First off, Panda isn’t named after a bear–it is actually the surname of the Google engineer whose ideas lay behind it. And, although it is about to celebrate its first birthday, it isn’t a single event wrapped in the past. Google Panda has ushered in a series of changes over the past year, with a couple of ranking algorthm updates interpsersed with more regular changes in the data that it depends on.

Panda is revolutionary because it adds a new ranking factor to Google’s algorithm–a quality score imposed on sites by human raters that decide whether the site would be worth visiting again, for example. Dozens of human raters might visit the same site and Google averages their answers. High quality sites get boosted in the rankings, with lower-ranking sites, well, not so much.

Now, this wouldn’t be terrinbly interesting if that is all there were to it. For even Google, with its vast resources, can’t afford to pay human raters to visit all the sites that reside on the Web–not when they need many raters to judge each site and when those sites change regularly and need to be re-rated. No, they needed something a lot cheaper than that approach.

Enter machine learning, a technology that looks for patterns in data. Instead of Google having to use human beings to rate every site, they instead rated a small number of sites and then applied those ratings to all the unrated sites that were similar to the rated sites. So, if your site wasn’t rated. but it has the same characteristics as sites that are lkow in quality, your site will be treated as low in quality.

You probably want to know what patterns Panda is looking for, so that you can avoid them, but no one is saying. In fact, the very way that the algorithm works makes it a difficult question to answer. Machine learning algorithms are trained with some of the human data that Google collected, and then tested on the rest of the data. So the algorithm keeps trying to find more and more patterns until it can actually preduct the answers that the human beings gave. At that point, the algorithm is unleashed on pages that have not been rated, assuming that the training it received against known answers will now allow it to predict the quality level of sites that have not been rated.

What this means is that, for the first time, what human being sthink of Web pages is an explicit ranking factor. So, if you’ve been just following some rote rules about how to optimize for search, you might be in trouble if people don’t actually like your pages. This is, alas, the fate of most search optimizers who are only trying to feed the Google beast what it wants, instead of creating a quality experience for searchers. Thise that give searchers what they want are now being rewarded more than ever.

Google is believed to be going after so-called “content farms” with Panda–low-quality sites produced at low cost by hack writers. But some marketers worry that there are other sites affected. Google reassured marketers that merely having a repeated product description from the manufactuerer is not considered content scraping, but searchers might find it to be a low quality exprerience when they have to look through so many stores and keep reading the same information.

Does this mean that Panda never downgrades a site unfairly? Hardly. All of this technology is imperfect, although Google is constantly tinkering with the training data and algorithms. In fact, Google is collecting lots of data from people pressing +1 buttons, and might find someday that those are all the human raters that they need–and they won’t have to pay anyone.

So, many more changes are still ahead. And if Google’s Panada update is successful, you’ll see Bing go in that direction, too, affecting 30% more of the U.S. searches. And who knows how Panda might evolve in the future. To check out all my slides from the event, take a peek at “Google Panda Update” on Slideshare.

Comments

Haven’t we Seen this For a While Now?

This may sound similar to what you have seen over the past while where certain results had a profile shot of a friend next to them because your friend had recently given it a social nod (liked, +1′d, etc.) but this is far more intense. The new additions are opt-out only and include any of the following content showing in results if it is relevant: images, video, and websites. In addition, these results can also include semi-private and private content such as content shared with you via a limited circle in Google Plus or directly from someone in your network; note that does not mean this is visible to anyone else on Google so don’t get freaked about privacy unless you plan on leaving your browser logged in to your Google profile.

Here is the official (and rather cool) video Google used to demonstrate the functionality all Google users will soon see:

The biggest and most alarming takeaway here is that the only social content found within these socialized search results is that of Google Plus! That’s right, Google did not include one iota of content from Twitter, Facebook, Flickr or any other social platform. Their excuse is painfully simple and transparent; the others won’t let them spider their content to that degree (the simple part) and Google is whacking them over the head for it (the transparent part). Welcome to a classic case of cause and effect except in this case Google has tread heavily on dangerous ground.

So, based on this news did you pick up how Google just dropped its pants for its competitors? The outright antitrust implications of Google integrating its own social network into supposedly neutral search results without catering to the competition – the pants loosen. The full bare arse is revealed when Google made this new functionality opt-out by default; in other words you are in whether you like it or not until you manually remove yourself. This, in my opinion is the dumbest move imaginable and has the social industry quickly aiming at Google’s privates.

