AB Analytics : Web Analytics and Optimisation

Far from average ecommerce conversion rate analysis - part three

Purchase Funnel & Conversion Rate Analysis

Like this article?  a) tell us  b) share it c) download it and d) keep updated

online ecommerce shopper

The final instalment of average e-commerce conversion rate series focuses website improvement through visitor navigation and shopping cart abandonment analysis.  

There are so many excellent techniques available to monitor and understand online shopping behaviour that this post developed into a "war and peace" novel length at one point but has been stripped back to a brief observation into funnel page analysis.

In line with the e-commerce theme, our focal point for discussion is the all-important purchase flow....


Purchase Funnel - Page Level Analysis

For a traditional e-commerce site, arguably the most important site area is the purchase flow. Identifying friction points in the checkout area can be achieved by analysing traffic volume between key steps or pages in a conversion process.  Web analytics tools offer a variety of default reports to display visitor navigation and funnel abandonment. However, reports of this nature make performance comparison over time difficult and potentially hide the best opportunity for performance step change. In fact, after reviewing the funnel reports it is not uncommon for people to still ask, "So where shall we focus our attention first?"

To derive insight, at least three shortfalls in visualisation approach need to be addressed:

1) Optimisation action priority indicator

2) Historic performance comparison

3) Audience segmentation

How to identify priorities -  Create a weighting for each funnel step. This can be done in Excel using the RANK() function (in ascending order) to generate an index value for the number of visitors who abandoned at that stage and then repeat this rank formula for the drop-off step percentage. Multiply the two ranks together and the step with the highest score is the starting point. The table below shows that although step 8 'verify payments' has the highest drop-off rate (80%), the number of lost visitors is only eight (8). On the other hand, 'check-out delivery address' page lost seventeen (17) visitors and has a drop-off rate of near half (49%) making it prime real estate for a conversion optimisation project starting location.



How to compare performance over times - Pull historic performance data to obtain context and combine with the priority action list. In the example below, SPARKLINES have been generated to plot the last 13 weeks of data and an Excel function called, REPT() is used to visualise traffic volume in the funnel column. 

Reporting at an aggregate site level can hide underlying issues specific to a particular audience (e.g. Internet Explorer 6, etc), so be sure to leverage segments or advanced filters to understand audience nuances. And finally, add some financial information (where available) to demonstrate bottom line impact and secure budget for conversion optimisation work. After all money talks... There are plenty more ways to spice up funnels but hopefully, this basic overview will generate ideas to evolve reporting in your business too.

Click here to download the Purchase Funnel excel report  




Pros and Cons of Purchase Funnel Analysis

One of the drawbacks using page level funnels is that visitors rarely progress through the site in a linear fashion and with a complex site, building separate reports for all the possible user journey permutations is unmanageable.  Using data filters or segments can help reduce duplication.

Web and App design technology that allows for dynamic page content (e.g. Ajax) also mean visitor interactions that occur in-page are not taken into account with simple page-to-page drop-off making friction detection more difficult.  To accommodate analysis at page interaction level, use Event or Form tracking.

The purchase funnel example above has benefits over-and-above the default reports found in most web analytics tools but, introduces a manual element to updating reports which can be resource intensive.  Where possible, minimise this effort by using tools to automate the data refresh process.  Users of Google Analytics benefit from a wide range of Excel plugins like Analytics Canvas, Excellent Analytics and Supermetrics.  Similarly, Adobe Analytics users have the ability to leverage Excel Client and Report Builder to automate reports.  There are also APIs for most tools to extract this data and store it for visualisation.

Page level funnel reports do help pin-point user experience friction locations which is an important step in the conversion optimisation process.  However, at every stage in the purchase flow customers generate questions and demand reassurances.  The data visualisation inform design decisions but, a business must cover the basics first.

  • Steps 1 Home >  2 Product Details
Q. What is our brand value proposition and why should a visitors be interested in our products? 

For home or product landing pages, use spilt testing to quickly determine which brand value proposition and page design layout work best. 


  • Step 2 Product Details > 3 Basket Add
Q. What type of benefits lead incentives can be offered to stimulate basket additions and is the price-point compelling enough? 

