Most people understand that testing is important, but as Conrad Saam of Avvo tells WebProNews, properly testing is the only way to get successful results. According to him, one of the biggest mistakes that SEOs make is that they rely too heavily on averages.

Saam says that everyone within a business needs to be on the same level with statistics. If they are not, smart decisions cannot be made.

He advises SEOs to conduct a Binary Test. In medical terminology, this would mean that a patient either lives or dies. In other words, there is no middle ground between the two. Saam says online calculators are very helpful since they provide users with the required sample size for testing. To utilize these calculators, he suggests doing a Google search for an A/B calculator for confidence intervals.

The second test Saam recommends is a T-Test. This test consists of an array of data with some variability associated with it.  In SEO, he says this test could be a ranking improvement of 1-10 or 1-100. A T-Test gives users the statistical confidence level based on two different collections of data, which determines if “A” is better than “B.” Avvo uses Excel to conduct this test.

Although some people believe that more tests are needed, Saam doesn’t necessarily agree. For example, he says what happens if you survey 20 people and find that the 10 men surveyed are 4′ taller than the 10 women surveyed? If this were true, most people would feel comfortable saying that men are taller than women. However, if the survey found that the men were 4″ taller than the 10 women, then people would be less comfortable saying men are taller than women.

“Depending on that variability on the data, you need different levels of sample size to have confidence in your decision,” says Saam.

Is your testing helping you make smart decisions?

According to Richard Zwicky of Eightfold Logic, formerly Enquisite, marketers are not effectively monitoring and measuring online attribution.  Many analytics products attribute all or most conversions to the last click, which isn’t always correct.

By applying attribution to the wrong place, marketers are not able to understand how different channels contribute to the sales process. Zwicky says this of marketers, “Basically, they’re undercutting their own opportunities.”

He goes on to stress the importance of finding the true sources of value in order to increase conversions and, ultimately, grow the business.

Zwicky also talks with WebProNews about a new Eightfold Logic product for link building. The product applies a dating service model to the process of link building. Just like an online dating service, users begin by filling out a profile that explains what they want in a link, instead of a date.

Once a match is found, the system contacts the party and advises it to review the match. At this point, if the user wants an introduction and both parties agree, the system connects the two.

“It isn’t designed to overwhelm you with a lot of links; it’s designed to deliver you high-valued, contextually relevant links, which… in the end, is really more valuable,” says Zwicky.

For more information on the product, sign up here.

Posted by SeanWF

This post was originally in YOUmoz, and was promoted to the main blog because it provides great value and interest to our community. The author’s views are entirely his or her own and may not reflect the views of SEOmoz, Inc.

This is my first YOUmoz post, and I would greatly appreciate your feedback. I will be actively responding to comments, and I know that we will get a great discussion going. Please comment with any critique, questions, or random thoughts that you may have. If you would rather skip the statistics, feel free to jump ahead to the discussion section.

Introduction

A couple of months ago, SEOmoz explored the relationship between a web page’s PageRank and its position in search results. They concluded:

Google’s PageRank is, indeed, slightly correlated with their rankings (as well as with the rankings of other major search engines). However, other page-level metrics are dramatically better, including link counts from Yahoo and Page Authority.

I was intrigued by the study, and vowed to investigate the metric using my own data set. Because all of my data are at the root domain level, I chose to focus on the homepage PageRank of each domain.

Methods

I averaged three months of data (November, 2009 – January, 2010), collected on the last day of each month for 1,316 root domains. Using Quantcast Media Planner, I selected websites that had chosen to make their traffic data public. To be included, websites had to have an average of at least 100,000 unique US visitors during this time period.

The domains selected for this study do not approximate a random sample of websites. Because of the way in which they were selected, they will bias in favor of sites with many US visitors, and against sites with very few. There may also be differences between Quantified sites with public traffic data, and non-Quantified websites. For example, Quantified domains are probably more likely to include advertising on their pages than sites without the Quantcast script.

PageRank

PageRank (PR) can only take eleven values (0-10). It is an ordinal variable meaning that the difference between PR = 8 and PR = 9 is not the same as the difference between PR = 3 and PR = 4. Like mozRank, it probably exists on a log scale.

The median and mode PageRank among websites in this study were PR = 6, with a minimum of PR = 0, and a maximum of PR = 9. However, only ten websites had PR < 3, and only seven had PR = 9.