Here are a few examples of the rush of heated headlines spurned by this news:

• Twitter Really, Really Hates Google’s New Google+ Integration by Jason Kincaid at Tech Crunch
• Twitter: Google+ Integration In Google Search Is “Bad” For Everyone by Matt McGee, Marketing Land
• “Search, Plus Your World As Long as it is Our World” by John Battelle
• “Twitter Dumps on Google for Pushing Google+ in Search” by Liz Gannes of AllThingsDigital
• Many more are found here (this link will show the latest) courtesy of TechMeme

What Google Could Have Done to Minimize Anti-Trust Exposure

I don’t understand when a company willingly puts itself in the path of legal harm which Google clearly has done by not including its competitors in Search Plus; perhaps it is just another clear example of corporations having too much money to care. At any rate, I would have done one or both of the following options to minimize exposure:

1. Included the content that Google does have access to from the other social platforms. For example, they can index public information from any of the platforms; there is a lot to be had and we see it often in everyday searches on Google. By adding this information they would have the ability to say, “well we would love to include more but we need to create better relationships with each company to get more of that data.” At least that position would have held more water than simply not trying which opens a downright palatial foothold for an antitrust lawsuit.
2. Not everyone has a Google Profile but the advantages of having one is made abundantly clear if you want to take advantage of Google’s vast array of free applications. Since Google’s entire future strategy relies critically upon pushing Google Profiles it just makes sense to make Search Plus a very bold opt-in option when signing up.  And for those who already have a Profile, just use ads to push the benefits of Search Plus and make it dirt simple to enable the option. Chances are that many will take the option and by going this route Google is free and clear of antitrust; at least I think so based on my admittedly puny knowledge of law.

Living With Google Search Plus

Comments

In a nutshell, the post explores how different factors can impact the click-through rate of any listing in Google’s search results.

Given this data is quite difficult to obtain, a dummy search page was built to replicate Google’s search results. Users were then encouraged to visit the page and select the search result they think best matched the query.

Test 1: Great White Sharks

The first test was set to determine if users would click on the first result out of habit, or if they searched for the best match for the query. To do this the search term ‘great white sharks‘ was used with top 3 search results for ‘sharks‘ inserted at the top of the page. This means that the best result for ‘great white sharks‘ was now sitting in position 4 as indicated below.

Here are the results with the percentage of clicks that each listing received,

As you can see, the results indicate that a combined total of 65% of people clicked on the first three results for ‘great white sharks‘ even though they were in positions 4-6 on the page. It’s also interesting to note that the two listings with the highest click through have the text “Great White Shark” in the title of the listing – a well-known ranking factor.

Test 2: The Wikipedia Effect

The second test was created to determine if Wikipedia listings still claim a higher click-through than other search results. In order to do this, the test was run with the search results for ‘barack obama‘ modified so that the number 1 listing (a Wikipedia entry) was moved down to position 2 and then position 3.

The results for Wikipedia in position 2 are on the left, and the results with Wikipedia in position 3 are on the right.

In the first set of results, even though the first listing is extremely relevant to the query, the Wikipedia listing still comes in with the highest click-through rate in the second position.

In the second set of results, the Wikipedia entry again receives an unusually higher percentage of clicks in position three, despite the second listing being a relevant whitehouse.gov link.

So what learnings can we gather from these tests? Well for the benefit of those who have skipped to the end of this post, here’s a summary of some of the key takeaways:

1. While listings in position 1 do get a fair share of clicks, the majority of searches will still spend time to look for the most relevant snippet, even if it is lower down on the page.
2. Having an exact match of the search phrase in the Title of the search result will result in stronger click through.
3. From the test results above, Wikipedia listings by their nature will attract a higher click-through rate.

If you’re interested in learning more about the test I encourage you to have a read of the original post over at SEOmoz.

Comments

Christine started by sharing why she thinks tools are so important:

• save money and time
• provide insight outside of your own site
• identify missed opportunities
• does the “grunge work” for you
• they make you more efficient and keep you sane!

1. Google is taking into account the personal blocking data (the block site option in results) from users that have a long and trusted profile; confirmed by Matt Cutts in his September 21st Q&A (the first answer on the linked page).
2. Right now, if you have a highly trusted profile and you have authorship markup (rel=author) on your articles/copy you will get representation in Google search results – by having your photo show up next to the article.
3. In order for this markup to work you need to have a Google Profile and it must be correctly associated with the sites you write on and your author page on the site has to connect back (a few hoops are necessary) to your Google Profile to finalize the association.
4. Google is integrating Plus into most (if not all) of their products – this was confirmed by Vic Gundrota on a recent Web Summit 2.0 interview with him and Sergey Brin.
5. In order to be on Google Plus you have to have a Google Profile.
6. Your Google Profile prompts you to connect all of your social profiles so Google knows your social fingerprint and can highlight content in search results that your friends have socially shared/liked.
7. Links are an important part of Google’s algorithms but they are heavily gamed and likely cause the majority of spam found in Google’s results.
8. If Google sees that others like your content then it has a better chance of appearing at the top of relevant searches.

