Grab & Go

In 2023, using a self-checkout kiosk often felt like a gamble. What should've been quick turned into a cycle of alerts, bagging errors, and waiting for help.

I joined a small team of designers and engineers to reimagine this daily interaction, used by millions, enjoyed by few. Our goal was simple: make self-checkout seamless, intuitive, and accessible for all.

Design by accretion

Self-checkout kiosks evolved from niche experiments to mainstream fixtures in supermarkets worldwide. What started as a quick, convenient option became the default for millions. But as usage grew, the system was patched together rather than redesigned. Barcode scanners, bagging sensors, and payment options were added incrementally, each solving part of the problem but not the whole. Over time, the interface became a mismatched collection of decisions, losing the sense of innovation it once had.

Our mission wasn't just to revive the kiosk's original simplicity; it was to create a system meeting the needs of today's diverse users, from tech-savvy millennials to older adults using touchscreens for the first time.

As Lead Product Designer and UX Strategist, I led UX research and strategy for the self-checkout redesign. I worked with a cross-functional team to identify pain points and create a new interaction flow. Collaborating with Ayan Ahmad Khan on UI design and Budhil Raj Patel on technical strategy, we conducted field research, iterated on prototypes, and defined the north-star experience.

Kickoff

At the start, we didn't have a clear brief or a defined vision of what "better" looked like for self-checkout. But we shared a strong hunch: the experience felt slow, unpredictable, and unforgiving. To move forward, I partnered with Ayan to observe how real shoppers interacted with kiosks in busy supermarkets. No scripts, no assumptions. Just watching, listening, and tracking frustration as it unfolded in real time.

Evolving expectations

Initially, issues like misplaced barcodes and extra taps seemed minor. But as we analyzed behavior, it became clear shoppers expected a seamless experience, not something they had to "figure out." As self-checkout became common, users developed a mental model of how it should work. Delays and unclear prompts felt like unnecessary friction, and frustration grew not just with the system, but with the effort required. Improving usability wasn't enough. We needed to rethink the experience to make it intuitive and effortless.

The discovery: most self-checkouts require additional effort

The data revealed friction in nearly every session. Users frequently retried barcode scans, adjusted bagging placements, or sought staff help. Mistakes led to hesitation, backtracking, and trial-and-error. The experience was far from the seamless, independent process it was intended to be.

Impact on revenue and customer experience

In high-footfall supermarkets, inefficiencies led to long queues, abandoned purchases, and frustrated customers, directly affecting sales and satisfaction. In a city like Delhi, with thousands of daily transactions, these inefficiencies caused significant weekly revenue loss, especially during peak hours.

Reframing the problem

The self-checkout system, instead of simplifying, adds friction at key touchpoints: unreliable scanning, confusing bagging errors, unclear feedback. Shoppers re-scan items, adjust groceries, or call for help, slowing transactions and increasing cognitive load.

Guided Checkout transforms self-checkout into a seamless, intuitive process. By reducing friction at every step, it makes smart real-time adjustments and gives clear, actionable feedback.

A more inclusive checkout experience

Traditional self-checkouts assume tech-savvy, unchallenged users. To address this, I designed for everyone, considering spectrums (temporary or permanent challenges) and situations (managing kids, rushing for a train). Moving away from one-size-fits-all, we prioritized flexibility and scalability, shifting from efficiency to inclusivity.

From degradation to adaptation

I created a Hierarchy of Needs framework to move from degrading features for tough environments to building inclusive solutions from the start. Instead of "How will this degrade in tough environments?" we asked "How can we design a seamless self-checkout for all users, regardless of constraints?" This kept a high quality standard: effortless, accessible checkout from high-tech urban stores to markets with limited infrastructure.

Self-checkout errors often arise from unclear user intent: mis-scanned items, accidental selections, confusion over feedback. Instead of manual correction, we focused on intelligent, adaptive assistance.

Smart Checkout Accelerators & de-risking

To optimize checkout, we focused on shopper intent. Asking "How would you like to pay?" at the right moment aligned with expectations: streamlining payment, reducing hesitation, and preloading likely options. I designed Smart Checkout Accelerators: predictive shortcuts offering 1-tap access to payment methods, bagging preferences, and receipt options.

To test our assumptions, we ran on-site usability sessions in Bangalore, Delhi, and Ahmedabad. To our surprise, not a single participant struggled with the adjusted flow. The scan-to-pay experience and quick payment shortcuts felt intuitive, even welcome.

Live Scanning: adapting to real behavior

Live Scanning emerged from a simple question: "Do users really need to follow a fixed checkout flow?" Instead of a rigid step-by-step process, Live Scanning adapts in real time, surfacing the right action (payment, bagging, receipt) only when needed. This shifted our mindset from "Here's how checkout works" to "How would you like to complete your purchase?" The result: a checkout flow that's faster, and feels more personal and intelligent.

In early phases, I believed more visibility meant more trust. So I layered in confirmations, breakdowns, and verifications to reassure users. But testing told a different story:

• Most users didn't read the breakdowns; they assumed the system was working.
• The excessive feedback slowed people down, creating friction instead of reassurance.
• The anxiety we felt as designers translated into the UI, without addressing real user concerns.

After three test rounds, I scrapped the complexity for a simplified, balanced interface: clear, but not overwhelming. The system became more invisible, letting users move forward with confidence, not caution.

We built a system that adapts, automating smartly and seeking input only when necessary. Three core principles carried it.

SmartCheckout: context-aware guidance

SmartCheckout is a dynamic guidance system mirroring how shoppers naturally interact. Using behavioral heuristics, it streamlines scanning, handles errors proactively, and simplifies payment. I modeled user behavior across real-world scenarios, revealing that intent, familiarity, and store environment heavily influence interactions. The long-term vision was a self-optimizing system adapting in real time; due to the complexity of full automation, we piloted a structured-yet-flexible version to validate key behaviors.