I am one of the last people at MoreVisibility who should be posting to our Analytics blog, but that’s exactly why I think I need to. If you are a regular reader of this blog, I apologize in advance, as the strategy I’m about to encourage is absolutely embarrassingly evident to you. Apologies also to Joe Teixeira, our most prominent Google Analytics expert for cluttering his blog with such low-level concepts.

That said, paying attention to Analytics is not for everybody, but it’s a skill set that someone related to your e-marketing efforts needs to possess if a program is going to be profitably managed in-house.

The options and triggers available today within Google AdWords and Microsoft AdCenter (soon to be powering SEM for both Yahoo and Bing) to incrementally nudge ahead the performance of your paid efforts are too significant to ignore.

Several steps need to occur:

• Properly coding your website and campaigns for Analytics
• Getting familiar with the data that populates
• Customizing the data to suit your program objectives
• Assessing Analytics data and making modifications to the campaigns based on past performance

The SEM industry has evolved to the point where there are typically a handful of companies in any industry / vertical that are adept at SEM and are effectively utilizing Analytics. Those are the firms to keep an eye on and learn from, but that will not realistically happen without an appropriate understanding and commitment to Analytics.

Whoa! The title of this blog post should have hit you like a ton of bricks. You see, I’m combining three different sub-industries together:

SEO = Search Engine Optimization
A/B testing = Well…A/B testing! (Like you can do with Google Website Optimizer)
Google Analytics = Web Analytics

Now, how in the world do we do A/B testing on our SEO efforts while using Google Analytics to evaluate the results?

On June 8th of this year, Google rolled out a new search index called Caffeine. This new search index collects and processes information across the web at a much faster rate of speed and accuracy than the previous search index. This means that you can update your website with new content and it is almost immediately indexed and available to be searched for on Google. Website owners who perform frequent updates to their site benefit the most, as Google will pretty much always have the latest version of your website in their search index.

A new search index that instantly updates itself also means that you can perform A/B testing on your organic search results and use a web analytics platform like Google Analytics to determine the rate of success or failure of your experiment. For example, let’s say that you are the owner of HardwareStore.com, and when a person searches for “hardware shop“, they see this organic listing:

As you know from reading our Search Engine Optimization blog over the years, the listings that you see in organic search results derive from the relevant web page’s “meta” tags (specifically, the title tag and the description tag). Before the launch of Google Caffeine, it would be several weeks, sometimes months, for your organic listings to be updated to reflect changes performed on your meta tags. Now, the changes are practically immediate. You can swap out the title and the description meta tags to deliver a different search engine result to your future website visitors who are searching for you on Google. Essentially, you’re running an A/B experiment to see which organic listing produces the best results.

Then what you can do in about one to two weeks after you’ve updated your meta tags is log-in to your web analytics tool and perform a date-range comparison to see what affect your change had on your site performance and task completion metrics, like in this example from Google Analytics:

As you analyze the above image, you can see that the change in your organic listings resulted in higher revenue and a higher conversion rate, which means that you should most likely keep the new meta tags for a while (but keep an eye on that average order value and crunch the numbers to determine if you really are more profitable in the long-term with a lower average order value and per-visit value).

Presto! You’ve successfully performed an A/B experiment with your SEO efforts, while using web analytics to measure the results!

About a year and a half ago, I wrote an article for our monthly newsletter that it “takes a village to raise a culture of web analytics” in an organization. One person – regardless of how motivated and hard-working he or she is – cannot create a culture of analytics and insights alone. It takes people from within your own organization, across organizations, and on your executive team to buy in to the analytics program and truly become an organization which values insights and data analysis.

I postulate that the same culture-driven approach can be applied to your company – no matter how big or how small – to create an environment where testing, experimentation, and optimization come first. As you know from reading our blog over the years, we here at MoreVisibility love Google Website Optimizer, a free A/B and multivariate experimentation tool. With Google Website Optimizer, you can create experiments rather quickly and efficiently and get excellent insights as to how your visitors are reacting to the new pages or new variations that you’re sending their way.

Creating a Google Website Optimizer experiment involves some important people that either work directly for you or that you’ve hired to do work for you (depending on your situation). It also involves six steps – from creating the account to installing the JavaScript snippets to launching your experiment – that you and your fellow colleagues, companies or vendors must participate in to ensure that your Google Website Optimizer experiment is a success.

