A/B Testing Reports
Project Objective
At The Home Depot, I was involved in A/B testing initiatives aimed at optimizing user experience and increasing conversion rates. These tests focused on content changes, form layouts, or pricing changes, with the goal of identifying user behavior patterns and making data-driven improvements.
Role: Analyst, responsible for gathering, analyzing, and presenting user behavior data.
The Strategy
The strategy centered around developing test hypotheses based on user behavior insights and business goals. Working closely with the content, research, and product teams, I helped run A/B tests, ensuring that we tested only one variable at a time to isolate effects. After gathering data, I analyzed the performance metrics and provided recommendations for future optimizations.
A/B Testing Flowchart (Click on the image to view Miro board)
Design Solution
Each test compared two or three variations of the same webpage. For instance, we tested different versions of a services page layout to determine which resulted in a higher completion rate. The team used Optimize and Mixpanel to track user engagement and identify key trends. From there, I synthesized the results into actionable insights that supported management team with design and content decisions.
To help illustrate the format and structure, I’ve included a slide with an example of one of our A/B test results and findings layout:
A/B Test Findings Slide
The Technology
Microsoft Office, Google Optimize, Mixpanel
Some of the key technology that was leveraged were:
A/B Testing Tool: Google Optimize was used to set up and monitor the tests.
Data Visualization Tools: Mixpanel and Excel were used to visualize performance metrics.
Project Management Tools: Jira was used to track test progress and outcomes.
Final Thoughts
The A/B tests provided insights that led to significant improvements in conversion rates and user engagement. While specific results remain confidential, these projects helped refine the company’s approach to user experience and data-driven design decisions.