Revamping the Filters and Search in Jumia's Catalog

Context and Challenge

After receiving recurring feedback from the customer success team regarding difficulties in using filters and searching the catalog, we identified the need to deeply investigate the issues faced by users. The primary objective of the project was to understand user pain points and propose improvements to optimize this experience. In this project, I was the UX/UI designer responsible for the entire process from start to finish. One of the highlights was aligning and collaborating with other teams, such as marketing, engineering, product, customer success, and content.

Discovery Objectives

To perform a more accurate analysis, we initiated a discovery process aimed at:

Evaluating the usability of the current screens.
Analyzing competitor solutions.
Generating data for both the current development and future decisions.
Avoiding common mistakes in design decisions.

Heuristic Evaluation

The first step was the evaluation. We conducted a heuristic evaluation based on Nielsen’s usability guidelines to identify critical issues. The main findings were:

Inconsistent Colors: Lack of standardization in color usage.
Unusual Behavior When Selecting Categories: The interaction does not follow a logical flow.
Inconsistent Terminology: Different terms are used for the same concept ("Clear All" vs. "Reset").

Issues with Information Input: Sellers register the information displayed in filters, leading to inconsistencies and making filtering difficult, because we had no guidelines to help them

Quantitative Survey

Now that we fould more about the usability, it was the moment to understand our customers perceptions, we conducted a quantitative survey to uncover:

Whether users actually use the filters;
If the filters assist in the search;
What information is most important to find a product;
Whether users find the current filters helpful.

We got 37 responses and had many finding, but some of them stand out:

Findings: Our customer

Based on the survey responses, our typical user:

Is between 18 and 30 years old.
Has been using Jumia for over 3 years.
Prefers shopping through the app.
Typically buys clothes/shoes, electronics, and phone accessories.
Accesses the marketplace with a clear idea of what they want to buy and goes straight to the search function.

Findings: Filters

The collected data indicated that filters are not considered very useful, with 43% of users rating their experience 3 or below. The most important information to assist in purchasing was:

Price
Express delivery
Seller location
Product rating
Brand
Category

Problem Statement

After a few alignments with the product team and based on the insights gathered, we formulated the following problem statement: "I am a Jumia customer trying to make a purchase but I can’t find a specific product because I don’t feel that the filters are helpful."

Exploration of the Causes

With the results in hand, we dove deeper to better understand the key issues:

Evaluate customers’ perceptions when using the search feature.
Identify UX and UI improvement opportunities.
Understand what information is missing.
Discover what customers expect from the filters.

Results

Preference for Contextual Filters: Customers prefer filters relevant to the current search (e.g., "Size" for fashion, but irrelevant for electronics).
Jumia Express is Valued: The option to filter by Jumia Express is important but not easily visible.
Percentage Discounts Not Considered Useful: Customers prefer to see only discounted items instead of percentage information.
Overall Perception: Many customers feel overwhelmed by too much information in the filters, but consider much of it irrelevant.

Co-creation

During my trip to Egypt, we conducted several workshops aimed at generating topics for the upcoming roadmap. One of these workshops was instrumental in creating the initial wireframes for the new filters. This particular workshop focused on developing a problem statement and working on a solution. Many of the proposed solutions highlighted challenges related to search functionality and filters, emphasizing the need to provide more space for product listings while prioritizing entry points for the most commonly used filters. The workshop participants included two product owners and six software engineers.

Solution

Thinking about all the insights, we finally had a solution: To enhance the filtering experience, we introduced context-sensitive filters that automatically adjusted based on the product category being searched, ensuring relevance and usability. The Jumia Express filter, which was crucial but previously hard to find, was given more prominence by placing it higher in the options and making it a quick access feature. Additionally, we replaced the percentage-based discount filter with a more intuitive "Discounted Items" option. Lastly, the filter layout was optimized, prioritizing frequently used filters for easier access while deprioritizing less relevant options.

Apps (iOS and Android)

Previous version

New

See prototype

Desktop

Usability Test

Once the concepts were finalized, it was time to validate if we had the best solution. We conducted moderated tests with 5 participants. The mais objectives:

Evaluate users’ general perceptions of the new filters concept;
Check if the approach with the “All Filters” page open is clear and helpful;
Test the effectiveness of the “Suggested Filters” on the top of the catalog.

Results

Highlighting Jumia Express: 3 out of 5 users noticed the emphasis on Jumia Express. Feedback: “I usually choose Jumia Express, the faster the better. That’s how I see it in other apps.”
Approval of the Filter Structure: 5 out of 5 users understood the structure and the meaning of each filter. Feedback: “I feel like I have everything I need, and it’s kind of guiding me.”
Satisfaction with the Suggested Filters: 4 out of 5 users praised the “suggested” filters visible at the top. Feedback: “I liked the suggestions at the top; it’s more direct and has the main filters I usually use.”

Conclusion and Next Steps

This study had an influence on other teams. Through our research, we uncovered frustrations related to the inefficiency of our search functionality, which led our backend team to start working on ways to improve search results.

Meanwhile, the content team, responsible for assisting sellers in listing their products, created guidelines to prevent inconsistent information in the filter options.

The feature is already on live, but the work is not done, our main objective today is track the success, to understand if the new filters are trully helpful and if they are actually being used. Points that we can validate using google analytics.