Click here to register: http://goo.gl/gDZai

We will cover the basics of infographic marketing and review a case study of one of our campaigns. This webinar will cover the following questions:

No related posts.

The security of your Hosting Service’s servers is another thing to consider. Linux Apache servers are some of the most secure and widely used servers on the market. A January 2011 Netcraft web server survey showed a growth of 10.1M hostnames added to Apache servers and the graphs clearly show that Apache servers are the preferred choice by far over all other types of servers. In addition, Linux runs 91.8% of the world’s super computers. (11/2010). Security can dramatically impact not only your SEO but your customers as well. Many of the design structures of PHP and Linux Hosting that seem difficult or cumbersome are in fact security related structures that make PHP hosting very stable and secure. The security of your Hosting Service is important to SEO because of the way websites are managed by hosting companies. If many of the other websites on your hosting service are poorly designed or have security issues your SEO can be negatively impacted. This is guilt by association as far as the algorithms are concerned. In a worst case scenario the entire host can be forced to shut down to deal with a security breech and if that happens your whole website could be down for an indeterminate period of time and this will definitely negatively impact your SEO and your business.

PHP is the workhorse programmer’s language for the web much more so than either Adobe’s Cold Fusion or Microsoft’s ASP. PHP is the most dominant server-side scripting language. Web-content management systems that are written in PHP include Joomla, Drupal and WordPress. Two of the largest websites that use PHP for user-side content are Wikipedia and Facebook. Google, Yahoo!, YouTube, Digg and Flickr all use MySQL and elements of PHP throughout their websites. One of the reasons these large players use PHP is because upload speed can influence SEO ranking. PHP is fast and it is platform agnostic meaning that it runs at the same speed on a smart phone as it does on a high-speed desktop. Scalability is also a major factor to consider when choosing your language and Hosting Service.

In many countries specific markers are used in host names. For Example, .uk for the United Kingdom and .it for Italy. To achieve high SEO rankings in these markets it is useful to use these county specific markers. For example Priceline, a company that has grown very quickly over the last five years, uses www.priceline.com for its U.S. customers but uses www.priceline.co.uk for the United Kingdom. If you are planning future growth for your company outside of the United States it pays to invest in PHP and Linux based hosting now- especially if you have any intention of entering the BRICK or European markets where the LAMP (Linux, Apache, MySQL, PHP/Perl) stack is common.

Using SEO friendly web hosting immediately opens up a wide variety of free tools and sources of code to you that other types of hosting may not. Moreover, because these are Open Source (free) you can apply what you need from the basic caches of code and then customize them to suit your needs. Enormous worldwide communities of computer experts are then at your fingertips to answer questions and offer technical advice. All of this support is free and overall the quality of help far surpasses any paid company support. You can then take the money that you are saving paying for proprietary software and apply it to your company. PHP also works seamlessly with many other tools and languages including Flash. Encapsulation allows you to easily outsource sections of code to specialists to get just what you are looking for in your web site. All of this means that your web site will be up and running smoother, faster and better and this will increase your SEO ranking.

PHP hosting is also often cheaper and sometimes free compared to other types of hosting. Free hosting services need to be carefully reviewed before you choose this route. Often your space will be limited on a free hosting site and you and your customers may be forced to tolerate pop up ads and delays. Depending on your needs, free Linux hosting may be a good choice but it’s not likely to do anything good for your SEO. You would be far better off taking the money you are saving in buying and updating software and spend in on a first rate PHP hosting Service. For more information about Search Engine Optimization please visit http://www.seomoves.org

Related posts:

1. When Do You Need Manged Web Hosting?
2. How To Upload Your Website with a Free Domain and Hosting Service
3. Making the Right SEO Hosting Decisions

Leadership isn’t just for those with the title or position of leadership. Every person, in any given situation, is a leader of something. You may not have the prestige or salary, but there is no doubt that you carry some leadership responsibilities, however small they may be.

Very few leaders start at the top. They start somewhere way down the ladder and work their way up. The difference between them and the next person is that they exhibit strong leadership characteristics. Now, not everyone is a natural born leader, but we all lead in various ways. Whether you’re trying to get to the top, or simply being successful where you are, there are several aspects of your own personal development that must be achieved in order be effective as a leader at any level.

John Maxwell provides 10 steps to developing courageous leadership:

“Convictions that are stronger than my fears”

A leader is one who overcomes their fears. This may be fears of stepping out, fears of trying something new, or even a fear of standing up to what you know is right. Most everyone has convictions but many are too timid to stand up when those convictions are challenged. To succeed as a leader your convictions must overrule your fears.