Next, let’s connect a few dots and make some educated assumptions:

1. If your content is shared extensively on social networks, especially Plus, and you have rel=”author” (AKA authorship markup), Google will credit your profile with more trust.
2. If someone links to your article or otherwise shares it, Google can see the authorship markup and will consider crediting your Google Profile – depending on the quality of person linking/sharing the article.
3. Links are a trust indicator and Google Profiles will be a trust indicator with a tougher signal to fake.
4. By creating content regularly that is highly shared and signed with your rel=”author” you will build greater trust for your Google Profile.
5. The more trust & credibility you have with Google, the better chance your content will have of appearing in the top search results.

My Conclusion
Build your Google Profile because there are many reasons to believe it will assist the ranking success of content you write and socially share. In order to build your profile you will want to give Google every reason to believe you are trustworthy and rel=”author” is one of the tools they have given you to do that. The sooner you get started, the longer your positive  Google Profile history will be and the more trust you can gain before your competitors wisen up.

Click on the above image for a much larger version.

Important Reminder: although many of these signals are already being noted by Google few of them have been integrated as of yet – at least officially. This article is based on educated speculation.

In order to give your brand the best chance at succeeding in the online world, here are 3 ways you can derive SEO value from your time spent on social networking sites:

Link between profiles
Think of social media marketing like the wheel of a bicycle. Each social profile is one of the spokes and your website is the central hub linking them all together.

You never want the visitor’s journey to end at any given social profile. By interlinking your social profiles with each other, as well as with your website, you are encouraging visitors to extend their interaction  with your company and your brand. The longer you can keep them engaged the better chance you have of getting them to convert.

Linking between profiles also gives you the chance to connect with your target audience on more than one platform, increasing the amount of touch points your brand has in their online lives. For instance, if someone connects with you on LinkedIn, why not send them a message inviting them to follow you on Twitter and to Like your Facebook page? You don’t know which one of these social profiles plays the most important role in their online social lives, so by  creating a loop between all of your social profiles you are helping ensure your message gets heard at least once.

Keep in mind that the end goal of social media marketing should be getting your social connections over to your site, not driving traffic from your site towards your social profiles. Don’t dedicate prominent real estate on your website to giant “Connect with us on Facebook!” buttons. Your site should focus on converting your visitors, not turning them into fans/friends/followers. Keep the “connect with us” buttons on your site, but don’t let them overshadow the other goals of your site.

Promote your content

Content marketing forms the backbone of your SEO and drives most of your online marketing tactics in general. But creating great content is only half of the battle. It doesn’t matter how unique, informative, inspiring or useful your content is if no one sees it. That’s where social networks become incredibly valuable from a marketing perspective. Social media marketing thrives on fresh content and gives your social connections a reason to interact with your social profiles. It keeps your brand top-of-mind and present in their online social lives.

Every time you (or one of your connections/readers) share a piece of your content on a social network that creates a valuable inbound link for your site. Not just ways to drive traffic, these social signals are being used by the search engines to determine the importance of your content. The more times a piece of content is shared across various social networking sites the more valuable it becomes and the better it will rank in the long run.

You don’t have to publish the whole blog post to your Facebook wall either. A small snippet and image is enough to attract the attention of your network. It’s a teaser to get them interested and give them a reason to head over to your actual blog/site to read your content.

Customize and optimize profiles

Social profiles can rank in the search engines like any other webpage. Make sure you take full advantage of this opportunity and properly optimize your profiles like you would your site. For instance, Facebook allows users to create custom URLs for their pages; this is a great place to target your most relevant keywords. You should also focus on targeting relevant keywords in your biography or info sessions.

Not every profile will allow you to post the same amount of information, so it’s important to ensure consistency across your profiles. Before you start getting really heavily involved in your social media marketing, write a few company biographies of varying length that all focus on the same core message. You want to present a unified brand across all of your social profiles so you don’t accidentally confuse your audience.

