CONTEXT
CONTEXT
Reordering groceries feels endless.
Reordering groceries feels endless.
Grocery shopping is already repetitive, and on Talabat, users usually search and browse for each item until their basket is full. Search algorithm is not as accurate and out-of-stock items are hidden, leaving users to wonder how else to get all of their groceries. To address this, I collaborated with a Data Scientist to build a personalized feature that nudges users to restock, showing their usual categories upfront to reduce repetition.
Grocery shopping is already repetitive, and on Talabat, users usually search and browse for each item until their basket is full. Search algorithm is not as accurate and out-of-stock items are hidden, leaving users to wonder how else to get all of their groceries. To address this, I collaborated with a Data Scientist to build a personalized feature that nudges users to restock, showing their usual categories upfront to reduce repetition.
TEAM
Product Designer (Me!)
Product Manager
8 Engineers
1 Data Scientist
Product Designer (Me!)
Product Manager
8 Engineers
1 Data Scientist
RESPONSIBILITIES
Secondary Research, User Research & Analysis, Interaction Design, Product Strategy
Secondary Research, User Research & Analysis, Interaction Design, Product Strategy
TOOLS
Figma, Dovetail, Miro
Figma, Dovetail, Miro
IMPACT
1.3%↑in returning users
1.3%↑in returning users
USER PROBLEM
USER PROBLEM
(Searching + browsing) X number of groceries + (out-of-stock) = frustration!
(Searching + browsing) X number of groceries + (out-of-stock) = frustration!
In interviews, users described grocery shopping as tedious and chore-like. To understand the scale of this pain point, I worked with the data team and found that Talabat users add an average of 12+ items from a single store, with occasional edge cases of adding items from another store due to out-of-stock items. I identified a couple of low-hanging fruits in the experience that could be improved.
In interviews, users described grocery shopping as tedious and chore-like. To understand the scale of this pain point, I worked with the data team and found that Talabat users add an average of 12+ items from a single store, with occasional edge cases of adding items from another store due to out-of-stock items. I identified a couple of low-hanging fruits in the experience that could be improved.
BUSINESS OPPORTUNITY
BUSINESS OPPORTUNITY
Weekly grocery shopper retention was below the 30% industry benchmark
Weekly grocery shopper retention was below the 30% industry benchmark
Instead of focusing on quick wins like showing out-of-stock items, I worked with the Product Manager to identify opportunities for a long-term reordering strategy, with the goal to reach the benchmark.
Instead of focusing on quick wins like showing out-of-stock items, I worked with the Product Manager to identify opportunities for a long-term reordering strategy, with the goal to reach the benchmark.
PROBLEM STATEMENT
HMW make it effortless for users to reorder groceries on Talabat so that the business can increase weekly shoppers' retention curve to the industry standard?
SOLUTION
SOLUTION
Usual categories at the right moment.
Usual categories at the right moment.
Users see their top five categories from their favorite store just before and after their usual reorder time, making it faster and easier to restock.
Users see their top five categories from their favorite store just before and after their usual reorder time, making it faster and easier to restock.
Shown only when it matters.
Using machine learning, Talabat learns each user’s unique reordering cycle and predicts their next order, surfacing the feature exactly two days before and after to reduce the clutter on the screen.
Usual items at a glance.
Usual items at a glance.
Previously bought items are shown at the top with a clear tag, helping users quickly restock before browsing for other products
Previously bought items are shown at the top with a clear tag, helping users quickly restock before browsing for other products
Shown only when it matters.
Using machine learning, Talabat learns each user’s unique reordering cycle and predicts their next order, surfacing the feature exactly two days before and after to reduce the clutter on the screen.
DESIGN CONTRAINTS
With competing priorities on the screen, space made it a design constraint
Leveraging on horizontal scroll
I designed the widget to be compact and used horizontal scrolling with clear visual affordances to maximize the available space.
Making content the main driver
I believed setting the right context through words in this widget was the most important.
I decided to surface categories instead of items based on the A/B test
Designing for a high-traffic screen meant I had to minimize risks. I based my decision to showcase categories based on an experiment that guided users to their usual refill categories, resulting in an 11.1% lift in category page conversions.
I believed microinteractions could elevate the repetitive experience
I designed smooth transitions, visual feedback, and playful animations in the user journey that made the shopping journey feel more engaging and less transactional.
SCALABLE DESIGN
Cross-vertical collaboration to scale retention across the company
I collaborated closely with the Food, Grocery, Marketplace, and Subscription teams to identify key use cases and scale them, driving increased retention across different user scenarios.
IMPACT
+5%
Grocery Adoption
+1.3%
Returning Users
TAKEAWAY
This project taught me the value of thinking big. By framing the vision beyond a one-off feature and pitching its potential for scalability, I learned that it not only inspires collaboration but also ensures we're building solutions that stick and last with the users.
OPPORTUNITY
HMW make it effortless for users to reorder groceries on Talabat so that the business can increase weekly shoppers' retention curve to the industry standard?
Copyright © 2025 Lucy Ji Soo Choi
I decided to surface categories instead of items based on the A/B test
Designing for a high-traffic screen meant I had to minimize risks. I based my decision to showcase categories based on an experiment that guided users to their usual refill categories, resulting in an 11.1% lift in category page conversions.
Leveraging on horizontal scroll
I designed the widget to be compact and used horizontal scrolling with clear visual affordances to maximize the available space.
Making content the main driver
I believed setting the right context through words in this widget was more important than the number of categories shown.
DESIGN RATIONALE
A crowded homepage with competing priorities made space
a design constraint
Cross-vertical collaboration to scale retention across the company
I collaborated closely with the Food, Grocery, Marketplace, and Subscription teams to identify key use cases and scale them, driving increased retention across different user scenarios.
SCALABLE DESIGN
IMPACT
+5%
Grocery Adoption
+1.3%
Returning Users
I believed microinteractions could elevate the experience
I designed smooth transitions, visual feedback, and playful animations that made the shopping journey feel more engaging and less transactional.
DESIGN RATIONALE
Leveraging on horizontal scroll
I designed the widget to be compact and used horizontal scrolling with clear visual affordances to maximize the available space.
Making content the main driver
I believed setting the right context through words in this widget was more important than the number of categories shown.
A crowded homepage with competing priorities made space a design constraint
Cross-vertical collaboration to scale retention across the company
I collaborated closely with the Food, Grocery, Marketplace, and Subscription teams to identify key use cases and scale them, driving increased retention across different user scenarios.
SCALABLE DESIGN
IMPACT
+5%
Grocery Adoption
+1.3%
Returning Users
I believed microinteractions could elevate the experience
I designed smooth transitions, visual feedback, and playful animations that made the shopping journey feel more engaging and less transactional.