If you have a website, one of the most crucial factors to its success is the continual examination and study of its Performance Analytics. Without knowing where your areas of success and failure lay, you will not be able correct any existing issues or capitalize on any potential opportunities. In short, you will not be able to form a marketing strategy because you will not have the data necessary to formulate said strategy. How do you know where to go if you fail to recognize where you currently stand?

Understanding and Using Google Analytics For Ecommerce


There are many ways to gather data and study a website’s Analytics, but by far one of the most widely used and easy to understand is Google Analytics. For that reason, it is worthwhile to learn the ins-and-outs of the system. A large community uses it, it’s continually updated and improved upon by Google, and a big benefit to most, it’s free. Due to the large user base, resources on the subject can be found everywhere online. So any problem or question you might have, has likely already been answered. This makes using Google Analytics easy, but more importantly, extremely worthwhile.

Let’s get into some of the important fundamentals to understand when looking into Google Analytics.

Measurement Plan, KPI’s, and the Dashboard

Understanding Google Analytics properly means first setting up a Measurement Plan, which should attempt to identify 5 primary points:

  • The company’s primary business objectives
  • The tactics and strategies to be implemented in order to achieve said objectives
  • The Key Performance Indicators (KPI) intended to measure success
  • Customer segments in order to have an intimate understanding of what causes success
  • Target goals for each KPI chosen which will be used to value whether the business is reaching their expected goals

Once your measurement plan has been formulated, you will then go on to create a custom Dashboard in Analytics, which will have all of your targeted KPI’s included. This is an excellent means to be able to monitor the daily health and standing of your website.

You should be sure to segment dashboards created to specific job levels and functions to monitor certain important areas of a business. For example, on a broader scale you may want to quickly be able to see revenue over a certain time period, revenue compared to a specific time frame, websites traffic, or the geographical regions your users are located in. On a more narrow scale and when focusing on lets say marketing, you may want to see where the individual channels traffic is coming in from, how landing pages are performing, or how certain products or categories are fairing.

The dashboards you create will be specific and individual to your businesses KPI?s, and the roles of the people viewing that information. If you use your Measurement Plan as a guide to understanding what will be needed in the dashboard, nothing will be left out and important information can be easily viewed and analyzed.

For example, if you are trying to briefly look at the behavior of New vs. Returning customers, you may want to create a dashboard that highlights Pages/Session, Bounce Rate, Average Session Duration, and Average Time on Page. In short, this will show you which type of user tends to engage best with your website.

New versus returning user behaviorWhat this data alludes to is that for this example website, returning customers view more pages, have a lower bounce rate, and have significantly longer average session durations than compared to new users. Remember that the Dashboards you create can be customized in any number of ways. Be sure to create specific Dashboards that give a good overview of KPI’s important to your business.

Analysis Preparation and Segmentation

You can’t always expect the Analytics dashboard data to align with what your exact expectations were. Of course, you should always be aiming high, but when the data is off or points to something being wrong, you can use Google Analytics features to inspect and identify what the problems are. One of the most important tools you should understand is Segmentation.

When you apply a segment, you are filtering data in real time. You will be able to compare subsets of analytics data to identify poor performing segments. You are able to Segment Users or Sessions. Users are defined as the people who visit, interact, and engage with a website. Where a Session is defined as a group of interactions that take place on a websites within a set time frame.

User Segments are able to span across multiple Sessions within a specific date range. This can be configured to display User traffic, sales, actions, and other metrics on a website over a specific period of time. You can then compare these Segments against one another.

For example, you can compare the traffic source of New vs. Returning users. This data will be Segmented by Direct Traffic, Organic Traffic, and Referral Traffic.

New versus returning user graphIn this example, by Segmenting the data in this way, we can identify that though Organic traffic has sent the most users to the website, those users are not as engaged with the website as those arriving from Direct or Referral traffic. Organic traffic has a much higher bounce rate, fewer page views, and significantly shorter average session duration. Based on this information, we would then investigate as to why user arriving from Organic search tend to have a poorer experience onsite compared to the other two user groups.