“Vision that is clearer than my doubts”

For any leader, vision is essential. A leader must be able to see where they are now, and look ahead to where they strive to be. While any vision comes with doubt, the doubt cannot be paralyzing to achieving the vision.

“Spiritual sensitivity that is louder than popular opinion”

Many people try to check their spirituality at the door when it comes to work and leadership, when in actuality they are inseparable. Spirituality is the core of who you are. Unfortunately, many allow trends, popular opinion, or even a louder voice in the room hold sway over what they truly know and believe in their heart. Spiritual strength is essential to establishing a firm moral foundation that cannot be blown over or toppled by the voices around them.

“Self-esteem that is deeper than self-protection”

Protecting oneself from outside forces and influences is a natural reaction. But sometimes people allow that to come at the expense of their own self-esteem. They protect themselves by going along and not standing out. This is contrary to true leadership. Leaders must be able to stand out and, by doing so, put themselves in a vulnerable position. Having the self-esteem to stick to your core convictions may leave you vulnerable, but no true leader ever succeeded under a roof of self-protection.

“Appreciation for discipline that is greater than my desire for leisure”

Greatness (or even desired goodness) can rarely be achieved without a measure of self-discipline. We all want and need leisure time, but those who stand head and shoulders above others almost universally have something in common. They are willing to sacrifice some of their precious leisure time for those things that help them grow mentally, physically, spiritually, and emotionally.

“Dissatisfaction that is more forceful than the status quo”

While I’m a firm believer in learning to be content where we are, there must also be a measure of dissatisfaction with things always being the same. Contentment helps us learn to survive and be happy with what we have. Dissatisfaction helps propel us forward to better things that we know can be achieved. While we cannot live in a state of unhappy dissatisfaction, we can use that dissatisfaction to grow our measure of success.

“Poise that is more unshakable than panic”

Nobody likes a panicky leader. While any leader may become worried or distressed, how they handle those situations says a lot about them. Keeping cool under pressure produces a calmness that spreads within an organization, allowing everybody to think with a clear head and develop strategies that will bring you through any crisis. Keep in mind, however that poise without action is just as devastating as panic… it just takes longer to feel the results.

“Risk-taking that is stronger than safety-keeping”

Leadership itself is a risk. There is no safety in standing up or stepping out when everybody else is just sitting around. Sometimes the risk is mental or emotional. Other times the risk will be financial. But there are very few profitable investments that don’t require some measure of risk. True leaders understand that risk is a part of the job.

“Actions that are more robust than rationalization”

It’s possible to rationalize your way out of anything. The problem is rationalizations reduce us to inaction rather than action. Nothing ever gets accomplished when we can find all the reasons not to do it rather than looking at why it needs to be done. Focus on the goals and find ways to to get there, instead of reasons not to try.

“A desire to see potential reached more than see people pleased”

Every person has potential for greatness. The biggest obstacle to such greatness is often those who we surround ourselves with. While we cannot put aside the needs of our friends and family for our own selfish ambitions, we cannot please everybody all of the time. Nor can you allow others to hold you back from achieving the success you deserve.

Leading is rarely ever easy. Some have natural ability, for others it must be developed. But every leader faces the same trials and struggles. Learning to overcome the roadblocks and other obstacles that often try to set us back is essential if we are going to reach our fullest potential.

Courageous leadership means finding ways to succeed regardless of our circumstances. It means putting ourselves out there, facing our fears, doubts and potential ridicule all for the greater good. While many people may not like what is required to become a leader, very few regret what they have to go through to achieve it.

Search Engine Guide : Small Business Search Marketing

Related posts:

1. 4 Easy Steps to SEO-Friendlier Blog Categories (WordPress)
2. Whip Your Landing Page into Shape in Just 12 Easy Steps

As many of you who read this blog know, I’m a terrible self-promoter. I actually feel guilty writing about, linking to and promoting the products and services that make payroll for the amazing SEOmoz staff and allow us to conduct cool research, produce awesome guides and build out spiffy office space. But, every few months, I manage to crawl out from under that shell. This time, it’s by request.

I’ve been hearing from a lot of our PRO members that they feel both overwhelmed and confused by all the offerings in PRO. I know it’s tough when there are 30+ pages on which unique types of PRO content exist and even the dashboard doesn’t link to all of them (that’s our fault for bad organization – I promise it’s getting better by the end of summer). Hence, this post is all about what to do in your first 15 minutes inside PRO to get lots of value that can actually move the needle on your SEO actions and search traffic.