Results

SEOmoz Metrics

Using Spearman’s correlation coefficient, I compared PageRank to several SEOmoz root domain metrics. Domain mozRank (linearized) was strongly correlated with PR (r = 0.62)*. This correlation was somewhat smaller than the 0.71 that SEOmoz reported in May, 2009. The disparity may be due to differences in methodology; SEOmoz used Pearson’s correlation coefficient, and did not linearize mozRank. Additionally, PR data in my study were probably measured over a smaller range of values, potentially weakening the observed dependencies.

*All reported correlations are significant at p < .01.

MozTrust was also highly correlated with PageRank (r = .62), with Domain Authority somewhat less-so (r = .55). The latter has since undergone some major changes, and this result may not reflect the metric as it exists today.

Search Engine Indexing

I performed [site:example.com] queries using Google, Yahoo, and Bing APIs to approximate the number of pages indexed by each search engine. Much to my surprise, PageRank shared the strongest correlation with the number of pages indexed by Bing (r = .52), instead of Google (r = .30), or Yahoo (r = .24). My first thought was that Google might not have reported accurate counts, a phenomenon often noted by SEO professionals. However, there is some evidence that may indicate otherwise.

If Google’s reported indexation numbers are inaccurate, we would expect the metric to have lower correlations with similar metrics. However, indexation numbers reported by Google and Yahoo share a fairly high Pearson’s correlation coefficient (r = 0.38). Both appear to share smaller correlations with Bing: 0.34, and 0.26 respectively. Even more interesting, SEOmoz metrics seem to have much stronger correlations with Bing’s indexed pages than the numbers reported by Google or Yahoo.

If Google is failing to accurately report the size of its index, we might expect that similar queries would also return inaccurate data. However, PageRank shares a high Spearman’s correlation coefficient with the number of results returned by a Google [link:example.com] query (r = 0.65). The strength of this relationship appears similar to those between SEOmoz metrics and PR mentioned earlier. PR’s correlation with the results of a Yahoo [linkdomain:example.com -site:example.com] query is somewhat smaller (r = 0.53).

If the number of pages Google reports having indexed is a relatively poor metric, we would also expect to find more variation between months than other search engines. However, I did not find this to be the case. In fact, Bing had by far the highest average percent change in the number of pages indexed, a whopping 355% increase per month. Google averaged an increase of 61%, and Yahoo an increase of only 2%.

While it is still possible that the number of pages on each domain that Google reports to have indexed is inaccurate, I see another potential explanation. Moreso than Yahoo or Google, the number of pages that Bing will index on any given domain is related to the quantity and quality of links to that domain. Perhaps, at least when it comes to indexation, Bing follows more of a traditional PageRank-like algorithm. After all, Google claims that PR is only one of more than 200 signals used for ranking pages. This theory is supported by the results of SEOmoz’s comparison of Google’s and Bing’s ranking factors.

Social Media

PageRank even shares fairly strong correlations with social media metric such as how many of a domain’s pages are saved on Delicious (r = 0.49), how many stories it has on Digg (r = 0.38), and even the number of Tweets linking to one of its pages as measured by Topsy (r = .38).

Website Traffic

Last, but certainly not least, PageRank predicts website traffic with somewhat surprising strength. As reported by Quantcast, monthly page views, visits, and unique visitors are all significantly correlated with PR. Google’s little green bar even correlates with visits per unique visitor (r = 0.18), but not page views per visit. However, putting this in context shows the value of a metric like Domain Authority.

Discussion

So what exactly does all of this mean, and why is it important?

First, despite being a page-level metric, homepage PageRank is actually a fairly good predictor of many important domain-level variables relevant to SEO, social media, and website traffic.

For instance, on average, websites with a PR = 7 homepage had 2.6 times as many unique visitors as those with a PR = 6 homepage, which in turn had 1.5 times as many unique visitors as those with a PR = 5 homepage.

Second, homepage PageRank is sometimes used as a proxy for a hypothetical “domain PageRank.” While technically inaccurate, this study supports the idea that the PR of a website’s homepage provides information about the domain as a whole.

While it may be limited to just eleven possible values, PR it is surprisingly good at predicting the relative number of inbound links to a domain reported by Google and Yahoo, as well as the relative number of pages indexed by Bing. The key word here is “relative.” As an ordinal variable, PR cannot be used to predict the actual values of continuous variables.

