Mixed-methods research and conversion rate optimization for a custom men’s shirting e-commerce client to uncover motivations for drop-off in the design and purchase funnel
Original Stitch | E-Commerce | B2C | Mixed Methods Research & CRO | 2016
I led a 1-month mixed-methods user research project for a custom men’s shirtbuilder e-commerce client as a Product Designer & User Researcher at ROIWorks, a growth marketing & design agency. I subsequently led a 3-month redesign to optimize their conversion funnel.
Challenge
Original Stitch was a men’s custom dress shirt builder that allowed for customization of fabric, style, and size. They were founded in 2012 and ROIWorks worked with them from 2016-2017.
ROIWorks audited the Original Stitch design and purchase funnel to identify opportunity areas. I used Google Analytics and HotJar video recordings to identify points of dropoff and then conducted qualitative interviews to understand customer motivations for dropping off.
Outcome
Based off the research findings, I worked with our team to implement a series of A/B tests to test our hypotheses, increasing conversion. Subsequently, I designed a new customization and purchase funnel. Since our engagement ended before the final implementation, I do not have metrics associated with the final outcome.
Original Design
Revised
PROCESS
Below, I describe the step-by-step process used to execute and analyze the research.
01 GOALS
02 PROTOCOL
03 RECRUITMENT & INTERVIEWS
03 ANALYSIS
04 DATA COLLECTION
05 FINDINGS
06 OPPORTUNITIES
07 SEEING INTO THE FUTURE
During our initial audit, we found the highest drop off in the funnel was on the cart page across desktop and mobile.
Cart dropoff isn’t an atypical behavior for an e-commerce site. We all love to shop until it’s time to pay. To understand the customer’s exit behavior, I used Goal Flow in Google Analytics to understand behavior across desktop and mobile platforms and uncovered the following:
*screenshot of client’s design & purchase funnel
The highest percentage of customers who drop off were returning to the shirt builder after visiting their cart. The 2nd highest percentage of customers were returning to the sizing step.
To dig deeper into user behavior, I analyzed HotJar video recordings of customers on the drop-off points to identify patterns in user behavior.
After analyzing video recordings, the following patterns emerged.
Customers were revisiting the sizing page to see the outcome of entering their measurements. The user flow bypassed the outcome of a result and took the customer directly to the cart page.
Customers realized they bypassed customizing the shirt style once they reached the cart.
The primary CTA led to the sizing step rather than the next step in styling the shirt. The secondary CTA led to the style customization.
After conducting 5 in-person usability tests with the web app on desktop and mobile, I validated the same patterns I observed in Google Analytics and HotJar.
Participants unintentionally bypassed the customization flow.
3 out of 4 participants clicked the ‘Enter Sizing’ button after selecting their shirt fabric versus clicking the ‘Next’ button.