Step 1: Find Your Big Missed Opportunities via Top Pages

When you run a report in Open Site Explorer, click to the “top pages” tab and browse through the list of the most-linked-to pages on your domain. You’re looking for two things – any troubling codes (302, 40x, 50x) and pages that have lots of links, but aren’t targeting competitive keywords for relevant search traffic. In the former instance, you want to get those pages up and pointing to the right place. In the latter case, you need to run that page through OSE, determine who’s linking to it and with what anchor text (there’s a tab for that, too), then see if you can put together good content to match the links & ranking ability. You can do all that, later – for now, just export the list to CSV, or make a note to revisit.

Elapsed time: 3 minutes

Step 2: Crawl 3,000 Pages on Your Site and ID Potential Errors

The new Custom Crawl Prototype will mimic a search engine spider and crawl up to 3,000 pages on any domain, then email you with a CSV of the results in 24 hours. It identifies duplicate content issues, HTTP headers, missing titles & meta descriptions, and many more potential SEO pitfalls. Get a report on a site or two and dig into the results tomorrow.

Elapsed time: 3 minutes 30 seconds

Step 3: Run Keyword Difficulty Reports for Your Top 5 Keyword Targets

How tough, relatively speaking, are the keywords you’re chasing and where might easy opportunities exist? Keyword Difficulty can help answer this question and provides a terrific CSV export of the top 25 sites/pages ranking for any query with metrics for each. Often just a report or two can help you identify keyword targets where small quantities of links or optimization effort can take you a long way. They’re also ideal for showing management/clients exactly how far you have to go to catch up with the competition.

Elapsed time: 7 minutes

Step 4: Uncover Some Easy Link Targets with Link Intersect

Tom Critchlow and I call the Link Intersect Tool “cheating,” because it’s just too easy to find good link opportunities. Plug in your site and at least 2 (up to 5) competing sites (or just sites that you think have relevant/acquirable links) and it spits back a list of sites, pages and metrics that link to 2+ of the competitors but don’t link to you. It’s like shooting links in a barrel! (that’s a thing, right?)

Elapsed time: 11 minutes

Step 5: Sign Up for a Webinar (or Download a Past Presentation)

I’ve personally run a dozen 60-90 minute webinars for our PRO members on topics ranging from “reverse engineering the SERPs” to “competitive link building” to “actionable analytics” and more. The feedback we get on these is overwhelming positive and we’re running two each month (one with a specific content focus and another reviewing members’ sites). The webinar archives contain video+audio downloads of the presentations plus a link to register for upcoming ones. If you like a more interactive/participatory learning environment, these are a great option.

Elapsed time: 12 minutes

Step 6: Track Rankings on a Few Dozen Key Terms/Phrases

My recommendation is to Track Rankings for 10-20 key terms you’re targeting, a handful of mid-range “nice-to-haves” and a healthy helping of long-tail keywords to help give a sense of how you’re performing across the keyword demand curve. When traffic fluctuates, it’s great to be able to see if rankings were the cause, or if other factors (demand, downtime, errors, analytics capture problems, etc.) could be the culprit. The best part about the current rank tracking system is the ability to choose between multiple engines on any TLD (and to select “entire subdomain” so it catches any page from your site in the top 50 results).

Elapsed time: 15 minutes

And, as a tease, here’s an early comp of what we’ve been busy with in 2010:

ETA: Late this summer

SEOmoz Daily SEO Blog

Related posts:

1. Overcome the Google Analytics Learning Curve in 20 Minutes
2. No-Fail Ways to get Links from Bloggers
3. A Few Ways to Get Fresh Links to Old Content

One of the great things about developing content for your website is that, with a little research, you can know exactly who your target audience is and how create content to meet their needs. Spending a few minutes before setting pen-to-paper, or fingers-to-keys, can tell you just about everything you need to know about what types of things people are searching for on the web. From that, you can determine what kind of content you need to reach your audience.

Using keyword research tools provided by the search engines and third party keyword platforms can help you a great deal in writing for your target consumers. Not only can you learn what keywords people are using, but keyword research can also help you craft your content using the words and phrases that your audience searches for most frequently. This helps you attract the widest audience possible while also focusing your words using higher traffic and better converting terminology.

There are three things that good keyword research will help you uncover: who your audience is, what they are interested in, and what their needs are.

Target Audience

Who is your target audience? Your research will tell you quite a bit about who they are by the searches they perform. Look at the keywords. Are they looking for business solutions? Information that will help them with a hobby? Or maybe something that will help them with their personal or professional education? Even looking for the same product or service, different searchers will use a variety of search words and qualifiers based on what interests and needs they have.