Session Segmentation on the other hand is where user behavior is tracked only for a single set time frame. You will be able to analyze which business goals or KPI’s were accomplished by a user during a specific Session. An example of a Users overall Session might look like:

  • Session Action 1 – User visits page
  • Session Action 2 – User visits a second page
  • Session Action 3 – User fills out email newsletter sign-up form
  • Session Action 4 – User shares page content on Social Media
  • Session Action 5 – User completes a transaction

A single User is able to open up multiple Sessions which can occur on the same day, week, or over the course of a month. Once a Session ends, which is based on time-based expiry and Campaign settings, there is then the opportunity for a new Session for that user to begin. By default, a single session is set to last 30 minutes. Remember, a Session is deemed as the actions a user takes onsite and determining which Sessions are important to compare against one another is dependent upon the business itself.

Using the same example as before, if we want to see the Bounce Rates for all traffic sources to investigate further why Organic traffic Bounce rates are so high, we would set Bounce Rate as the additional metric.

Audience overview session bounce ratesBoth User and Session segments can be built using dimensions, metrics, session dates, and sequences of user actions. When first viewing your Audience Overview, notice your Segments is automatically set to All Sessions and can be changed to show specific and other desired segments. You are able to show up to 4 segments at any given time and this is extremely useful for comparing data. Once these have been set, Segments will apply to other reports as well so there is no need to change them repeatedly.

Beyond the Segments that are automatically included in Analytics, you are also able to create your own customized Segments. You will be able to name your customized Segments and choose specific criteria that the Segment is to report on. The criteria fields include Demographics, Technology, Behavior, Date of First Session, Traffic Sources, and Ecommerce. Advanced options also include Conditions as well as Sequences of User Interactions. As you begin to ask more complex questions about your Users behavior, you can use Segmentation to isolate the right data you want to analyze and identify opportunities to increase your websites overall performance.

Other Important Analysis Tools

Multi-Channel Funnels: The Multi-Channel Reports shows how marketing channels (the sources of traffic to your website) work in conjunction with one another to generate conversions. A conversion does not typically occur upon a users first visit. They will shop around and compare prices before finally committing to a certain brand or product. In that time, users will visit various other websites, and even perhaps find your website or your content in another form, such as a paid advertisement or email marketing piece.

Multi-channel funnels overviewMulti-Channel Funnel reports then are generated from the conversion paths that directly lead to a transaction or sale. The conversion paths and the data sets used to determine the report are, but are not limited to:

  • Paid and Organic Search
  • Keywords Used in Search
  • Referral Sites
  • Affiliate Sites
  • Social Media Networks and Platforms
  • Email Newsletters

Multi-channel conversion visualizerIn the Multi-Channel Visualizer, you can see percentages of total conversions for all possible and tracked traffic sources. The Visualizer to the right gives a representation in the form of circles for the overall importance of each traffic source, and how they worked in conjunction with one another to help complete a conversion by a user. Note, that the areas where circles overlap are traffic sources that worked together in funneling traffic to your site that converted.

For instance, the center where all traffic source circles meet comprise 1.6% of all traffic that converted. This means that for 1.6% of all conversions, users landed on your website from each of the 5 traffic sources before converting. Organic, Referral, and Paid Search traffic combined accounted for 4.75% of total conversions. Where Users who converted only after arriving from Organic Search comprises nearly 55% of total conversions.

These numbers for combined traffic sources that converted are shown when you hover over the circles with your mouse cursor. Once you begin to digest this information and understand how traffic sources work together, and which are the most profitable for your business, you can then dig further into analytics to discover possible solutions to help increase conversions.

The Multi-Channel Funnel Report shows every channel that should be credited with playing a role in an actual sales conversion. It’ll explain how often a specific channel helped or assisted in completing a conversion as well as the assumed value of that conversion. You can learn the length of time it took for a conversion to be completed, and in sum, know everything that happened from a users first interaction with your website to their last interaction. In the end, Multi-Channel Funneling is important to study and understand as it greatly helps to identify any problems occurring or opportunities that are available within your various marketing channels.

Attribution Modeling: An Attribution Model is a set of rules that determine how conversion paths are to be measured. In essence, how credit for completed sales and conversions are to be assigned and their overall value. The value itself will change depending the Attribution Model used. For example, if a certain marketing channel creates conversions (ie. Paid Advertising) rather than visits (ie. Email Marketing), or vice versa.

The Model Comparison tool for Attribution Modeling allows you to compare how different Attribution Models impact the valuation of your marketing channels.

Model comparison toolYou may find that though you intended a specific marketing channel to perform one, it is actually performing in another way that is still worthwhile. Without testing and comparing your Attribution Models, you may be lead to believe that a certain marketing channel isn’t performing well, though it may simply be performing well in another manner.

In the image below, you can see that we’ve taken a closer look into the Organic Search section of the Model Comparison Tool. From here we can dig in further to find out which Search Engine?s are sending us the most traffic that converts. In our example below, the grand majority of traffic that converts arrives from Google.

Model comparison tool filter viewHow deep you study the Model Comparison Tool is up to you. Understanding that an opportunity exists is important to capitalize on and knowing that a problem is occurring is important in being able to correct it. These two reasons primarily, for everything Analytics has to offer, are the greatest reasons for its use. Knowing whether or not some action in the past worked for you greatly determines the proper course of action to take in the future. In the end, this is what you are searching for and seeking to uncover.

Secondary Dimensions: Google Analytics reports can display a subset of all the available dimensions, and custom reports are able to display even more than the few that are automatically displayed. There are Primary Dimensions and Secondary Dimensions.

Primary Dimensions are built into every Analytics report. It displays the most relevant information for the chosen data set. For example, in the Language report, the primary dimension initially is Language. In some reports you are unable to change the Primary Dimension, though this option is still available for most.

Primary dimensions filteringSecondary Dimensions can also be included in most reports. These help to compare data sets against each other and more than one Secondary Dimension can be included on a report. These can include the Acquisition, Advertising, Behavior, Custom Variables, Ecommerce, Social, Time, and User Dimensions. Depending on the data you hope to use, compare, and study will determine the exact Dimensions, both Primary and Secondary, to be used.

In the example image below, we’ve added Country/Territory as the Secondary Dimension to compare against the Language (Primary Dimension) the user has set in their browser preferences. This helps to understand the preferred language in use depending on the country the visitor is visiting from. This is important to note as perhaps a large percentage of users prefer a different language than what you use on site. It may make sense if it appears enough users drop off due to a language barrier, to make your website bilingual, or even create custom landing pages for certain user language preferences. Obviously this all depends on the number of users and the value inherent in offering your content in another language. But nonetheless, this is very important to watch and keep track of.

Secondary dimensions filtering

Analyzing Your Traffic Sources

In order to understand how effectively you are targeting, enticing, and capturing customers, you want to look at and analyze your traffic source data. Traffic source refers to the origin of a referral to a website. The main potential sources include: Search Engines, Referring Sites, Email or Newsletters, and Direct Traffic.

Analyzing Traffic Source data and information will alert you to what is actually happening during sessions that were initiated by specific traffic sources. For example, which traffic source generates the most revenue, has the greatest conversion rate, drives the highest-valued transactions, or allows for enhancing the user experience the best? Without analyzing your traffic sources, you won’t have the information you need regarding which traffic sources are most important to your business. Which should you target and capitalize on and which should you whole-heartedly avoid. Examining and studying your data will highlight these areas for you so improvements can be made.

