01
She just needs a few things.
After a long day, she wants to do her grocery shopping.
Increasing Purchase Completion Rate On Bazargam
An end-to-end UX redesign focused on uncovering and resolving usability issues across Bazargam's shopping experience, from session analysis and usability testing to redesigned user flows and UI.
Methods:
Cognitive Walkthrough
Session Record Observation
Usability Testing
UI Redesign
MY ROLE
Product Designer / 6-person engineering team (3 backend and 3 frontend developers)
I was the sole designer on the project, owning the end-to-end product redesign from research and testing to UX flows, and high-fidelity UI. Product requirements came from engineering leadership, company stakeholders, and research findings I identified throughout the process.
Designer
Tech Lead
Developers
CONTEXT
The intent was simple; but the journey wasn't.
Bazargam is an FMCG marketplace where small UX frictions would interrupt the purchase journey. From choosing a store to navigating back, finding available products, resolving cart conflicts and finally reaching payment, the user had to repeatedly figure out what the product expected from her.
01
She needs a few things.
After a long day, she wants to do her grocery shopping.
02
She finds what she needs.
But adding it to the cart isn’t quite that simple.
03
She picks a store.
She needs to choose where she wants to buy from.
04
She goes back.
But navigating back doesn't always take her where she expects.
05
She checks availability.
Some products aren't available in the selected store.
06
She resolves the conflict.
The cart needs to be adjusted before she can continue.
07
She gets ready to pay.
After all the checks, she finally reaches checkout.
08
And finally... payment.
After all the clicks, check and returns, she makes it.
COGNITIVE WALKTHROUGH
We started by walking through the core journey from the user's perspective.
Instead of looking at isolated usability issues, we examined the journey step by step from a new user's perspective. At each step, we asked four questions to identify potential points of friction.
Q1
Will users try to achieve the right outcome?
Q2
Will users notice that the correct action is available?
Q3
Will users associate the correct action with the result they're trying to achieve?
Q4
After the action is performed, will users see that progress is made toward the goal?
01Store Selection
The flow could trap users in a dead end
The flow could trap users between store selection and product details.
Supporting evidence
- Users entered an endless "Store Selection → Product Details" loop when available stores were outside their delivery area.
02Add to Cart
Actions did not always lead where users expected
Several interface elements suggested actions that produced different outcomes than users expected.
Supporting evidence
- "Continue shopping from this store" opened the cart instead of returning users to the store.
- Editing the selected store navigated to the store instead of allowing users to change it.
- Delivery method appeared editable even when pickup was unavailable.
03Navigation
Back navigation broke users' sense of place
Back navigation in most of the pages repeatedly broke users' sense of place.
Supporting evidence
- Pressing Back from the cart returned users to the Cart List instead of the previously visited store.
04Checkout
Errors showed the problem, but not the way forward
The interface reported problems but failed to help users recover.
Supporting evidence
- Unavailable products and pricing errors appeared without a clear next step.
- When delivery coverage changed, users were directed to change their address even when the issue came from the store's service area.
INITIAL ASSUMPTIONS
But these were assumptions, not findings. We needed behavioral evidence.
01
Navigation
Users may struggle to find the right category because back navigation behaves inconsistently.
02
Information architecture
The IA may not match users' mental models of FMCG shopping.
03
Product discovery
Users may have to navigate more than necessary to find a product.
BEHAVIOR VALIDATION THROUGH SESSION OBSERVATION
We took these assumptions to real user sessions. Many users skipped categories entirely and went straight to Search.
More than 30 real user sessions were reviewed to understand whether the interaction breakdowns identified during the walkthrough also occurred in practice.More than 30 real user sessions were reviewed to understand whether the interaction breakdowns identified during the walkthrough also occurred in practice. What users actually did confirmed some of our concerns and revealed another pattern we hadn't anticipated.
Instant "No Results" Message
Users immediately encountered a "No results available" message before typing anything.
Availability Not Visible in Results.
Product availability was not visible inside search results, requiring users to open product pages simply to discover items were unavailable.
Enter Instead of Suggestions
Many users pressed Enter instead of selecting instant suggestions, leading to a 404 page because no search results page existed.
We initially thought product discovery was mainly a navigation problem. Session recordings showed that for many users it wasn't. They bypassed navigation and went straight to search.
USABILITY TEST
Usability testing quantified where task completion broke down.
Session recordings showed us where behavior broke down. Usability testing helped us understand the reasoning behind it.
Task 01
Find a specific product and check its info.
Success Rate
80% (4/5)
Misclick Rate
40% (2/5)
Assisted Completion
20% (1/5)
Task 02
Add multiple products and complete purchase
Success Rate
40% (2/5)
Misclick Rate
100% (5/5)
Assisted Completion
60% (3/5)
PROBLEM SPACE
Five patterns consistently disrupted the shopping journey.
By combining the walkthrough, session recordings, and usability testing, we found five recurring patterns behind the friction.
01
Finding products was harder than expected.Search became the shortcut
Many users bypassed categories and went straight to Search, treating it as the fastest route to a known product. But when Search itself introduced friction, that shortcut became another barrier to purchase.
Search
Category Navigation
Product availability
02
Product selection wasn't always straightforward.
Once users decide a product is worth buying, they expect the next step to preserve that decision. Introducing another unresolved choice at this point interrupts purchase momentum and forces them to reconsider something they thought they had already decided.
