Reasons of user drop-off sales funnel on e-commerce
This research project was focused on the challenges and complexities for users when exploring the product detail page. Through a rigorous process purely quantitative, I identified the key issues and the main reasons why users abandon the purchase process (sales funnel).
Client:
Private
Role:
UX Researcher, Data Analyst
Date:
Sep - Dec 2024
Industry
Travel Arrangements
Project overview
Product Detail Page drop-off
Large proportion of users who land on the product detail page doesn't complete a purchase, and also all of the are coming from different acquisition channels (e.g. organic traffic, SEO, paid media, etc.). It is necessary to find the main issues and fix it to improve the conversion rate.
Image: sales funnel illustration
My role
UX Researcher and Data Analyst
I worked mostly individually on this project, with the support of the store product manager and growth manager during the analysis process, also from a data analyst co-worker that helped me to shape more detailed data.
Project Context
Under what context did the project started?
This page presents a lot of issues we identify during the last UX audit, also several users mentioned confusion around specific information shared there, which were seen as critical for their purchase decision.
Client's Context
What was the company's situation?
This research was a strategic decision based on the findings of previous research, but which did not fully find the reasons why users decide to leave the page and not buy the product, but some relevant problems had been identified from which many hypotheses arose to investigate and also to test with growth.
Project Impact
The results will guide to the prioritization of improvements and adjustments especially focused on features that generate the largest impact on sales and revenue.
Timeline
Planning
Recruitment
Execution
Analysis
Reporting
Research Planning
Hypothesis
If we identify the user problems or reasons for non-purchase on the PDP, we will be able to define actionable steps to increase the number of sales by addressing user issues more efficiently.
This hypothesis is based on: ee have identified that a large proportion of users who land on the product detail page and abandon the purchase process. We believe this presents an opportunity to understand the reasons for non-purchase and define actionable steps to better meet user needs.
Challenges and Goals
Have an holistic approach about what is happening with the users that are landing in the page and don’t complete a purchase
Find the ways to streamline the user flow to reduce friction and increase conversion rates
Have information to decide what to implement to support users in their decision-making process
Success Criteria
This research must provide a greater understanding of the specific reasons why users drop out of the PDP, as well as identify the frictions and barriers they face.
Discover the users’ behavior and needs when they arrive at the page, as well as any usability issues that may be present.
Differentiate between different type of user behaviors.
Discover the user perceptions of the value proposition, their trust in the product, and the expectations they have when they reach the product detail page.
Image source: survey planning with segmentation of different user traffic
Methodology
Research Execution
Strategy
Validation with Quantitative Data: in this case focusing on quantitative data would be more beneficial due to the richness of information this approach can provide.
Identification of Problems and Reasons for Abandonment: gain a deeper understanding of the specific reasons why users abandon the page, as well as to identify the frictions and barriers they encounter.
User Behavior: Analyze behavior patterns and any usability issues that may be present. This analysis will be supported by a review of recordings, heat maps, clicks and statistical data.
Segmentation and Differentiation: I also considered segmenting users by platform (web, mobile, iOS, Android) and market by country (e.g., United States, Europe, Asia) to better understand behavioral and needs-based differences by segment. Additionally, segment users by origin and destination, due to perceived differences in pricing and trust based on destination, and differentiate between returning and new users to gain a better understanding of their behaviors.
User Expectations and Perceptions: Assess user perceptions of the value proposition, their confidence in the product, and their expectations when they land on the page.
Image representative illustration
Sample, segmentation and it outcome
Desired outcome
Analysis and synthesis process
Segments, variables and trends
To conduct the analysis, it was necessary to cross-reference data to identify trends to highlight similarities and differences.
First, I ensure the data accuracy by cleaning the dataset and standardizing formats, after I checked the data an charts to define the sample of the different segments, and I did and descriptive analysis I calculated averages, medians, and percentages for each segment.
After, I searched for differences and trends between segments and identify relationships between variables, also identify outliers or anomalies.
From the beginning to the end, I utilized data visualization to observe and highlight those findings and support the insights.
Image source: new user segmentation chart
Project Deliverables
New user segmentation
I identified s a diverse range of behaviors and different obstacles users are facing. I suggested to consider this insights to refine targeting and provide better decision making support to users based on their needs.
Market Insights
I was able to identify different user insights depending on the different segmentations I defined and new ones I identified.
Image source: likert scale chart report
Sharing and activation.
My storytelling methodology:
The insights were clustered and summarized using my own data storytelling framework, using the data to emphasize the message, adding directly actionable to every insight.
It was necessary to run 2 report meetings of 45 minutes each one, due to the largest amount of information and big amount of charts. This comprehensive approach allowed the research team to assess user perceptions, behaviors, and challenges related to the experience in the e-commerce.
Also, I ran an brainstorming workshop to redesign solutions with all the teams involved.
Next steps and Recommendations
Improve the cognitive load and communication to reduce time of decision-making process
Improve the communication from specific identified topics to reduce user confusion and abandonment rates, prioritize the easy and effective readability and navigation
Opportunity of tailoring the message by language or culture for improve conversion rates, the segments and markets identified are trying to solve different problems
Lead workshops to brainstorm new solutions ideas with all the teams involved: design, research, product, marketing, SEO and support, to test those new solutions.
Image source: report insight
Project Impact
With this results I helped to enhance the company's data-driven design and optimization by addressing the following improvements:
Increase the conversion rates by: reducing the friction on the navigation and increase the checkout completion.
Enhance the User Experience by: usability improvements and addressing user needs.
Increase Revenue and Profitability by: maximize ROI on traffic and generate higher average order value (AOV).
Reduce Cart Abandonment by: addressing abandonment reasons and implementing engagement strategies.
Reflections and learnings
It's important to note that users' behavior and expectations can be significantly shaped by the acquisition channel or previous step in the customer value journey (e.g., Ad, Google Result, Affiliate campaign, etc.), so it is important to consider it in the segmentation and would be a next point to consider on the iteration of this research and also the customer value journey, anyways it’s challenging to segment users by acquisition channel from ux research. For future surveys, we could include questions like 'How did you find out about us?' or 'Where did you discover us?' to get an approximation of the acquisition channel. While this wouldn’t be completely accurate, it could provide some valuable context.
