top of page






Represented the team by communicating weekly with industry partners on project updates, scheduling weekly meetings, and shared resources. 

Prototyped experiences at low and high fidelity through concept sketches, wireframes, and interactive prototypes. 

Conducted observations 

Led the team in the affinity mapping and journey mapping process.






Visually Impaired users in being able to use grocery store kiosk independently with the addition of an AR Scanner and Image Recognition

SUS Score: 74.1

No. of Clicks: 5

Time on Task: 34 Seconds

1.3 billion people are estimated to live with some form of visual impairment. They are often unable to perform simple tasks due to the lack of accessibility features.


Currently, the NCR Corporation grocery store kiosks do not have many accessibility features for visually impaired users. Many necessary accessibility features are missing from their grocery store kiosks. Our job was to work to add accessibility features to the SS90 kiosk for users that are visually impaired. 

Pain Points

Kiosks are currently not adaptable.

Kiosks are not easy to use or to recover from errors for users with visual impairments.

Kiosks do not provide appropriate feedback to help users navigate the system.

Understanding Users

Audible directions to follow at self-service kiosk

Clear feedback to understand my actions


Completely unable to use kiosks due to the lack of detailed audible directions

Depend on others to complete grocery shopping and checkout


Unable to adjust components on self-service kiosks

Need to use assistive technology on my phone in order to use kiosks

A way to customize the self checkout experience

Better accessibility features on self-service kiosks



35 / Single / Musician



43 / Married / Librarian

Born blind

Legally blind

How Might We 

How might we make self-service kiosks easier to navigate and recover from errors?


How might we enable self-service kiosks to provide clear feedback to our users?


How might we make self-service kiosks easily customizable?


We used our research to help us conduct an informed brainstorming session. Together we considered different ways to solve the pain points identified.


After considering the feasibility and creativity of each idea, we narrowed our ideas down to three different potential solutions that could address the identified "How Might We" statements.


Users loved the elimination of barcodes by using Image Recognition to scan items. They were rushing to try this feature again! When trying to pair the phone to the kiosk, users wanted more feedback on how to complete this action. 

Users thought the system seemed intuitive and fresh. Many of our users were smiling while interacting with the kiosk

Potential Solutions

We used the graph and the ideas in the optimal zone to help us create three possible solutions. Next, we created detailed sketches and wireframes of each potential solution. 

We referred back to user research, user pain points, and the "How Might We" questions in order to make sure we stayed focused on user needs. Finally, we got feedback on each idea from users.

Tap to Change Accessibility App


Potential Solutions

This application allows users to tap their phones to self-checkout kiosks in order for the kiosk to be instantly customized according to the accessibility features on the user's phone. Users can also add items to a shopping list in the app before or while shopping.


  • Users loved the ease of being able to tap their phone to the self-checkout kiosk and enable it to instantly customize to their needs.

  • It was unclear to users how to navigate through the app.


AR Navigation & Item Checkou​t

This application uses Augmented Reality to help users navigate the store. It also allows users to scan aisles and sections within aisles in order to identify the correct products and add those products to their shopping cart.


  • The ability to have in store AR navigation and checkout at our user's fingertips was well received.

  • Unfortunately, concerns were raised about the potential safety threat walking and utilizing this solution could pose. 


Smart Cart

Smart Cart is a smart shopping cart that scans items and reads them aloud as you put them in your shopping cart and adds them to your virtual shopping cart. Once the cart gets to the self-checkout kiosk, it will automatically upload all the items in the cart to the kiosk. 


  • The ability of the smart grocery cart to read items aloud and instantly add them to the user's virtual shopping cart on the kiosk was very helpful.

  • Some users worried about the time it would take to add potential purchases into the shopping cart just to be able to identify the item.

Iteration 2

Our findings from iteration 1 led us to add a new feature to our iteration 2 prototype, an AR Scanner. This Scanner came from one of our original ideas. Users with visual impairments were still having trouble identifying the correct items and price of items in the grocery store.

Mr. Merchant Kiosks Identified

Users with visual impairments can now use the Mr. Merchant app to help them find self-checkout kiosks that they can customize! The app helps them find grocery stores near them and shares the availability of Mr. Merchant kiosks.

The UI of the app can be adjusted according to the accessibility settings on the user's phone.