Store selection
Product availability
Add-to-cart
Store/service-area constraints
03
The cart and checkout didn't preserve the user's progress.
Users build an expectation of what they are buying as they move through the journey. When product availability, store context, required actions or error states change later in the flow, they have to reconstruct that context instead of simply progressing toward payment.
Store selection
Price/stock changes
Product availability
Add-to-cart
04
Errors didn't provide a way forward.
When purchase conditions changed, users were told what went wrong but not what to do next.
Store selection
Product availability
Add-to-cart
Store/service-area constraints
05
Actions didn't behave as expected.
Navigation and interaction cues often led users somewhere different from where they expected to go.
Cart Navigation
Product availability
Add-to-cart
Store/service-area constraints
PERSONA
Two Different Behaviors,
One Underlying Failure.
Whether users were searching for a known product or exploring for a deal, they expected Bazargam to preserve their purchase intent as they moved through the journey. Instead, UX frictions repeatedly forced them to stop and figure out what to do next.
Routine Shopper
EFFICIENCY DRIVEN
“
I know what I need. I just need the fastest way to get it.
Core Goal
Find required products quickly and complete the purchase with minimum friction.
Pain Points
- Confusing zero-state search feedback
- Inability to recognize store delivery service zones early
- Getting trapped in store-switching loops after choosing out-of-boundary outlets
Promotion Explorer
VALUE DRIVEN
“
Show me good opportunities, and I'll decide what I need.
Core Goal
Discover valuable deals and decide what is worth purchasing.
Pain Points
- Hidden coupon eligibility rules at product views
- Friction between browsing discounted items and checking their local store limits
- Overlooked fields during checkout due to low prominence
Two different goals, one shared point of failure – the experience could not hold a consistent shopping context across the journey.
IDEATION & PRIORITIZATION
We looked for solutions that worked for users, fit Bazargam's constraints, and supported business goals.
Desirability
What should we solve for users?
Feasibility
What could we realistically change?
Viability
Why is it valuable for the business?
SOLUTIONS
Design decisions were shaped by three perspectives: user evidence, business needs, and implementation realities.
Research set the direction. The Three Lenses kept each response grounded in a user need, a real delivery context, and a product hypothesis.
01
Make product discovery and evaluation easier.
Make product discovery and evaluation easier.
What we built
We improved Search and product information to make finding and evaluating products easier.
Why this decision held up
DESIRABILITY
Users needed a faster way to find and evaluate products, especially when Search was their preferred shortcut.
FEASIBILITY
Builds on the existing product-indexing layer already powering categories.
VIABILITY
Better discovery could reduce search abandonment and increase add-to-cart opportunities.
02
Make product selection predictable.
Make product selection predictable.
What we built
We clarified store selection, product availability, and add-to-cart actions to make product selection more predictable.
Why this decision held up
DESIRABILITY
Users needed clear product, availability, and store context before committing to add a product.
FEASIBILITY
The solution had to account for store/service-area logic and product availability states already present in the product.
VIABILITY
Making purchase conditions clearer could reduce failed or interrupted add-to-cart attempts.
03
Keep users oriented from cart to checkout.
Keep users oriented from cart to checkout.
What we built
We redesigned the cart and checkout experience to preserve context and make the path to purchase clearer.
Why this decision held up
DESIRABILITY
Users needed to maintain a clear sense of their choices and progress as they moved toward payment.
FEASIBILITY
The redesign required rethinking existing Cart and Checkout actions, including removing or redesigning actions that did not support the intended flow.
VIABILITY
A clearer, more focused path to payment could reduce friction before purchase completion.
04
Give users a clear way forward when things change.
Give users a clear way forward when things change.
What we built
We redesigned availability and error states to give users clearer feedback and a way forward.
Why this decision held up
DESIRABILITY
Users needed to understand what changed and what they could do next when purchase conditions changed.
FEASIBILITY
Recovery states had to work with existing availability, store, product, and checkout conditions.
VIABILITY
Giving users a clear recovery path could prevent avoidable dead ends during purchase.
Across the redesign, we also aligned interaction patterns with users' existing mental models, making navigation and key actions more predictable.
Pattern 05
IMPACT
How we would measure success...
By tracking key product events and analyzing the user funnel, we can measure how the redesign affects progression from search and product discovery to cart and checkout.
Add-to-cart Rate
Expected Impact
Clearer product information, more visible product availability, and more accessible add-to-cart actions should reduce hesitation and friction at the point of product selection.
Search Abandonment Rate
Expected Impact
Improved search feedback, clearer search results, better product visibility, and a proper recovery path for unsuccessful searches should help users continue their shopping journey.
Cart Abandonment Rate
Expected Impact
A clearer cart state, better visibility of product and availability changes, and more predictable interactions should reduce uncertainty and friction between cart and checkout.
Checkout Conversion Rate
Expected Impact
Reducing unnecessary friction, clarifying required actions, improving error recovery, and making the checkout state easier to understand should help users progress through checkout.
REFLECTIONThe screen wasn't the problem.
The screen wasn't the problem.
The journey was.
The redesign started with individual usability issues. But as evidence accumulated, they revealed a larger problem: Bazargam couldn't consistently preserve the user's shopping context from discovery to checkout. I learned to look beyond individual screens and understand the journey as a whole. From now on, I'll understand the journey first, then redesign the interface.