Traffic source viewMulti-Channel Traffic Analysis: Before we were measuring success based off of conversions and revenue, but there is another way to evaluate your sites traffic channels. We can look how well certain channels assisted other channels in generating conversions. This supplementary and supportive assistance is important for every website and just because a certain channel is not directly generating sales itself does not mean it is not helping others to accomplish that same goal.

The best example of this is Display Advertising. Display Ads on the Google Display Network primarily assist other traffic channels in generating transactions, conversions, increasing brand awareness, and simply engaging with users early on in their buying cycle. This is called Assisted Conversions and it is important to keep it in mind. You would hate to cut marketing funds on a certain traffic channel because you wrongly believed it had no value. To avoid over looking these conversion assistants, use the Attribution Modeling Tool to help understand the over arching worth for specific channels that help with conversions.

When doing Multi-Channel Traffic Analysis there are four major questions you will want to be asking yourself.

  1. Of all your Traffic Channels, which engage best with the customers throughout their buying cycle?
  2. How well does the relationship between your marketing channels and campaigns work to generate conversions?
  3. Can you target anything in your channels or campaigns that is performing poorly and improve upon it?
  4. If a channel does not generate direct conversions, what is the actual value of that channel?

Answering these questions will ensure you’re keeping yourself on the right path at all times when doing a Multi-Channel Traffic Analysis. You can use these questions, in essence, as a guide.

Analyzing Your Customers

Understanding your current customers greatly helps to define and target potential future customers. Learning about customers who have completed a sale or conversion on your website is the first step to creating a better overall marketing strategy. After analyzing your customers who have completed a purchase, you should be able to answer these 4 necessary questions:

  1. For customer who converted compared to those that didn’t, what was the behavior of the user?
  2. For customers who converted compared to those that didn’t, what were the demographic characteristics?
  3. What technology platforms were being used, and did any of them make completing a purchase easier or more difficult?
  4. For users who did convert, are there any other noteworthy characteristics to be concerned with?

Analyzing Customer Shopping Behavior: You can have the best products around, perhaps even at the lowest cost, but little will matter if the user experience is poor throughout the purchasing process. Your websites overall design, functionality, and navigation are more important than your actual products are themselves. What good is an item if a potential customer is unable to complete a transaction or purchase it?

The Google Analytics Shopping Analysis report allows you to better understand how your customers engage with the content onsite related to initiating, abandoning, and completing transactions for products. Landing pages, Calls-To-Action, Internal Link Structure, and any other number of factors all play a vital role in directing, or funneling, traffic to specific pages where a conversion can be made. From the Initial contact to the Completion of a Sale, it is important to observe a buyers behavior. In this way you are able to spot possible problems or note potential opportunities that exist.

To further help understand your customers’ behaviors, you can check the Overview and Product Performance reports. These will provide insights into the data surrounding revenue generated and conversion rates for your products.

eCommerce overview Product performanceYou can investigate to see what the average order size is, the average order value, and refunds that had to be given back to customers. Perhaps one of the best metrics you should be watching is the rate at which users added a product to their shopping cart and actually completed a transaction after viewing specific product-detail pages. To the astute marketer, you will soon realize which are your most profitable and valuable pages that generate the greatest number of conversions.

Success Begets Success: When you are evaluating your sites online sales, you will want to study how products, product categories, as well as the overall process, from first encounter to final sale, are performing. Typically, the first things most people studying their analytics will notice is where they are succeeding. They will notice things such as increased revenue from the month or year prior for a specific product, an increase in the number of new or returning visitors who have completed a purchase, and any other number of factors. It’s always nice to first learn what you are doing right, because typically that warm fuzzy feeling diminishes as soon as you realize all the work that still needs to be done for things that are performing poorly.

But beyond making you feel good, there is another reason you want to find things you are successfully doing. If revenue has increased, a certain poorly performing product is doing better, or you notice an increase in traffic types that are converting, you want to dig into that information, understand it, and study it. This will help you to uncover whether there are any opportunities that exist where a prior successful process can be replicated.

Were the conversions received due to highly engaging content you created? Did a new paid ad you’ve created drive a dramatic amount of traffic to a landing page that ended up resulting in a sale? Did something receive a lot of attention on social media that resulted in increased converted traffic? Would a certain successful marketing campaign that was seasonally targeted the year prior be worthwhile to try again this year? Would that same campaign plan work for other seasons or for different products?

The number of questions you can ask yourself when studying your successes will lead you to better understand them, and more importantly, replicate them. Often times in marketing, and especially for SEO, success begets success. Do what you know already works, and improve upon that. It’s much easier and worthwhile than trying to create something successful from scratch. That however is not saying to avoid trying new things. Just don’t ignore what is already working, and more specifically, don’t neglect to attempt to make better the successes you’ve already achieved. Everything can always be improved upon 100% of the time, as nothing is, or can ever be, done to absolute perfection.

Product Performance and Placement

Under Product Listing in Google Analytics, you will be able to see how products perform based on their positioning in product lists on a webpage. There are many metrics that can be studied here, but initially you are shown which products are placed on what pages and the rank they are placed in specific lists on those pages. You will also be able to see how many impressions the product received, how many times a product was actually clicked on, and the associated click through rates.

Product placementStudying how your products performed based on their placement on pages and lists is important. You may notice that even though a product is continually at the bottom of every product list on every page, it has the highest click through rate. This may lead you to raise it higher in the list to attract even more clicks. Conversely, you may notice another item that is continually at the bottom of every product list which you wish would perform better. Placing that product higher in the list would help to draw more attention to it and subsequently clicks.

The important point to remember about where you place products in lists is that typically the human eye reads a webpage like they would a book, or a list. From top to bottom, left to right. So with that in mind, always place the products you deem most important in areas where a user will spot it before anything else. You not only have the power to guide where on your site a user visits and the path they take to get there, but you also are able to guide their eyes on individual pages and determine what it is they will be seeing first upon landing on a page. Keep this in mind when adding products to lists.

It’s said that the first spot in search results receives around 30% of all traffic, the second spot 15%, and so on. After the tenth spot, and on all subsequent pages thereon, traffic is reduced to a mere fraction of the total. Using this model, and your exact statistics can be studied when doing this in Google Analytics, we should be lead to believe that in a product list of 10 items, the first three slots will receive around 40% of the total traffic simply because they were placed higher.

This was of course is an example, and if you sell one hot item out of 1000, users will no doubt be looking specifically for that item so it’s placement wouldn’t matter and these results would be skewed (though its performance would certainly benefit from being placed higher in lists). As you’ll quickly note, product placement is critical. A website only has so much “Real Estate” or “Shelf Space” on a single page, so you have to make the most of the space you?re given.

Shopping and Checkout Behavior Analysis

The Shopping and Checkout Behavior reports provide an excellent way to interpret and understand the behavior of your customers and users. These will help you to measure the performance of your website and marketing efforts, understand why customers may not be purchasing certain products, and provide insights into the navigational path users go through during their checkout process.

Both the Shopping and Checkout Behavior reports watch four steps within the Customer/User cycle.

Shopping Behavior Analysis

  1. All Sessions compared to Shopping Activity
  2. Sessions with Product views compared to Cart Additions
  3. Sessions with Car Additions compared to Cart Abandonment
  4. Total Sessions with Transactions

Checkout Behavior Analysis

  1. Account Sign in and % of Drop-offs
  2. Shipping and Shipping % of Drop-offs
  3. Billing Address and % of Drop-offs
  4. Payment Info and %of Drop-offs
  5. Total Sessions with Transactions

First investigating this data will inform you to where attention needs to be paid. Both of these reports will quickly highlight at what point users dropped out of the purchasing cycle. It’s important to note whether a sale was not completed prior to a checkout, or during it. Once you find certain issues, you can investigate where exactly where those issues lay within the shopping or checkout process that leads to potential customers dropping off or abandoning cart.