Finally, this study provides evidence that SEOmoz’s domain-level metrics may be good (and possibly better than PageRank) predictors of variables important to search, social media, and web analytics. This, as well as all of the results of this study should be interpreted within the context of the included domains (high-traffic, US-centric, and publicly Quantified).

I hope you enjoyed reading my post, because I certainly enjoyed writing it. I intend to write many more based on your feedback. If you found this post interesting or valuable, I would greatly appreciate your thumbs up by clicking the icon below.

Do you like this post? Yes No

Posted by randfish

After last week’s Whiteboard Friday on the penalties paid links can incur, I got several questions about whether paid/spammy links could be used as a weapon to potentially harm someone else’s rankings. In this post, I’ll walk through why this is rarely the case, how you can defend yourself from potential scenarios and why this isn’t a great tactic to employ against your competitors.

Can Paid Links Be Used as Weapons in the SERPs?

The short answer is "almost never." But, as is typical in the SEO world, there’s a lot more in the long version.

In general, it’s very, very hard to bring down a white hat site/page ranking well in the search results. Although Google isn’t perfect at catching spam (e.g. our recent video featuring the success of some very obvious paid links in a well known network), they seem to be surprisingly excellent (almost prescient) at detecting the intent of links. My suspicion is that sites who buy links to prop up their own rankings have very different patterns than those who have competitors buying links to them. These patterns exist on the sites themselves, in other sites registered to the owners, in link footprints and in usage/search behavior.

It could, in fact, be that the "penalties" many SEOs often ascribe to paid links are in fact the result of a much more sophisticated analysis by Google looking at multiple aspects of a site’s presence before making a determination of the link intent. Given that, in nearly 10 years of SEO, I’ve only heard of two reasonably verifiable instances of "Google-bowling" (the process of pointing bad links at a site or page to hurt it’s rankings) working, my guess is that Google’s webspam team has developed some very impressive methods here.

Many SEOs have also suggested that a certain "bar of trust" can be achieved in Google, after which, negative links may be devalued, but likely don’t cause penalties or rankings drops. This makes a lot of sense to me (though it’s nearly impossible to prove), since "Google-bowling" is largely defeated and even good sites who stray into black/gray hat link building will simply find themselves wasting money, rather than being removed from the results (which could, for many popular brands/sites, cause a loss of relevance in the results for users).

Thus, if you are trying to wield paid links as a weapon against your ranking competitors, it’s far more likely to work against the new(ish) site ranking #65 for your keywords rather than those who’ve earned their way to the top spots with white hat techniques.

Defending Yourself from Potential Link Attacks

Have you recently broken the heart of a black hat link broker’s son or daughter? Stepped on a link farmer’s superhero cape? Talked smack about a nefarious panelist at an SEO conference not realizing they were just around the corner? The best defense, in this case, is a good defense (don’t go buying and renting links to others; you’re only enriching the spammers).

Many, many SEOs and webmasters worry a tremendous amount about spammy links pointing to their sites and pages. By and large, this isn’t a concern and it happens to every site on the web. Just look at some of the spamtastic links that point to SEOmoz (via this Yahoo! query):

If you see a collection of scraper sites filled with pharmaceutical, financial, legal, real estate and other questionable links with surprisingly well-optimized anchor text appearing in Google Alerts or your 24-hour reputation monitoring queries (e.g. http://www.google.com/search?as_q=seomoz&as_qdr=d&num=100 - which queries Google for all pages mentioning "seomoz" in the past 24 hours) don’t panic. If you exist on the web, you’re going to attract these types of links and the search engines will not punish you for it, even if you’re a relatively new, untrusted site.

However, if you start acquiring links that look an awful lot like they’re part of an intentional, paid link network (great anchor text, pointing to internal pages on the site, coming from footers and sidebars that contain other irrelevant, anchor-text rich links), there may be some cause for concern. Your best course of action is to submit a spam report to Google from your own, verified, Webmaster Tools account, noting that you have nothing to do with the links and want to make sure Google doesn’t think you’ve created, endorsed or paid for them.

This action is rarely necessary or worthwhile, but if you’re highly concerned about competitive conduct, it’s not a bad route to take. Of course, you’ll want to make sure you don’t actually engage in any black/gray hat activity yourself or it could trigger the wrong kind of review by a webspam team member.

Should I Buy Links to Push Down My Competitors?

Not unless you feel the link brokers of the world are more worthy than your favorite charity.

