How To Measure Conversion Rates for E-Commerce Sites [Google Analytics Tutorial – With Pictures!]

Jan 26, 2014 | How to

E-Commerce sites are all about conversion. It’s all about taking a customer from a certain step in the decision-making process, and moving them through a set of steps until they purchase. Increasing revenue can be as easy as improving or eliminating some of that process. So how do you measure conversion rates on E-Commerce sites? Often, as an online marketing agency, we are focused on “converting” people from somewhere else online (google search, social media, other websites) to our clients’ sites. But there is also immense value gained from optimizing the conversion of traffic on your site through the checkout process. We thought we’d share this valuable process for analyzing your on-site conversion rates to determine where on your site you are losing the most customers.

How many people put something in their cart but didn’t check out? How many people registered for accounts but never came back? How many customers do you lose at each stage of their checkout process? These are the types of questions E-commerce companies need to know to make informed decisions about marketing and website design/UX. We recently analyzed one of our clients’ websites, and we wanted to show you how to create custom reports and custom segments in Google Analytics so you can get more actionable data from large, complicated E-commerce websites.

These are the basic steps we outline below:

  1. Setting up a custom report within Google Analytics
  2. Preparing for your advanced custom segments reports
  3. Creating your advanced custom segments

Step 1: Setting Up Custom Reports Within Google Analytics

First, login to Google Analytics and click on the customization tab in the top navigation bar.

Customized Reports in Google Analytics

Then, create a custom report. Depending on what you want to measure, you will choose different “metrics” and “dimensions.” Since we are looking for data on how unique visitors navigate our client’s site, we chose Unique Visitors as our Metric and Page as our dimension.

Google Analytics Custom Report

After creating this custom report, what you should see is a page-by-page detail of your site’s unique visitors. Now this information isn’t particularly useful – you can see unique visitor data in your normal reporting view. However this gives you a bit of a “sandbox” where you can distill the vast amount of information down to only the metrics in which you are interested.

Basic Custom Report - Google Analytics

This basic custom report in Google Analytics shows us how many unique visitors came to each page in our selected date range.

Step 2: Preparing for Custom Segments

Along those lines, the next step is to create custom segments for the discrete steps in your site’s conversion funnel. Here’s what you need to get started:

1. A standard URL taxonomy. For example our client uses index/[page] for shopping pages, index/cart for cart pages, cart/signup for registration pages, etc. (see below) The pages need to be unique or you won’t be able to separate your customers in different stages of purchase intent.

Example of a client's URL Taxonomy

Example of a client's URL Taxonomy

Example of a client's URL Taxonomy

2. The URLs roots/directories/keywords of the steps in your customer’s shopping process. This is what you will use to create an “Advanced Sequence” in your custom Segments. This also comes directly out of step #1 above.

3. A workflow of your most common customer paths. This helps you decide which steps are most important to measure. Here is what our customer flows look like:

Customer Workflow

Here we laid out the process that a customer goes through

Step 3: Setting Up Your Custom Segments


Now you are ready to set-up customer segments! First click on the little button above the “overview” tab (circled in red below).

Getting started with customer segments

That will bring up a page that looks like the below screenshot. Click “Create New Segment” (circled).

Custom Segments in Google Analytics


That will bring up a box that looks like this:

Segments within Custom Reports

This is what you should see when you create a new segment

First add a second step by clicking “Add Step.” Then Click on “Ad Content” and change it to select “Page.” Then choose the pages that you want to measure traffic between. For example, if you want to measure unique visitors between cart/index and cart/shopping, add those pages as steps 1 and 2, respectively.

Selecting Advanced Sequences

Select the page taxonomy and path you want to examine (e.g. Home Page to Cart)

Very important: Make sure to choose “Immediately followed by” – otherwise you are just looking at unique visitors that hit 2 pages at any point in their experience.

Google Analytics Custom Segment within custom report

Make sure to choose “immediately followed by”

And voila! After you add the steps you are trying to measure, you should end up with something that looks like this:

Custom Segments and Reports in Google Analytics

Within the custom report dashboard, we have identified custom segments for each step of the website conversion funnel.

This shows us very clearly where our client’s audiences are dropping out of their purchase process. The important part of the above graphic is the lower half of the page, where it shows the % of total traffic that has visited each page. For example we can see that almost 8,000 (42%) visitors dropped off between their cart and their login pages. Also it seems that another 35% of visitors dropped off between the login page and the payment page. These vital insights tell us where our customers are leaving, even after they’ve begun the process of buying something.

Now we can get to work on making changes to the site to increase the conversion rate from page-to-page, or better yet, even eliminate certain steps altogether.

[If you are interested in reading more about on-site analytics for E-commerce companies, check out our friends at Growth Spark – they know all about conversion-optimized design and analytics]


Submit a Comment

Your email address will not be published. Required fields are marked *