You can use the research to weed out a lot of people simply because you know you don’t provide what they seek based on the words they use in their search. They may be looking for a niche you don’t provide or a variant that you are unable to supply. Either way, by focusing on those terms you can help, while moving away from those you can’t. You’ll find yourself reaching out to a greater percentage of your target market.

Areas of interest

Next, you need to use your research to learn what it is that your customers are interested in. Depending on who they are, each visitor is often searching because they have a specific interest that needs to be satisfied. Some may be looking for information, others education, and still others might only be looking for ideas.

Using this research you can uncover the interests of your audience and use that information to build content that speaks to those interests. With this knowledge, you may be able to create a page, or even multiple pages of content. By looking at specific interests, you are able to engage with your audience on their terms, within the confines of their area of interest. This will help you produce better content that has a stronger chance of converting.

Needs to be met

People are needy! Most searchers are doing so because they need to get answers, solutions, or information. Figuring out what your target audience needs is critical to ensuring you are able to create content that provides them with the answers.

When writing your content to meet visitor needs, you may have to cover a lot of ground. Each searcher wants to know, “what’s in it for me”, and it’s your job to tell them! It all boils down to letting them know what benefits they’ll get from what it is you have to offer. But the benefits won’t be the same for every person. Or rather, the desired benefit won’t be the same, so be sure to hit as many benefits possible.

As you work through your keyword research, you’ll find that there is a lot of crossover between these three categories. Some industries clearly cater to one type or another.

Some business people are looking for ideas, some for information, and still others may be looking to build up their education. Similarly, the same can be said of students and hobbyists as well. You don’t have to be a student to look for education, or a hobbyist to want some new ideas. You need to determine the degree of crossover and whether there is enough to go after those in a category different from your primary audience.

Using your research to uncover all the keyword gems will help you determine the course of your content and maybe even who it is that you want to attract to your site. Some sites can be a catch-all, but many times you’ll find that trying to appeal to everyone appeals to no one. Only you can make this determination.

Keyword research will help you determine how best to reach your target audience. Without it, you’re just struggling around in the dark.

This post was inspired from The Princess Bride themed presentation I gave in early 2010 at SEMpdx’s Searchfest titled Inconceivable Content: The Dread Pirate Robert’s Guide to Creating Swashbuckling Content, Pillaging the Search Engines, and Commandeering a Treasure Trove of Conversions. If you enjoyed this post you also might enjoy other posts inspired from the same. Search for “inconceivable content” on this blog to find them all.

Search Engine Guide : Small Business Search Marketing

Related posts:

1. Keyword Research – Basics For SEO
2. How To Do Keyword Research With Google Adwords Keyword Tool
3. How To Handle Keyword Research

“Improvements to the bot enable more efficient crawling, and increase the ability to crawl content on sites not optimized for search,” he says.

Rick DeJarnette has more about the change on the Bing Webmaster Blog:

Instead of the old msnbot 2.0b showing up in your server logs, the updated user agent will be:

Mozilla/5.0 (compatible; bingbot/2.0 +http://www.bing.com/bingbot.htm)

The HTTP header From field will also change as shown below:

From: msnbot(at)microsoft.com

will become

From: bingbot(at)microsoft.com

If Bing finds separate sets of directives for Bingbot and for other crawlers, directives for bingbot will take precedence, the company says.

I find the part about increasing the ability to crawl content on sites not optimized for search to be particularly interesting. I wouldn’t exactly call this an invitation to ignore SEO. Obviously Google is still the biggest search engine anyway, but even as far as Bing is concerned, good SEO practices will likely still help your rankings.

Also keep in mind that optimizing for Bing is becoming increasingly important. Not only is Facebook giving more reason for people to search (where Bing provides the web results), but the Yahoo/Bing integration will be here (likely) before the holidays.

WebProNews – SEO

Related posts:

1. Microsoft Brings Facebook Twitter Sharing to Bing Shopping Search
2. Facebook & YouTube Are the Most Visited Sites at Work
3. Choosing the Right Keyphrases – Especially for the Smaller Sites!

Last week, I asked the musical question, “Is Your SEO Strategy to Barely Avoid Spamming?” One of the commenters told me that he waited and waited in that article to find out what the line was between spamming and ethical behavior, but I didn’t tell him. So, that’s what we tackle this week.

Spam has an easy definition: it’s anything the search engines don’t allow. You can read the search engines’ terms of service (for example, Google calls them quality guidelines)

All the usual suspects are there, ranging from tricking the search engine by showing the search engines a different page from what the searchers see, all the way to creating fake links or buying links to fool the engines into thinking your page is better than it is.