Seriously, the chances you’ll have a negative impact are far lower than the changes you’ll actually help (again, I refer back to our paid link WB Friday experiment in which the obvious link network had positive effects, even on the brand new site). The money is far better off spent on editorial content, public relations, social media campaigns and white hat SEO efforts for your own stite. Bringing someone else down may seem temporarily, emotionally satisfying, but it’s the wrong way to approach SEO (and life in general, if I may be so bold).

Looking forward to the discussion in the comments and happy to talk through the filtration processes and failsafes (or at least, my speculation) Google may employ.

p.s. The new Beginner’s Guide to SEO has more on understanding + recovering from search spam penalties.

Do you like this post? Yes No

Posted by Kate Morris

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)

Do you like this post? Yes No

a guest post by Larry Brooks of Storyfix.com

10.      This is a huge community.  As in, ginormous.  Literally four corners of the world, anyplace with digital cable and a Fed Ex partner. 

Which means my frequently sarcastic American humor doesn’t always play places like Klagenfurt and rural Kirgizstan.

9.        Online sarcasm is itself risky business.  One writer’s sarcasm is another’s snarky… a word which probably doesn’t play in Kirgizstan, either. 

8.        Never write a post about the need to double and triple check for typos that has a typo in it. 

One word: crucified.  Still smarting from that one.

7.        “Know Thy Audience” isn’t a cliché.  It’s the natural law – the physics – of marketing.

I’m a blogger who posts about fiction writing and sells a few writing ebooks while I’m at it.  The majority of readers here are online entrepreneurs who’d rather hear about blog-related marketing than how to write the next Salzburg Times bestseller. 

Many of whom, by the way, have a story in them.

6.       Darren Rowse really is the nicest guy on the internet.  A total pro, too.   I’ve tested this theory with a wide breadth of technical cluelessless and naiveté, and you can add patience to those first two.

He doesn’t just let anybody onto this site, which means you not only earn your admission ticket (lest you wonder, I was invited to post here twice a month), you earn your keep, too.  And it’s all fair. 

5.        The company you keep defines you.  Choose wisely. 

In this case, being on Problogger has upped my online exposure and, merely by association, my chops in the online world.  My brand.  Which means, the pressure is on.

This, too, is natural law in the online world.

Because the same crowd that throws in on that count can slap you back to reality with one missed swing.  (That being three metaphors in one sentence… don’t try this at home.)

4.        It’s okay to get personal.  And I’m not talking about dating or social media sites (getting too personal on those venues can also get you arrested). 

A blog is usually an ancillary tool in an otherwise pointed branding and marketing strategy, which means it doesn’t need to exclusively spew bits and bytes (digi-speak for features and benefits) or self-serving bluster that doesn’t smack of commonality. 

People are attracted to commiseration, empathy and the voyeuristic joy that comes from reading about the sheer misery of others in like-minded situations.

3.        There’s one in every crowd.   Try not to be that guy.

You could blog about the reliability of death, taxes and gravity and somebody will post a comment endeavoring to make you wrong (one self-proclaimed “blogging superstar” tried to refute my theories about writing and publishing contemporary fiction by quoting Cervantes, who published his last book in the year 1615 … but that’s another site). 

That which doesn’t kill us either makes us stronger or simply pisses us off. 

2.        You, the blogger and the commenter, put the UNITY into community.   That’s why this venue is unique in all of the history of human communications.

And the most valuable thing I’ve learned here on Problogger is…

1.        I have a lot to learn.  That’s why we’re all here, isn’t it? 

One of the best ways to learn – albeit with a resource like Problogger on your daily to-do list – is to just keep writing.  On your own site, and on others if they’ll have you.

And if that’s not common ground, perhaps we’re all in the wrong place.

Larry Brooks is the creator of Storyfix.com, an instructional site for fiction writers and those who proof them.

Post from: Blog Tips at ProBlogger.

10 Things I’ve Learned From Posting on Problogger

Share This

As mobile grows, we also see the rise of local search. Dylan Swift of Yelp spoke with WebProNews about this phenomenon recently at SMX Advanced. He explains how companies such as Yelp are helping to close the gap between the PC and local business. These companies are doing this through the mobile device.

From Yelp’s mobile application for the iPhone, it can see that there is a phone call happening to a local business every 5 seconds of every day. Another statistic they are seeing is 27 percent of all the company’s searches are happening through their app. In addition, a little less than 1 million mobile searches for directions to a local business have occurred in the last 30 days.