Ensure 'add-to-basket' buttons are prominently featured above the fold and test compelling call-to-action messages. Check product availability too in case of low inventory levels. 

  • Step 3 Basket Add > 4 Register
Q. How much information do visitors really need to provide in order to purchase?  

Determine what information is essential to know about the customer in order to ship the product vs non-essential demographic and marketing based data.  Don't forget to clearly mark fields that are mandatory and consider dropping optional ones.  If you're not already using a 'Guest' checkout option, try it for speedy shoppers and balance the amount of information requested during registration with that which can be collected later in the lifecycle.  

  • Step 4 Register > 5 Billing or 6 Delivery Address
Q. Are the Billing and Delivery address form pages designed for easy completion?

Postcode lookup widgets that semi-automate form completion are a frequent pain point because the rules engine may not be configured for free text data entry. For UK postcodes allow visitors to enter incorrect case and/or spaces (e.g. W1 5NT, W15NT, W1 5nt) and provide international address information should shipping allow (e.g. US zip code numbers 90210). 

Allow for different billing and 'delivery address' options and simplify form completion by placing a simple check-box on the 'billing address' form that copies the information provided to the 'delivery address'.  Experiment with the choice of shipping options and cost. 

  • Step 7 Payment Details > 8 Verification
Q. Can completion rates be increased by asking for less information on payment forms?

Add a hidden lookup / function to the 'payment details' form and determine credit type from the details entered into card number field. This helps visitors by removing the 'credit card type' field and speeds completion.  Implementing the code depends on the server-side language used on site and here is a JavaScript example by Northstar.  A list of major credit card provider identification types can also be found on Wikipedia here.  

  • Step 8 Verification > 9 Confirmation
Q. Do visitors need to be transferred to another site for credit card verification?  

Build a verification iframe, or widget that is hosted on your own site domain and ensure that a secure protocol (https:) is provided for all pages where personal information is entered as with previous steps in the checkout purchase flow.  This method provides a consistent experience and additional tracking continuity for visitors progressing through to order confirmation 'thank-you' page(s) as seen in the eBuyer example below.

ebuyer verify by visa example
 
  • Step 9 Confirmation
Q.  Does the user journey stop when visitors complete a purchase?

Drive additional sales by displaying cross-sell and up-sell content on the "thank-you" confirmation page. Include a customer benefit (e.g. voucher or discount on next purchase) to incentivise sharing the product(s) purchased with others.  Deliver purchase confirmation emails quickly and be sure to include all details that help customers follow up in the event of any problems with the call centre or support.  These sound obvious right, but some e-commerce sites still don't do this well (if at all). 

Happy Optimising!

What other techniques are you using to improve site conversion? 

We would love to hear your opinion in the comments section.  


About the Author

Alex Brown is a Digital Analytics and Site Optimisation expert who works as an independent freelance consultant.  The opinions shared in this blog post are based on personal experiences gathered over a decade of data crunching and technology evaluation.  The author makes no attempt to be grammatically, politically (or otherwise) correct.  Spelling was never a strong point and for practical reasons, not all vendors in the market are referenced in the article - no hard feelings.

Related Products & Services

With in-depth knowledge of both the market place and major technology vendors, allow us to help select, negotiate, support and manage web analytics implementation for your business.  For a confidential and informal discussion to see how our solutions can improve your business today contact us.

Resources
(In Alphabetical Order)
AB Analytics Purchase Funnel Report Example
Adobe AnalyticsExcel Client
Adobe AnalyticsReport Builder
Analytics CanvasAnalytics Data Extraction
Excellent AnalyticsAnalytics Data Extraction
North Star Web DesignCredit Card Detection
SupermetricsAnalytics Data Extraction
WikipediaCredit Card Identification

Like this article?  a) tell us  b) share it c) download it and d) keep updated

Online e-commerce shopper image courtesy of Marin / Freedigitalphotos.net

Comments: 1

  • Robbert Jun 27

    thanks for the useful insights! What I am trying to find is some benchmark data that indeed shows the conversion drops for each step of the check out process - specifically the billing/delivery address part. You show some data in your article - is this real data? and do you have any indication on what a typical conversion drop would be or where I can find this information? Thanks in advance! Robbert

Add a comment

Email again: