

We supported all our inbound marketing with great outbound tactics:

1. Twitter: We ran sponsored tweets for our premium content (eBooks). We saw some really great CTR numbers for these. I highly recommend them.

 

2. LinkedIn Banner Ads: We ran some advertising on LinkedIn targeted at our core personas developed above (linkedIn has some great targeting options like target by job title). Again, we saw a far higher CTR from these ads (those offering content) over those just advertising a product.

 

3. Google Display Network: We are currently rolling out the same type of ads (those offering our premium content) on GDN.

5. Experts

Reaching out to thought leaders in your market is a great way to produce some highly valuable content. We were lucky enough to have some great experts involved in the initial content, who shared their expert advice with our audience and were kind enough to share our content with their own.

The Results

The project was launched officially on January 3rd, 2012 and we have seen some great results already. The feedback we have been getting back on our social channels around the content is great.

But we have also seen great results in terms of our business metrics (keep in mind we are in B2B):

• Traffic for January was up 80% YoY
• Traffic from social sites was up 2500 %
• We have over 6500 people signed up to our newsletter
• Our eBook has been downloaded over 10,000 times (generating 10,000 leads)

Our inbound marketing experiment has really shown us how impactful this stuff can be. We are currently working on similar sites in France, Germany and also new topics sites for EMEA.

So it's Onwards and Upwards!!

6 Changes Every SEO Should Make BEFORE the Over-Optimization Penalty Hits - Whiteboard Friday

Howdy SEOmoz fans. Welcome to another addition of Whiteboard Friday. This week we've been hearing a lot of chatter in the SEO blogosphere and on Twitter and on the forums about this new potential Google penalty that's coming down the line around over-optimization. Now, one of Google's representatives mentioned at a conference, South by Southwest, down in Austin, Texas, about a month ago actually, that Google would be looking into penalizing over-optimized websites and folks who have engaged in over-
the-top SEO.

There's been a lot of speculation around when that's coming out, whether that's coming out. There are a few things happening, actually, this week and last night about, "Hey is this already something we're seeing?" Seer Interactive, right, Wil Reynolds' fantastic SEO company out of Philadelphia had this penalty, and people were wondering whether that was related to this. Not really sure.

But before this penalty hits, for goodness sake, SEO folks, let's make these changes to our websites because we could be in real trouble if we don't impact these things beforehand. I think these are some of the most likely candidates to be hit by Google's over-optimization penalty, some of the most likely patterns they're going to try and match against in this upcoming change. So let's talk through them.

Number one, your titles need to be authentic. They need to sound real. They need to sound like a human being wrote them that was not intending necessarily simply to rank for phrase after phrase. I'll give you a good example. Bad: web design services, web design firm space brand name, whatever your brand name is, web design. What does it sound like? It sounds like all you're trying to do is rank for keywords, not show off your brand name, especially if this is your home page or those kinds of things. You're repeating keywords three times. Web design is in this title three times. Think about whether a normal human being would read that title and think, oh yeah, that sounds legitimate. No, they'd think to themselves there's something fishy here, something spammy, something's wrong, something manipulative. Try instead, probably equally effective, if not more, brand name web design Portland Spiffiest Design Services. Now look, I've got the word "design services," which you wanted to get in here. I've got the city where you are that you're trying to target, got brand name web design, right, sort of branding myself as the product and the keyword. Much, much better.

Try and look through your sites and see if this is a potential issue. I've seen tons of sites where SEO folks have just gone overboard again and again. Don't get me wrong. I used to do this too. One of the crappiest things about this is, even if your rank, your click through rates go down. So you can rank in position two or three and be getting less than the people below you, because people don't think that these are legitimate titles and they perceive them to be manipulative, especially if you're targeting more higher end, savvy or sophisticated technology customers.

Number two, manipulative internal links. I see this a lot on side bars, inside of content, where people have taken all of the instances of a particular word or repeated it throughout the side bar or in the footer, those kinds of things, and are pointing with exact match anchors to the same page over and over again. Now, we all know as SEOs that the first anchor text link counts and only one on the page is going to pass that value. Linking repeatedly to the same page with the same anchor is not helpful for SEO, and it makes our sites look really spammy and manipulative and questionable to someone who's browsing it. Why would we want to hurt our conversion rates like this, and why would we want to point out to the engines that, hey, over here, I'm trying to manipulate you? What are you thinking? This is crazy.

Instead, go with logical, useful, change it up when you're linking to pages, maybe a couple of times, in some spaces. You have a blog post and it mentions a page on your site that you want people to actually go to and that you think is useful in context. Great, link over there. Fine, use the anchor text. Maybe use a modified version of the anchor text, a little longer, a little shorter, a little more natural sounding, and you're going to get these same results, but you're going to do it in a much more effective way. You're not going to be at risk of whatever is happening with this over-optimization penalty.

Number three, cruddy, link filled footers. I see this all the time still. You're just having a bunch of exact anchor links down in here that no one would actually really click and that come in lists. I often see them in light gray on light gray so that it's not particularly easy to read. Use your footer wisely. Use your footer to link to the things that people expect to find in the footer. If you really need to get anchor text on pages, find natural ways to put it in the real menu at the top, in the content itself. Don't be trying to mess around and throw footer links site wide, across things. This 2002, man. We're ten years later. It's like at least a decade past that.

Number four, text content blocks built primarily for the engines. You know how sometimes you get to a page and there's good content, usable stuff, an image, a call to action, and then weirdly there's this block of junk. It's this block of blah, blah, keyword, keyword, blah, blah, blah, keyword, keyword, blah, blah, blah. Why is that there? Why does that exist? Does that really work? Does that really trick the engines? Yeah, it tricks them into thinking that they should penalize you. Get that out of there. Rewrite that stuff, man. Seriously, this is going to cost you far more than it's going to help you. If you've got those spammy blocks of text in your pages, that have no purpose other than to get your keywords or some keyword into the text, and it's not actually helping anyone, it's not a good call to action, it's not helping your conversion rate, it will actually drive people away from you. Why are you trying to rank if not to get people to do good things on your site, and like your brand, and appreciate you and come back again and again, and tell their friends, and share it socially, and link to you? Don't be putting this stuff in here. This is dangerous for all of those reasons, and super dangerous given this over-optimization penalty that's potentially coming down the line.

Number five, back links from penalty likely sources. So this is one of the toughest ones because it's really hard to control if you've already gotten links from these places. But you can see with those 700,000 Google webmaster tools, pings that they sent everybody that said, hey, it looks like you've done some manipulative linking, and that kind of thing. Be really careful for all of these, link networks, anything that says private link network, or I have a link network and I'll place your site on it, or building up a network of sites that you then interlink to one and other. Come on. There are so many better ways to get links. You're putting a lot of time and effort and energy into building all of that stuff. You can do so many authentic things with that time. This is time terribly spent. Comment spam, especially those that are sent though automated software blasts, so you think of your XRumer or your SENuke, the article marking robot, or whatever, that's going to submit your site to tons of places or find open holes in the web where they can leave comments and link spam and that kind of stuff. Forum signature links, this is actually one where I suspect it's one of the places where Google really gets to know, hey, this guy clearly is a manipulative, black hat/gray hat SEO, because look, they're pointing to the same site where we found all the link spam from forum signatures, particularly on webmaster sorts of boards. That clearly indicates that's their site and their trying to rain for it, and all that kind of stuff. They've got a long profile, and they keep linking to all these things from their forum signatures. Just be very cautious about this. I'm not saying don't link to it, but maybe don't use your exact match anchor text or try to make it more of a branding play, try and make it more authentic feeling. Certainly participating in communities is a great thing. Just watch that.

Reciprocal lists, right, people are emailing each other back and forth and saying, "Hey, I'll put you on my list of links. You put me on yours. Oh, and we'll do it 20 times and we'll form this big reciprocal circus that's going to get all of us penalized." How great is that?

Article marking sites, I've talked about article marketing in the past. Generally when you see, hey, we're an article marketing site and we can help you rank higher, and submit your content to us and we'll link out, and the same is true for SEO focused directories, anytime you see a site that is essentially extolling the virtues of participating there, or contributing there, as being primarily related to the link and the anchor text and the page rank you're going to get, you can bet your sweet hiney that Google does not want to count that. That's exactly what they're trying to prevent, and I'd worry, whether it's this penalty or a penalty that Google makes in the future, that this is the kind of stuff that gets hit.

Last one, number six, large amounts of pages that are targeting very similar, kind of modified versions of keywords and keyword intents, with only slight variations, slight variation being the key here. So think:
used cars Seattle, used autos Seattle, pre-owned cars Seattle. Why are those three different pages? It sort of feels like keywordy, SEO-y, spam, right, and then there are pointing exact match anchors at all of these. This is the same page. You can target all three of these keywords very nicely on one page that's called Used and Pre-owned Cars/Autos in Seattle. Right, one page, good, you've got it. You've combined all of the things. You want to have that great user experience there. You don't want to have to build that three times. You're not trying to build a bunch of spammy anchor texts to each one that's pointing from each of the different ones. The used cars Seattle page has a link to the used auto Seattle's, it's sort of like, "What?" From a user perspective, "Why is that there? What is the difference between a car and an automobile exactly? I don't understand why these two exist." This kind of thing is something where I think it's a very clear pattern match that the engines can detect. Looks like they did some research and then just built a page for everything, and then they pointed links at all of them. Its manipulative, right. This is the kind of thing, also, that will get you in trouble.

So, one, one, two, three, four, five, six. Six things you should change, and even though I'm not the Count from Sesame Street, you should still pay careful attention to these, because I'm super nervous that when this penalty going to come out, there are just going to be so many webmasters and SEOs who are doing this kind of stuff, and I don't know which one Google's going to hit on this time and what they might hit on in the future. But I just want you to be okay. I want your sites to do well, and this is such bad stuff for user experience too. So please avoid it. Be careful. Good luck to you, and we'll see you again next week for another edition of Whiteboard Friday. Take care.