In other words, Yelp is helping bring consumers and local businesses together. With the continued adoption of smartphones, Swift does not see mobile traffic dropping for a while. For example, the company saw 1.8 million unique visitors through its mobile app in the last month, which is up from less than 1 million 9 months ago.

Yelp also has an app for the iPad. Swift says it is similar to what users see on their website, since they have a more real estate to work with on the iPad. Although the majority of Yelp’s mobile traffic comes from the iPhone, Swift pointed out that the company is investing in Android as well.

Posted by bhendrickson

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:

Do you like this post? Yes No

A Guest post by Cori Padgett from Big Girl Branding.

Perfection is for losers.

There, I said it.

And I can say that because I used to be a bit of a perfectionist. I’m pot, you’re kettle, and we’re both freakin’ black. Feel better?

And frankly, I still am if I’m honest about it. You could say I’m a “recovering perfectionist”.

Seriously, I think it stems from my slightly O.C.D. tendencies. Tendencies like the insane urge to eat the same amount of M&M’s on each side of my mouth. Only green on the left, only blue on the right.

Or the ridiculous compulsion to leap out of bed at midnight, knowing I already locked the front door… but feeling compelled to check it one more time, “just in case”.

Or my friend who literally will follow the cord from the iron all the way to the wall, KNOWING it’s not plugged in, but making extra sure it isn’t anyway…

Yep…

We’re nuts. Isn’t everyone?

At least a little?

OK, fine, maybe that’s just me… sheesh.

Honestly, striving for perfect is a losing battle. The only perfect being in this world is the Good Lord Himself, and last I checked… I’m not God, how about you?

No one is perfect. And when you’re constantly seeking perfection in everything you do, you’re doomed to a life of dissatisfaction, discontent, and stagnation because you are unable to move forward on any of your goals.

And if you’re a writer and a blogger like me… well let’s just say that if I tried to reach perfection in everything I did… I’d likely still be cleaning vacation homes for a living and praying all of my bills got paid.

So hold on a sec while I give imperfection a big juicy kiss on the lips in monumental gratitude for saving me.

Muah! Now seriously, let’s talk more about you.

• If you’re struggling each day within the chokehold of perfection…
• If you’re wasting time and valuable resources trying to get something “just right” before you do anything with it…
• Or if you find yourself deleting pages and pages of work because you suddenly decided it “wasn’t any good”… Nay, scratch that… let’s call it like it is shall we? Because it “wasn’t perfect”…

Then I’m sorry friend, but it’s time.

Time that is, to don your steel-toed boots (“sh%# kickers” we call ‘em in the Dirty South) and kick perfection’s rosy little ass to the curb! Preferably with a resounding splat for good measure.

You with me? Good, let’s get started.

The First Swift Kick- Set Attainable Goals

I’m talking about blogging and writing here, but really this can be applied to just about anything. Perfectionists tend to make a habit of setting unrealistic goals for themselves. And then when they don’t measure up or reach those goals, they’re ashamed of themselves.

They call themselves quitters, or “stupid”.

They put themselves down, and engage in some pretty negative self-talk… telling themselves they just aren’t good enough to get where they want to go… if only they could do “this” they’d be so much better at “that”.

I mean really… would you talk to someone else like that?

Not likely.

Would you call someone stupid because they couldn’t do something perfectly?

Probably not.

So why in the ever-lovin’ world would you talk to yourself that way? Just sayin’.

Stop setting goals for yourself that you don’t believe in. Strive for excellence in all you do, not perfection. Yes, you should always stretch yourself. And yes goals you set for yourself should feel slightly scary… slightly out of reach even.

But they shouldn’t feel impossible.

For instance, I have a goal to reach 2,000 new subscribers to Big Girl Branding in the next 6 months. (Feel free to help a girl out btw!)

Is that goal attainable? I believe it is.

Is it slightly out of reach for me right now? I’d say so, as right now between RSS, Email, and my newsletter, I’ve only got about 250 subscribers. (Thanks to ALL of you, BIG hugs!)

That means I’ve got about 1750 more to go before I’ll reach that goal. And you can bet that if I got hung up on creating “perfect” blog posts, and having a “perfect” design, and making my newsletter a “perfect” mix of humor, smarts, and useful info… I probably wouldn’t even have any readers, much less a subscriber!

I’d be totally stuck.

Totally immobilized.

Living in fear of being judged.