Simple Checkout

The checkout process has been simplified in order to help users avoid errors. They are able to also quickly call for help if needed. The UI has been optimized for readability as most of our users prefer black text on a white background.

Users liked that the AR Scanner helped them to identify their items while grocery shopping. They also enjoyed the elimination of barcode scanning with the use of Image Recognition by the kiosk. 


User testing was completed via Expert Testing &  In-person moderated usability testing. We asked Usability & Accessibility experts to participate in a Cognitive Walkthrough and a Heuristic evaluation. We also had visually impaired users & users with unimpaired vision to participate in usability tests onsite at NCR's Usability Lab.


Key Takeaway: Be mindful of scope. 

We were so excited and passionate about solving all of the needs and pain points of our users. However, it became clear that we did not have enough time to address each issue within the scope of our one semester project. I learned to focus on the identified "How Might We" questions that were created from narrowing down the pain points. Then, it's important to design for those problems well.  Future iterations and user testing of the first iteration can be used to add on more functionality.


August 2019 - December 2019



Adobe XD, Figma

Anjali, Yujin, and Nektar​

I helped redesign grocery store kiosks made by NCR to make them universally accessible to all.

of our users need to use the voice assistant on their phones five or more times a week.

of users get their own groceries each week.

of users have screen magnification or screen readers set on their computers.

of our users wanted to be able to adjust the text display options on kiosks.







from our research

The Mr. Merchant app allows users to pair their phones to the kiosk in order to change the visual display on the kiosk to better meet their needs. Accessibility settings in Mr. Merchant app are sent to the Mr. Merchant kiosk in a quick process using NFC technology.


Accessibility Settings

The NFC pairing between phone and kiosk is a novel interaction, so additional feedback (audio/visual tutorials) could be useful to guide users through this process.


The AR Scanning feature was well received. However, users wondered if the AR Scanner could account for scanning and pricing multiple items at a time, and what would happen if it picked up items in the background.


The placement of the Accessibility menu option is not consistent with existing settings conventions in users' phones. It was hard for them to find in the app. Should they go to settings or accessibility?


The kiosks aren't really something I use. I get frustrated, because sometimes the voice works and sometimes it doesn't. I'd love to be able to use it one day though.

User that is Fully Blind 


To better understand users with visual impairments, we observed them using assistive technology. We also observed people using the XK7 & the SS90 kiosks at NCR. We saw immediate difficulties with the available accessibility features.


We studied self-service checkout kiosks and software that currently have accessibility features. We compared the ADA Compliance, support for Assistive Technology products, customizability of the kiosk screen, and ease of use. From our analysis, we learned that many existing products have been unsuccessful for users with visual impairments.

Competitive Analysis

Next, we interviewed users to get insights into how individuals with visual impairments used assistive and regular technology, shopped for groceries, and how they received assistance with technology when in stores.



Finally, a survey was created in order to find out users' likes and dislikes in regards to self-service kiosks in grocery stores. We also asked preliminary questions to uncover the specific details of their preferences in regards to assistive technologies.

Affinity Map

An Affinity Map was used to organize the information we learned from our users. They were unable to use self-service kiosks in their current state, but we wondered if there was something to be learned from the assistive technologies they already used.

Pain point identification helped us to narrow down on problems that were within our scope to solve.



After receiving feedback from usability tests, I saw a need to scale down the amount of features available to users. We added many features that were needed, but we did not put enough emphasis on addressing our "How Might We" questions. I decided to complete iteration 3 after the project ended.

Iteration 1

We chose to go with the Tap to Change Accessibility solution with a few extra integrations. This solution will allow users to access grocery shopping information before arriving to the store and add items to their shopping list.


  • 3D map to help users find items in the store.

  • Tap to pair phone to kiosk to send customized accessibility settings to kiosk.

  • Computer vision to recognize items when they are held in front of the kiosk in a new scanning interaction that eliminates barcodes.


Iteration 3

The second iteration was completed in Fall 2020 after receiving feedback on iteration two. Iteration 3 is shown in detail below. All features have an accompanying speech option that reads the text on the screen.

Change the Kiosk's Accessibility Settings

The Mr. Merchant kiosk also allows users to change the accessibility settings right from the kiosk. Users can also see a preview of what the screen will look like as the accessibility settings are updated. 

bottom of page