Today, I’m outlining the six critical steps that need to be taken to successfully create and initiate your Google Website Optimizer experiment. Let’s begin!

Step 1: Deciding What to Test
People Needed: Site Owner, IT / Webmaster, Marketer, Web Analyst, Web Designer
Objective: To come up with a crystal-clear picture of what will be tested (what page, what section, what page is the conversion page) and what type of experiment will be run (A/B or Multivariate).
What You Need to Know: This first step should be a meeting with everyone that will be involved in this process, because it is a team effort. Everyone from the owner of the website to the IT guru should be heard and should have something to say about the upcoming experiment. At this step, you’ll decide on two major things: what you will be testing and what type of experiment you want to run with. If you want to do a simple experiment involving an original page and a variation (or “B” page), then the A/B experiment is your best bet. If you have lots of ideas that you want to test on a single page, go with the multivariate experiment. If you’re not sure, you can’t miss by choosing an A/B experiment.

Step 2: Creating the Necessary Pages or Variations
People Needed: IT / Webmaster, Web Designer, Site Owner
Objective: To actually create and upload the variation pages or variation sections (image and text) to use in the forthcoming experiment.
What You Need to Know: Once everything has been decided on, your web designer will actually create the variation page(s) or variation sections for use in your experiment. Once the web designer has done their work, the site owner should be involved to give the green light on any images or mock-ups to proceed. Once the site owner OK’s the variations, they need to be uploaded to the web server (Google Website Optimizer needs to verify that the variation page(s) exist).

Step 3: Creating the experiment in Google Website Optimizer
People Needed: Web Analyst or Marketer
Objective: Creating the Google Website Optimizer experiment and verifying that the pages to be used are on the web (uploaded to your web server).
What You Need to Know: The person who will be in charge of creating the experiment in Google Website Optimizer, which is almost always the web analyst or the marketing person, will go in and start creating the experiment within Google Website Optimizer. They will enter in the URLs of the original, variation, and conversion pages for verification, and grab the JavaScript snippets for the IT / Webmaster to install.

Step 4: Installing the JavaScript Snippets
People Needed: Web Analyst or Marketer, IT / Webmaster
Objective: Working with the IT department or Webmaster to install and verify the JavaScript snippets on the experiment pages.
What You Need to Know: The IT / Webmaster will follow the on-screen instructions provided by the web analyst or marketer and install the JavaScript snippets that need to be installed on all pages involved in the experiment. They will work with the web analyst or marketer to confirm that this has been done exactly as outlined by Google Website Optimizer.

Step 5: Reviewing the Experiment
People Needed: Web Analyst or Marketer
Objective: To review the entire experiment set-up and preview experiment pages before launch.
What You Need to Know: This is the time to review that everything looks good and that there are no design flaws or broken HTML on the experiment pages. If everything looks okay, the web analyst or marketer will launch the experiment!

Step 6: Launching the Experiment
People Needed: Web Analytics or Marketer, Site Owner
Objective: To activate the Google Website Optimizer experiment!
What You Need to Know: Once the Google Website Optimizer experiment launches, it’s important that the owner of the website is made aware that some visitors to his or her site will start to experience different variations of the site. It’s also important to let the experiment run its course naturally (without influencing it by making changes to the web pages involved in the experiment). After a few weeks, everyone on the Google Website Optimizer experiment team should meet and review the report data that the web analyst and marketer will love to analyze and derive insight from.

By following these six steps with the members of your organization, you are destined to succeed, while at the same time, creating a new culture of optimization and experimentation in your company!

Every website owner using Google Analytics has access to two unique reports that can display an insight into the sales cycle of your online business. These reports are called the Visits to Purchase and the Days to Purchase reports, and they are found within the Ecommerce report section. You will find the reports at the very bottom of the navigation menu (the last two reports in the Ecommerce section).

As you can see in the screen-shot below, a horizontal bar graph represents the number of visits it takes users to purchase an item from a particular website’s shopping cart within a given date range. While most visitors on this particular website purchased something after 1 visit (17.93% of all purchases), there are many other groupings of visits that have contributed revenue and transactions for this website, including the group all the way toward the bottom of the screenshot (201+ Visits, 14.32%).