Living in fear of being seen as “imperfect”.

And if I set that goal even higher, say 10,000 subscribers in 6 months… I’d likely be completely stalled because the goal I set for myself wasn’t believable to me and I’d be so hung up on being perfect that my inner mantra would end up being something along the lines of “I’ll never get there” or “It’s too hard” or “I’m not ready”.

Bull.

It will be hard. I’ll probably never be 100% ready. But I will get there.

And so will you, wherever “there” is.

But I know that you’re smart enough to know that you won’t get there overnight, and you won’t get there by getting stuck under the thumb of perfection. So set incremental, attainable goals for yourself that feel just mildly out of your comfort zone, and then get rockin’ with the action taking.

Set about making them a reality.

As you do that, repeat after me… “Action will get you everywhere, perfection will get you nowhere.” “Strive for excellent, not perfect.” Now say that three times fast! (Kidding.)

Then get busy creating excellence in all you do, and let go of that ridiculous notion of perfection.

Capisce?

The Second Swift Kick- Enjoy the Process of Achievement

Really!

When is the last time you (speaking to you perfectionists here) stopped long enough to appreciate where you are right now?

To appreciate what you’ve accomplished already? As a perfectionist there is a tendency to be constantly looking for new and better ways of doing something. A bit like the “grass is always greener” syndrome. You’re never quite satisfied with anything “as-is”.

So you need to make it a habit to pat yourself on the back for every milestone that moves you a step closer to your goals, whatever they may be and appreciate the moment you’re in.

For me, that means appreciating the fact that I have 250 subscribers that read my blog already! That’s no small feat, especially when you’re starting out.

And people these days are stingy with their time and their emails… so if they are taking the time to keep up with you and your blog… that’s a huge compliment! Treat it as such and call yourself a winner, because even when you don’t feel “perfect” you’re still pretty freakin’ awesome.

Just tell yourself “Cori said so!” if ever you’re feeling doubtful.

The Third Swift Kick- Connect With Other People

Preferably other imperfect ones. (And in case you forgot… that means everyone is fair game!)

Honestly, sometimes when you’re feeling stuck in the spinning abyss of an “I can’t do this, it’s not perfect, everyone will hate it!” moment…

Slow your roll man!

Stop what you’re doing (or trying to do) get up and walk away. Get out of your house, go have coffee with a friend or three, and cop a squat in a park somewhere to see how the rest of the world lives. It’s a sure bet that things aren’t perfect for the rest of the world either!

(I know, I know! You… meet sledgehammer wrapped up neatly in “no one’s perfect” stickers!)

You’ll even be singing the “No One’s Perfect” theme song before I’m done with you.

Really though, connecting with other people can help you see that you’re not the only one that struggles with the insane urge to be perfect all the time. Trust me there are a ton of us out there!

You can probably safely bet that Darren, despite his pretty massive successes in the blogosphere has never been perfect. (Sorry Darren!)

And heck, Bill Gates went after what he wanted with barely a plan in place! That’s so far from perfect it’s laughable, but just look where he is today.

And sitting down with friends who are equally crazy but imperfect can help you see that perfect isn’t necessary for success. The only thing necessary for success is the ability and willingness to do what needs to be done when it needs to be done in the absolute best way that you can right then.

No more, no less.

The Fourth and Final Swift Kick- Accept Mistakes for What They Are

A somewhat painful learning experience. As a recovering perfectionist, I know there is often a tendency to view mistakes and screw-ups as failures.

It’s now time to change your point of view.

Mistakes are not failures they are lessons learned.

The only time a mistake is a failure is when you don’t walk away with new knowledge about yourself and your goals. When you don’t walk away with new self-awareness, I’m afraid that means you’re doomed to repeat those mistakes, sometimes over and over again until you get it.

And guess what it means when you repeat a behavior but expect a different outcome?

C’mon!

It means you’re insane, so stop that!

Mistakes are a part of life, a part of learning and growing. I’ve made more than my fair share over time and it’s a sure bet you will too. I can state this with absolute assurance because I already know that perfection is never attainable for mere mortals like you and me, despite our best efforts.

But it’s time to realize that you don’t have to be perfect. Loosen the chokehold a bit. You don’t have to be a perfect blogger, or a perfect writer, or a perfect mother or father. You don’t have to run a perfect business, or have a perfect home.

Just make it your goal to always give your best; to strive for “excellent” in everything you do… and you’ll get as pretty darn close to perfect as you’ll likely ever be.