You’ll hear all sorts of different names for these tricks: hidden text, link farms, paid links, cloaking—there are even more sophisticated techniques. Negative SEO is a way for your competitor to set up a link farm pointing to your site so that Google punishes you, thinking you did it. I know people setting up fake social media profiles so that they can fool Google into thinking many people are talking about their content.

You can check out all these ideas and more, to try to figure out where the lines are. As we discussed last week, you can decide that your strategy is to understand just what is allowed and what isn’t, so you can go right up to the line. (Or maybe just a tad over it.)

Image by Grumbler %-| via Flickr

I believe that is a mistake. Even if you think you know exactly where the line is, it is in the end a judgment call by each search engine. And it is dumb luck whether you are caught or not. And honestly, no one but the search engines really know where the lines are.

I think we can all do better. Next time, ask yourself if what you are doing is good for all three parties in the search transaction. Is this tactic good for you (the search marketer), good for searchers, and good for the search engine, too? If it is, keep it up. Whatever you are doing is not only not spam, but it is something good for everyone involved, so it will eventually be rewarded—perhaps immediately.

But any time you are doing something that does not work for everyone, it will eventually be branded spam or it will be made ineffective, or both.

A few years ago, all the smart SEOs were talking about “PageRank sculpting,” a technique that helped you control the level of quality that each page on your site has in the eyes of Google. All you needed to do was to carefully control which pages linked to which other pages on your site and voila!–you improve the search rankings of the pages you want. And it worked–for a while.

But who was that technique helping? It was helping the search marketer, but not really helping Google or the searcher, so eventually search engines stopped calculating their ranking algorithms the same way and the technique stopped working.

Today, many smart SEOs tell you that you must buy links to get the rankings you need. I know a few companies that do little else but sell links on content networks that are undetectable by search engines.You might or might not think that buying links is unethical, but it is clearly bad for the search engines, because the “wrong” pages (not the ones searchers might want) are ranked higher. So, the search engines are fighting back. Many SEOs ask me, “How will they stop it? No one can see the money change hands.” My answer is that I don’t know for sure how they will stop it. I am sure that the search engines use algorithms to sniff out the dicey links and give them less weight. But the search engines have a secret weapon.

They control their ranking algorithms. If it ever gets to the point that the search results are being too heavily affected by paid links, the search engines can stop valuing links completely (or as heavily). They can rank the pages based on page views, social media activity, or just about anything else that strikes their fancy. They can rank based on all these factors so that when all the factors don’t agree, the search engines downgrade—then you’ll have to fake links and fake social media activity and fake page views for it all to work (or even more stuff). The search engines haven’t done that (yet) because links still work well enough. But if enough people start buying links, then the search engines will find a way to regain the balance in the marketplace, because their very business depends on it.

So, the short answer is to stick to things that help everybody. Better content that contains the right keywords helps searchers understand things better. Be helpful or entertaining, because then you’ll attract links, activity, and every other visible indicator of attention—that marks your site as being of high quality. Everybody wins.

But if it feels clever and tricky and like something that you don’t want your competitors or customers to know you are doing, stop. If it isn’t against the search engines’ rules, it soon will be. If it is working now, it soon won’t be.

Search Engine Guide : Small Business Search Marketing

Related posts:

1. Examine Your Site’s Text, Reduce Chances of Search Engine Confusion

We have all been there once or twice, maybe a few more than that even. You just launched a site or a project,  and a few days pass, you login to analytics and webmaster tools to see how things are going. Nothing is there.

WAIT. What?!?!?!

Scenarios start running through your mind, and you check to make sure everything is working right. How could this be?

It doesn’t even have to be a new project. I’ve realized things on clients’ sites that needed fixing: XML sitemaps, link building efforts, title tag duplication, or even 404 redirection. The right changes are made, and a week later, nothing has changed in rankings or in webmaster consoles across the board. You are left thinking “what did I do wrong?”

A few client sites, major sites mind you, have had issues recently like 404 redirection and toolbar PageRank drops. One even had to change a misplaced setting in Google Webmaster Tools pointing to the wrong version of their site (www vs non-www). We fixed it, and there was a drop in their homepage for their name.

That looks bad. Real bad. Especially to the higher ups. They want answers and the issue fixed now … yesterday really.

Most of these things are being measured for performance and some can even have a major impact on the bottom line. And it is so hard to tell them this, even harder to do, but the changes just take …

Patience

That homepage drop? They called on Friday, as of Saturday night things are back to normal. The drop happened for 2-3 days most likely, but this is a large site. Another client, smaller, had redesigned their entire site. We put all the correct 301 redirects for the old pages and launched the site. It took Google almost 4 weeks to completely remove the old pages from the index. There were edits to URLs that caused 404 errors, fixed within a day, took over a week to reflect in Google Webmaster Tools.

These are just a few examples where changes were made immediately, but the actions had no immediate return. We live in a society that thrives on the present, immediate return. As search marketers, we make c-level executives happy with our ability to show immediate returns on our campaigns. But like the returns on SEO, the reflection of changes in SEO take time.

The recent Mayday and Caffeine updates are sending many sites to the bottom of rankings because of the lack of original content. Many of them are doing everything “right” in terms of onsite SEO, but now that isn’t enough. The can change their site all they want to, but until there is relevant and good content plus traffic, those rankings are not going to return for long tail terms.

There has also been a recent crack down on over optimized local search listings. I have seen a number of accounts suspended or just not ranking well because they are in effect trying too hard. There is a such thing as over optimizing a site, and too many changes at once can raise a flag with the search engines.

One Month Rule

Here is my rule: Make a change, leave it, go do social media/link building, and come back  to the issue a month later. It may not take a month, but for smaller sites, 2 weeks is a good time to check on the status of a few things. A month is when things should start returning to normal if there have been no other large changes to the site.

We say this all the time with PPC accounts. It’s like in statistical analysis, you have to have enough data to work with to see results. And when you are waiting for a massive search engine to make some changes, once they do take effect in the system, you then have to give it time to work.

So remember the next time something seems to be not working in Webmaster Tools or SERPs:

1. If you must, double check the code (although you’ve probably already done this 15 times) to ensure it’s set up correctly. But then,
2. Stop. Breathe. There is always a logical explanation. (And yes, Google being slow is a logical one)
3. When did you last change something to do with the issue?
4. If it’s less than 2 weeks ago, give it some more time.
5. Major changes, give it a month. (Think major site redesigns and URL restructuring)

SEOmoz Daily SEO Blog

No related posts.

We recently posted some correlation statistics on our blog. We believe these statistics are interesting and provide insight into the ways search engines work (a core principle of our mission here at SEOmoz). As we will continue to make similar statistics available, I’d like to discuss why correlations are interesting, refute the math behind recent criticisms, and reflect on how exciting it is to engage in mathematical discussions where critiques can be definitively rebutted.

I’ve been around SEOmoz for a little while now, but I don’t post a lot. So, as a quick reminder, I designed and built the prototype for the SEOmoz’s web index, as well as wrote a large portion of the back-end code for the project. We shipped the index with billions of pages nine months after I started on the prototype, and we have continued to improve it since. Recently I made the machine learning models that are used to make Page Authority and Domain Authority, and am working on some fairly exciting stuff that has not yet shipped. As I’m an engineer and not a regular blogger, I’ll ask for a bit of empathy for my post – it’s a bit technical, but I’ve tried to make it as accessible as possible.

Why does Correlation Matter?

Correlation helps us find causation by measuring how much variables change together. Correlation does not imply causation; variables can be changing together for reasons other than one affecting the other. However, if two variables are correlated and neither is affecting the other, we can conclude that there must be a third variable that is affecting both. This variable is known as a confounding variable. When we see correlations, we do learn that a cause exists — it might just be a confounding variable that we have yet to figure out.

How can we make use of correlation data? Let’s consider a non-SEO example.

There is evidence that women who occasionally drink alcohol during pregnancy give birth to smarter children with better social skills than women who abstain. The correlation is clear, but the causation is not. If it is causation between the variables, then light drinking will make the child smarter. If it is a confounding variable, light drinking could have no effect or even make the child slightly less intelligent (which is suggested by extrapolating the data that heavy drinking during pregnancy makes children considerably less intelligent).

Although these correlations are interesting, they are not black-and-white proof that behaviors need to change. One needs to consider which explanations are more plausible: the causal ones or the confounding variable ones. To keep the analogy simple, let’s suppose there were only two likely explanation – one causal and one confounding. The causal explanation is that alcohol makes a mother less stressed, which helps the unborn baby. The confounding variable explanation is that women with more relaxed personalities are more likely to drink during pregnancy and less likely to negatively impact their child’s intelligence with stress. Given this, I probably would be more likely to drink during pregnancy because of the correlation evidence, but there is an even bigger take-away: both likely explanations damn stress. So, because of the correlation evidence about drinking, I would work hard to avoid stressful circumstances. *

Was the analogy clear? I am suggesting that as SEOs we approach correlation statistics like pregnant women considering drinking – cautiously, but without too much stress.

* Even though I am a talented programmer and work in the SEO industry, do not take medical advice from me, and note that I construed the likely explanations for the sake of simplicity

Some notes on data and methodology

We have two goals when selecting a methodology to analyze SERPs:

1. Choose measurements that will communicate the most meaningful data
2. Use techniques that can be easily understood and reproduced by others

These goals sometimes conflict, but we generally choose the most common method still consistent with our problem. Here is a quick rundown of the major options we had, and how we decided between them for our most recent results:

Machine Learning Models vs. Correlation Data: Machine learning can model and account for complex variable interactions. In the past, we have reported derivatives of our machine learning models. However, these results are difficult to create, they are difficult to understand, and they are difficult to verify. Instead we decided to compute simple correlation statistics.

Pearson’s Correlation vs. Spearman’s Correlation: The most common measure of correlation is Pearson’s Correlation, although it only measures linear correlation. This limitation is important: we have no reason to think interesting correlations to ranking will all be linear. Instead we choose to use Spearman’s correlation. Spearman’s correlation is still pretty common, and it does a reasonable job of measuring any monotonic correlation.

Here is a monotonic example: The count of how many of my coworkers have eaten lunch for the day is perfectly monotonically correlated with the time of day. It is not a straight line and so it isn’t linear correlation, but it is never decreasing, so it is monotonic correlation.

Here is a linear example: assuming I read at a constant rate, the amount of pages I can read is linearly correlated with the length of time I spend reading.

Mean Correlation Coefficient vs. Pooled Correlation Coefficient: We collected data for 11,000+ queries. For each query, we can measure the correlation of ranking position with a particular metric by computing a correlation coefficient. However, we don’t want to report 11,000+ correlation coefficients; we want to report a single number that reflects how correlated the data was across our dataset, and we want to show how statistically significant that number is. There are two techniques commonly used to do this:

1. Compute the mean of the correlation coefficients. To show statistical significance, we can report the standard error of the mean.
2. Pool the results from all SERPs and compute a global correlation coefficient. To show statistical significance, we can compute standard error through a technique known as bootstrapping.

The mean correlation coefficient and the pooled correlation coefficient would both be meaningful statistics to report. However, the bootstrapping needed to show the standard error of the pooled correlation coefficient is less common than using the standard error of the mean. So we went with #1.

Fisher Transform Vs No Fisher Transform: When averaging a set of correlation coefficients, instead of computing the mean of the correlation coefficients, sometimes one computes the mean of the fisher transforms of the coefficients (before applying the inverse fisher transform). This would not be appropriate for our problem because:

1. It will likely fail. The Fisher transform includes a division by the coefficient minus one, and so explodes when an individual coefficient is near one and outright fails when there is a one. Because we are computing hundreds of thousands of coefficients each with small sample sizes to average over, it is quite likely the Fisher transform will fail for our problem. (Of course, we have a large sample of these coefficients to average over, so our end standard error is not large)
2. It is unnecessary for two reasons. First, the advantage of the transform is that it can make the expect average closer to the expected coefficient. We do nothing that assumes this property. Second, as mean coefficients are near to zero, this property holds without the transform, and our coefficients were not large.

Rebuttals To Recent Criticisms

Two bloggers, Dr. E. Garcia and Ted Dzubia, have published criticisms of our statistics.

Eight months before his current post, Ted Dzubia wrote an enjoyable and jaunty post lamenting that criticism of SEO every six to eight months was an easy way to generate controversy, noting “it’s been a solid eight months, and somebody kicked the hornet’s nest. Is SEO good or evil? It’s good. It’s great. I <3 SEO.” Furthermore, his twitter feed makes it clear he sometimes trolls for fun. To wit: “Mongrel 2 under the Affero GPL. TROLLED HARD,” “Hacker News troll successful,” and “mailing lists for different NoSQL servers are ripe for severe trolling.” So it is likely we’ve fallen for trolling…

I am going to respond to both of their posts anyway because they have received a fair amount of attention, and because both posts seek to undermine the credibility of the wider SEO industry. SEOmoz works hard to raise the standards of the SEO industry, and protect it from unfair criticisms (like Garcia’s claim that “those conferences are full of speakers promoting a lot of non-sense and SEO myths/hearsays/own crappy ideas” or Dzubia’s claim that, besides our statistics, “everything else in the field is either anecdotal hocus-pocus or a decree from Matt Cutts”). We also plan to create more correlation studies (and more sophisticated analyses using my aforementioned ranking models) and thus want to ensure that those who are employing this research data can feel confident in the methodology employed.

Search engine marketing conferences, like SMX, OMS and SES, are essential to the vitality of our industry. They are an opportunity for new SEO consultants to learn, and for experienced SEOs to compare notes. It can be hard to argue against such subjective and unfair criticism of our industry, but we can definitively rebut their math.