I’ve picked some UK high street retail examples to help display the issues raised, but please note that we have no connection with any of these websites – so this is an outside perspective. There may be logical reasons for the examples which we are unaware of, but these have been used in order to highlight where SEO mistakes are commonly made.

Non-descriptive URL structure

Ideally you want to keep your URLs concise and keyword descriptive. So automatically generated, ID-based URLs aren’t going to help your SEO, unless you’re aiming to rank for g474502s2 - in which case Next.co.uk have dominated market share!

Long and messy URLs generated by CMS

Some content management systems really make a mess of URLs. From an SEO perspective you want to have full control over re-writing category-level URLs such as this one on Argos:

Linking to multiple homepage URLs

This is a common mistake – which is getting better across many sites, but if you click the logo or homepage link on some sites, you’ll find that rather than getting sent back to the root domain, you’re taken to a duplicate copy of the page on a new URL. See this example on House of Fraser:

Poor title tags/meta descriptions

I’ve worked with a CMS before that didn’t allow you to edit title tags at all – that was a bit of a problem! Hopefully your site won’t be quite that bad, but too often people just think about SEO for generating rankings – what about click through rates and conversions though?

Crafting an enticing title tag and meta description should be as important as writing a high CTR, converting AdWords ad – notice the difference between these two listings for Marks and Spencer – surely M&S would prefer you click on the natural free listing given the choice!

No user-generated content/reviews

For conversion rates alone, having reviews and user-generated content is an excellent way to boost your site’s performance. See this case study on how onlineshoes.com increased sales by 119% due to user reviews. But it’s also a great way of adding extra content to your products – giving the search engines that extra 200-300 words of unique and what should be well-optimised copy (because it’s about the product) could well be enough to make a significant increase in search rankings.

It could certainly be worth testing at the very least for a lot of brands, for example Ted Baker:

Forgetting about branded product search

One of the first things I check with our e-commerce and retail clients is branded search results. It’s often just taken for granted that you will be ranking for your branded keywords, so it’s assumed that non-branded search and first-time visitors is the main target. However, this isn’t always the case and it definitely shouldn’t just be assumed – these are almost certainly going to be your top converting keywords, so a small amount of effort here can easily pay off to ensure that you’re generating the majority of traffic – which let’s face it you deserve, it’s your product after all!

It’s amazing how many brands don’t rank for their own products though – check out these results for Sony W510 12MP which are dominated by Argos and Amazon:

Lack of static on-page content

Many websites struggle when it comes to having good, optimised content deeper in the site. For example, product pages which have very little descriptive text written about them could be much better optimised for search. See this example from Monsoon, which showcases the product reasonably well, but does little towards telling users and the search engines about it:

Pulling search results in as category pages

As above, sometimes category pages are very weak on content and often these are just search results which are being pulled into a page. Yes it may do a job for the user – but surely a bit more text here would help to give the search engines a bit more to go on. It doesn’t even have to be too detailed – a quick description underneath “Mens Hats, Gloves & Scarves” on the Debenhams site here would be a big improvement to optimise for the phrase “Mens Hats”, which they currently bid on using PPC, yet fail to rank in the top 50 positions in Google organically for:

Webpages & content too image-based

From the websites I’ve reviewed today, I’ve actually been quite impressed that most of these have now moved away from having content which is too image or flash-based. This is a clear SEO issue to avoid, as you want your site’s content to be as well optimised as possible – which means it should be text-rich. Topman is an example of a site which hasn’t quite got there yet – the only text currently on their homepage is navigational:

Duplicate content – same product, multiple categories

I’ve seen several retail sites in the past where they have caused duplicate content issues by having category-level subfolders within the product URL. Here’s one example from Blacks, where they have a product which is listed under two different categories, so they’ve ended up with two URLs for what is exactly the same product:

Because they sit under both categories, the URLs are duplicated – so ideally it’s normally best to avoid using category-level subfolders in product pages – see Amazon for an example of this. Also, canonical tags are there to help get around this issue if it exists – but ideally you’ll want to have each product page in a single location. Hope that makes sense, but Dan’s written a much more detailed post on product URLs causing duplicate content issues – so you should read that one if it doesn’t!

So those are the top SEO mistakes we’ve found retail websites are still making – a big thanks to Malcolm Slade, Rishi Lakhani, Paul Rogers, Stuart Turner, Ashley Hayward, Daniel Bianchini, Ian Galpin and Edwin Hayward who contributed via Twitter. And if you have any questions or comments on what you’ve found to be the biggest challenges, it would be great to hear about this in the comments.

Comments