And on that note…

“Excellent” is pretty damn good in my book, so let me know if you need to borrow it.

Cori is a wildly hire-able freelance ‘ghost’ as well as the creative brains and dubious brawn behind her blog Big Girl Branding. If you’d like to harness her creative brains and dubious brawn to guest post on your blog, just stalk her on Twitter and ask her. I’m “almost” sure she doesn’t bite. Well… like 95% sure.

Post from: Blog Tips at ProBlogger.

Blogging, Steel-Toes, And Kicking Perfections A$$

Share This

This column is written by Kimberly Turner from Regator (a great tool that gathers and organizes the world’s best blog posts). – Darren

Thanks, as always, for stopping in for our weekly list of the ten most blogged-about stories, provided by Regator. This week, we’ll use posts about these hot topics to discuss thoroughness in blogging. “Thoroughness” can be a vague term, so I’ll define a thorough post as a post that tells the reader what they would want to know about a given topic and does not leave them with unanswered questions. Let’s take a look at some great examples:

1. World Cup – Flavorwire’s “First Person: Scenes from England’s World Cup Fever” uses thirteen photos and accompanying text to paint a vivid and complete portrait of England’s World Cup fever.
2. iPhone 4 ­– In “Word on TheStreet is that you shouldn’t buy an iPhone 4,” TUAW does a point-by-point rebuttal of a post from TheStreet.com. Posts or articles you disagree with can be a rich source of inspiration, just be sure to adequately address the points made in the original during the course of your rebuttal post.
3. Stanley McChrystal – In “What Gen. McChrystal should have known about Rolling Stone’s reporter going in,” Slate’s Press Box blog spends more than 1,000 words elaborating on why McChrystal should not have agreed to take part in the Rolling Stone profile then adds a level of completeness by providing a dissenting opinion and asks readers to discuss the issue.
4. Father’s Day – You need not be reporting on news to provide a thorough post. Miche G. Hill’s “My Dad: A Father’s Day Story” uses personal anecdotes and experiences to build a connection between her readers and her late father.
5. Gulf of Mexico – Many blogs were quick to put up posts indicating that a federal judge had blocked Obama’s proposed drilling moratorium, but “Judge Strikes Down Obama’s Offshore Drilling Ban” from Treehugger went a step further by providing quotes from the judicial opinion and the White House press secretary, speculation on why the judgment was made, and a link to supporting documents. Providing these extra elements requires research, but the time spent is likely to strengthen your post and increase your credibility.
6. Toy Story 3 – While many other posts on Toy Story 3 mentioned the tear-inducing nature of the film as part of a broader review, Cinematical’s “Why Does Pixar Make Growing Up Feel So Bad?” focuses in on that particular aspect of the blockbuster. If a topic seems too large to cover in a thorough manner, consider honing in on one particular aspect and covering that aspect well.
7. Supreme Court – Like number 5 above, SLOG’s “R-71 Case: Supreme Court Rules Petitions Can Be Released” demonstrates that it is built upon solid research and was not just dashed off in haste.
8. Miley Cyrus – Speakeasy’s “Miley Cyrus’ ‘Can’t Be Tamed’: Review Revue” combines reviews from various sources to create a one-stop post for those interested in how this pop star’s latest album is being received. Pulling together information from various sources can be helpful to your readers—so long as you also provide them with ample original content.
9. Wimbledon – The Guardian’s “Wimbledon 2010 Live Blog: 23 June” may be one of the best examples of thoroughness ever to grace the Blogosphere. When Xan Brooks was assigned to the seemingly enjoyable task of live blogging Wimbledon, he almost certainly never expected the longest match in the history of tennis. Although he was, by the end of the 11+ hour match, rambling about zombie players and hearses, the champion never gave up.
10. Kevin Rudd – Rather than simply linking to Kevin Rudd’s farewell speech video, Jack Marx’s “Kevin – too human, too late” analysed it, commenting on everything from the former Prime Minister’s eye contact and body language to the reaction of his son during the video.

Do you think about the thoroughness of the posts you write? Please share your thoughts in the comments!

Kimberly Turner is a cofounder of Regator.com and Regator for iPhone as well as an award-winning print journalist. You can find her on Twitter @kimber_regator.

Post from: Blog Tips at ProBlogger.

Blogosphere Trends + Thoroughness in Blogging

Share This

Newer Posts »