To that end, here are rebuttals for the four major mathematical criticisms made by Dr. E. Garcia, and the two made by Dzubia.

1) Rebuttal to Claim That Mean Correlation Coefficients Are Uncomputable

For our charts, we compute a mean correlation coefficient. The claim is that such a value is impossible to compute.

Dr. E. Garcia : “Evidently Ben and Rand don’t understand statistics at all. Correlation coefficients are not additive. So you cannot compute a mean correlation coefficient, nor you can use such ‘average’ to compute a standard deviation of correlation coefficients.”

There are two issues with this claim: a) peer reviewed papers frequently published mean correlation coefficients; b) additivity is relevant for determining if two different meanings of the word “average” will have the same value, not if the mean will be uncomputable. Let’s consider each issue in more detail.

a) Peer Reviewed Articles Frequently Compute A Mean Correlation Coefficient

E. Garcia is claiming something is uncomputable that researchers frequently compute and include in peer reviewed articles. Here are three significant papers where the researchers compute a mean correlation coefficient:

“The weighted mean correlation coefficient between fitness and genetic diversity for the 34 data sets was moderate, with a mean of 0.432 +/- 0.0577″ (Macquare University – “Correlation between Fitness and Genetic Diversity”, Reed, Franklin; Conversation Biology; 2003)

“We observed a progressive change of the mean correlation coefficient over a period of several months as a consequence of the exposure to a viscous force field during each session. The mean correlation coefficient computed during the force-field epochs progressively…” (MIT – F. Gandolfo, et al; “Cortical correlates of learning in monkeys adapting to a new dynamical environment,” 2000)

“For the 100 pairs of MT neurons, the mean correlation coefficient was 0.12, a value significantly greater than zero” (Stanford – E Zohary, et al; “Correlated neuronal discharge rate and its implications for psychophysical performance”, 1994)

SEOmoz is in a camp with reviewers from the journal Nature, as well as researchers from MIT, Stanford and authors of 2,400 other academic papers that use the mean correlation coefficient. Our camp is being attacked by Dr. E. Garcia’s, who argues our camp doesn’t “understand statistics at all.” It is fine to take positions outside of the scientific mainstream, although when Dr. E. Garcia takes such a position he should offer more support for it. Given how commonly Dr. E. Garcia uses the pejorative “quack,” I suspect he does not mean to take positions this far outside of academic consensus.

b) Additivity Relevant For Determining If Different Meanings Of “Average” Are The Same, Not If Mean Is Computable

Although “mean” is quite precise, “average” is less precise. By “average” one might intend the words “mean”, “mode”, “median,” or something else. One of these other things that it could be used as meaning is ‘the value of a function on the union of the inputs’. This last definition of average might seem odd, but it is sometimes used. Consider if someone asked “a car travels 1 mile at 20mph, and 1 mile at 40mph, what was the average mph for the entire trip?” The answer they are looking for is not 30mph, which is mean of the two measurements, but ~26mph, which is the mph for the whole 2 mile trip. In this case, the mean of the measurements is different from the colloquial average which is the function for computing mph applied to the union of the inputs (the whole two miles).

This may be what has confused Dr. E. Garcia. Elsewhere he cites Statsweb when repeating this claim. Which makes the point that this other “average” is different than the mean. Additivity is useful in determining if these averages will be different. But even if another interpretation of average is valid for a problem, and even if that other average is different than the mean, it neither makes the mean uncomputable nor meaningless.

2) Rebuttal to Claim About Standard Error of the Mean vs Standard Error of a Correlation Coefficent

Although he has stated unequivocally that one cannot compute a mean correlation coefficient, Garcia is quite opinionated on how we ought to have computed standard error for it. To wit:

SEOmoz Daily SEO Blog

Related posts:

1. Google vs. Bing: Correlation Analysis of Ranking Elements
2. Bing vs. Google: Prominence of Ranking Elements

Google’s Matt Cutts talked more about Mayday at SMX Advanced a couple weeks ago. He said that it was designed to try and spot signals of quality on pages and sites that would be good for users, and that auto-generated pages and content farms tend to get hit the most by Mayday.

More WebProNews Videos

Barry Schwartz at Search Engine roundtable is pointing to a WebmasterWorld thread indicating that there may have been another tweak on June 23rd and 24th that had a big impact on some sites’ rankings again.

Cutts’ advice to webmasters affected by Mayday in the first place, was basically to improve quality. I’m going to go out on a limb and suggest that this advice would probably still apply.

WebProNews – SEO

Related posts:

1. Mayday Update Confirmed by Google
2. Mayday! Your long tail is going to get shorter, but maybe fatter
3. Google incorporating site speed in search rankings