Revamping Feature Engagement through User Research
Iterative improvement is essential in an agile environment to maintain progress and confidently determine which hypotheses positively impact product metrics. Even smaller scopes need a careful UX design process. In this case study, I show how I focused on evolving an underutilized feature during the consideration phase of the customer journey to improve engagement metrics and positively affect the conversion funnel.
Youcom is a brand specializing in fashion and youth lifestyle. The focus of the Product Tribe is to evolve our digital channels to ensure a delightful omnichannel experience across e-commerce, the app, and physical stores. The squad I worked in is dedicated to the end-to-end online journey, specifically for desktop and mobile.
Work Context
Role: Product Designer
Team: Product Designer, Product Owner, Developers and QAs
Duration: 4 months
Period: End of January/2024 to end of May/2024
01
Overview
Problem Statement
Data collected by our feedback capture tool highlighted that clarity about sizes was one of the major reasons cited by clients for hesitancy during their navigation journey. This issue directly impacts conversion rates and is also the most frequently reported reason for exchanges and returns, as identified in our Customer Sucess data.
Although sizing is multi-faceted, making it an epic in our roadmap, we hypothesized that improving the use of the virtual fitting room feature would positively address the identified problems. The Product Manager prioritized this feature in alignment with the OKRs to be worked on for the quarter.
Goals
Increase usage of the feature
Increase conversion rates
Reduce the number of exchanges and returns related to sizing problems
02
Discovery Phase
Firstly, I leveraged usage data of the virtual sizing and size chart features to establish a baseline and determine if the planned improvements would positively impact the metrics. I began with desk research on sizing issues and associated technologies, along with benchmarking similar tools to grasp market possibilities. This provided a broader view of the problem to be solved.
Hypotheses
Clients didn't acknowledge the button's existence due to delays in display.
The button position on the desktop version had readability issues, remaining unnoticed.
The naming of the feature, "virtual fitting room," might not be clear to all customers.
An in-depth analysis of our third-party tool for the size guide, including the size chart and virtual fitting room, revealed a problem related to delays in showing the buttons during navigation on the product page details. This led to the first hypothesis that some clients didn't acknowledge its existence, hence not using it.
Regarding the feature's visualization, the second hypothesis was that the button positions on the desktop version had readability issues, remaining unnoticed by users even when displayed.
The third hypothesis was that the naming of the feature, "virtual fitting room," might not be clear to all customers, who do not feel encouraged to try it.
03
Solution Exploration
To explore the first issue, the squad leads (me, the Product Owner, and the Dev Lead) organized a meeting with the third-party team that provides the solution to discuss technical aspects. We decided to change how the button is displayed to optimize and speed up the load time. This task was assigned to the development team as it required code adjustments.
To iterate on the second hypothesis, I designed alternatives for positioning the buttons along with their appearance. Visuals were validated with the design team through a design review, and feasibility was checked with the Product Owner and a developer.
To address the third hypothesis, I brainstormed names based on benchmarking and our communication guide and validated them with the marketing team. Then, I elaborate a quick survey to collect data about sizing within our brand. I included an open question about how customers ensure they choose the right size and single-selection alternatives for naming the button and results section. This survey provided significant insights that were leveraged not only in the evolution of this initiative but also fed other ongoing research and activities of the tribe, reinforcing the culture of shared knowledge.
One-page report shared with the Tribe, featuring the hypothesis and actions, research methodology, survey results, user comments, frequency of themes, and mapped opportunities.
04
Testing
The survey results showed a draw between the current name and one of the proposed names, so we decided to A/B test the alternatives to attain confidence in the changes. For the desktop version, we conducted an A/B/C test, considering the need to experiment with different positions.
From this phase, the task followed the development pipeline, adhering to the downstream process and backlog.
Version B in Mobile showing before and after size consulting
Version A (top) represents the current solution. Version B displayed buttons in new position but with the same naming, and Version C (bottom) featured both new positioning and new naming.
05
Results & Impact
After 10 days of testing, version B (the new naming) proved to be the best option for mobile access, achieving a 31% conversion rate for users interacting with the feature. However, the results from the desktop version's A/B/C testing remained inconclusive. We were also aware that the initial results might be significantly influenced by the improved performance of loading the feature’s buttons. Therefore, we extended the test period to collect more usage data and establish accurate and stable results.
We ended the test after 50 days. The new desktop button position proved to be the best approach. Although the results regarding the feature's naming didn’t show much difference, they were slightly better for the new name. Combined with the mobile results, representing over 80% of our user base access, we decided to implement the new name for the entire base, as it was proven to increase interactions, contributing to higher conversion rates. The project's duration considers the kick-off meeting, research phase, iterative validations, and final implementation.
We successfully met the first two goals by testing the solution with a reduced customer base and expect further improvement with its extension to the entire base. The next step is to keep tracking data to understand how it will affect the after-sales stage, as it requires a longer period to observe the full impact of this action. Continuous monitoring is essential to evaluate if the improvements lead to sustained reductions in exchanges and returns and to ensure the enhancements positively influence overall customer satisfaction.
06
Takeaways
In the first alignment meeting, the classic suggestion was brought up: “make the button more prominent!”.
The product page detail is a crucial moment in the user journey where they make purchase decisions based on useful information. I didn’t want to add another element competing for users' attention. Nevertheless, I kept the suggestion as one possibility for design evolution but decided to investigate the problem further before drawing a solution. I believe that we, as designers, play a pivotal role in product strategy. Even though we are experts in usability and user experience, we also have to manage stakeholders’ expectations and make them know that their ideas are being considered. After all, they have knowledge that we need to complete the whole picture. In the end, the most efficient solution will be proven through testing by our users and positively moving metrics.