Days to Purchase in Google Analytics

You can use this data to learn more about your customers’ behavior online, specifically, when they are on your website and purchasing something from your Ecommerce system. Is your website able to turn sales around in fewer or greater visits? Does it take many days (weeks or possibly months) before customers buy something from you? Have you applied an Advanced Segment in Google Analytics and compared segments, such as New vs. Returning?

Using this data can help your marketing and website optimization efforts as you learn about your website’s sales cycle. Most website owners would love it if every visitor converted on their first visit, but that isn’t always going to be the case. The easier, more competitive, and user-friendly your website is, the faster someone will become a customer of yours.

Let me take you back to the days when you were an under-21 college student, figuring out who you were and what you wanted to be when you finally grew up. For some of you this may be a lifetime ago, and for others, it may have seemed as if those days happened yesterday (literally, yesterday).

Most college students must take one, if not two courses in mathematics during their college careers, regardless of their degree program. Most of the time, elementary statistics is the course selected, probably because it’s the easiest math elective to take for most people. In short, lots of people have an elementary knowledge of statistics. So, why are average-oriented metrics put on such a pedestal?

In elementary statistics, you most likely learned about the four measures of center and about outliers. If you don’t remember, that’s OK, it’s probably been a long time since, or you probably weren’t a math person and wanted to forget everything you had learned as quickly as possible.

The four measures of center are mean, median, mode, and midrange.

Mean – The mean is what you know as the average. It is calculated by taking all of the values in a set and dividing them by the total number of values in that set. The mean is very sensitive to outliers (more on outliers in a little bit).

Example: The mean of 1, 3, 5, 5, 5, 7, and 29 is about 7.8571.

Median – The median is not the same thing as the mean, even though in popular parlance, the two terms are often used interchangeably. The median is the number that is in the middle of a data set that is organized from lowest to highest or from highest to lowest. The median doesn’t represent a true average, but is not as greatly affected by the presence of outliers as is the mean.

Example: The median of 1, 3, 5, 5, 5, 7, and 29 is 5 (the number in the middle).

Mode - The mode is the number that repeats most often in a data set. It’s seldom used in statistics as a reliable measure of center.

Example: The mode of 1, 3, 5, 5, 5, 7, and 29 is 5 (it repeats 3 times – the other values only appear one time each).

Midrange - The midrange is calculated by adding the highest and lowest values of a data set together, and dividing the sum by 2. The midrange is hardly ever used as a measure of center.

Example: The midrange of 1, 3, 5, 5, 5, 7, and 29 is 15 (29 + 1 = 30; 30 / 2 = 15).

With four different measures of center, I’ve been able to come up with four different correct calculations for an average. Each measure of center has its benefits and present different sensitivities to the presence of outliers. Depending on the set of data, the measure of center may lose strength and implied value because of how it is calculated and how it is used.

Outliers – Outliers are numbers in a data set that are either way bigger or way smaller than the other numbers in a data set.

Example: In the 1, 3, 5, 5, 5, 7, and 29 data set, the number 29 is an outlier because of how much greater it is than all of the other numbers in the set. 29 is the only number that doesn’t “fit” in this set.

What is the meaning of all of this?
The meaning of all of this is to take your averages (average order value, average conversion rate, average time on site, and others) with a tiny grain of salt. Use average-oriented metrics cautiously and with skeptical optimism, as the presence of a mere few outliers in your data can distort the figures and not provide a true representation of what is really happening.

Take this extreme example of the revenue of five separate orders placed on a web site:

$4.94
$4.39
$7.01
$6.33
$553.93

Your “realistic” average order value here should be $5.67 (the four “normal” values added up and divided by four). But if we’re looking at a report from a web analytics tool, it would report the average order value as $115.32. Clearly, there is a massive difference between $5.67 and $115.32.

To obtain real insights that will help your web site and your organization, you’ll have to dive much deeper beyond the averages to really exact meaningful information and data. Know your measures of center and your outliers, so that you can decide if your averages are realistic representations of what’s happening on your web site.

Until next time, I will leave you with one of my favorite all-time quotes, which fits right into this topic. Think about it the next time you’re obsessing over averages:

“A statistician drowned while crossing a river that was on average six inches deep”.