
New AI SaaS Product to increase user acquisition for Honda Motors.
Web & Mobile A.I. Product
What
I worked as a UX designer with Honda R&D team on designing a new AI vehicle maintenance platform. My role included participating in ideation workshops for product strategy, synthesizing research with teams, and then leading the design of the web and mobile platform.
Why
We wanted to find a new revenue opportunity focusing on the Gen Z customers. As the automobile industry is navigating towards EVs, we wanted to create a solution for our customers to attract them & build customer loyalty and build trust with the brand.
How
I ran multiple ideation workshops and design sprints, collaborating closely with the research & engineering teams, working in Agile.
Company
Honda Motors
Duration
5 months, 2023
Team
Designer, I. Clark
PM, A. Paulisik
Service Manager, T. Jones
Engineer, E. Clepper
VP, Honda Research, D. Detwiler
Contribution
Competitor Analysis
Ideation Workshops
UX Flows
Prototyping
UI Designs
in brief
It's like having your very own vehicle expert right on your smartphone or Laptop!
Enterprise App for Businesses (B2B)
The platform empowers dealership mechanics to efficiently manage vehicle repair requests while simplifying data sharing between vehicle owners and Honda.
Mobile App for vehicle owners (B2C)
A mobile application for Honda vehicle owners to identify and diagnose problems using A.I. capabilities. It's like having your very own vehicle expert right on your smartphone!
impact
Personalized and faster onboarding by 30%
The new progressive onboarding experience for the Honda App for vehicle owners. I pinpointed a chance to enhance our onboarding process for new users, simplifying it to boost conversion rates by reducing the 10 step process to 7 steps.
CONTEXT
Vehicle owners do not trust the repair prices at dealerships leading to low user acquisition
Research Activities
Interviews of Gen Z using directed storytelling method to validate ideas, Co-design session with Mechanics, dealerships affinity mapping.
Workshop Facilitation with Gen Z vehicle owners. We found that vehicle owners face challenges in trusting dealerships with the price, process and time, and it was observed in interviews and internet research sessions, in YouTube comments, Reddit and Facebook groups.
process
Low Risk, Known Problem → Test-and-Learn
By pushing our ideas to the users, we watched, learnt and refined our designs through iterative design. For the project, the risk was low and the problem was a familiar territory for Honda, so adopted the test and learn approach.
Our research was open-ended in the beginning
I approached the project by rapidly sketching ideas & testing with the end users. Because of the open brief, I made user research and business canvas model presentations to spark interest by balancing both user & business problems that we could solve.
Agile Process
I worked in multiple sprints collaborating with designers, PMs, Engineers, getting feedback from the stakeholders at every step. My role was to communicate clearly the identified user problems & solutions to the VP and the R&D team.
key INSIGHT 1
Gen Z wants to know the seriousness of the vehicle issue before taking it for repair
Initial sketches showcasing ideas to solve process, price and time problem. We found that vehicle owners face challenges in trusting dealerships with the price, process and time, and it was observed in interviews and internet research sessions, in YouTube comments, Reddit and Facebook groups.
key INSIGHT 2
Service managers at dealerships find it hard to know the type of customer that they are dealing with
Initial sketches showcasing the user preferences like communication. A big part of service managers job is to build relationships with the vehicle owners. However, our insight informed us that Gen Z do not want to engage with the managers and prefers to skip dealerships visit because of the communication.
prototype
We decided to make a high 'visual' fidelity prototype to test if Gen Z would pay for diagnosis service.
High visual fidelity prototype created in Figma. A big part of service managers job is to build relationships with the vehicle owners. However, our insight informed us that Gen Z do not want to engage with the managers and prefers to skip dealerships visit because of the communication.
challenge faced
Our solution cannot work outside the dealership model
Storyboard for the problem. We encountered a huge set back in the middle of the project, after learning that Honda cannot operate outside the existing dealership model. This means dealerships need to be part of the solution. We decided to visit a couple of dealerships to understand their pain points as next steps.
next steps
Go back to drawing board
Once we discovered our concept was ineffective, I encouraged my group to compile our insights and reevaluate the issue from a new perspective. I drew multiple UX flows, re-evaluated the value proposition and tested with the users working in a design sprint, that we ran for 3 days.
DESIGN solution
A diagnosis app for vehicle owners and web and mobile app for dealerships


patterns and practices for a.i. interactions
I researched and designed the new basic mechanisms for enhancing the A.I. experience for our users.
01: Auto-completions
Ghost text for Autofill activity to support the user's with prompting as the first basic mechanism.
02: Select
Brushes for selecting a text and applying a transformation to it as the second basic mechanism.
03: Chat
Users explore the different outcomes until they are satisfied by the outcome as third mechanism.
01: design for optionality
Auto-completions for Autofill activity

AI Ghost text Modal. Get Autofill responses based on issues with similar cars.
02: Design for Transformation
Select for fine tuning

AI Brushes Interface Modal. Highlight text to get additional options to fine tune the prompting.
03: Design for Iteration
Chat for exploration

LLM based Chat's design Pattern. Use chat to explore the common issues.
prioritization - design vs engineering
It decreased "computing load" but increased the "cognitive load" for our Users

This had the benefit for lowering the cost since we would not be running multiple things at once but only one prompt, however for the user, it would add to the cognitive load with the popup modal. Hence, we decided to drop this interaction modal because it affected the core user experience.
CORE EXPERIENCE
Diagnose vehicle issues with AI

Self-diagnosis for vehicle owners through a mobile app interface that employs text, noise, and image recognition in conjunction with DiagnosisGPT. Easily identify the vehicle issues using the DiagnosisGPT interface developed on the Honda vehicles diagnosis dataset.
CORE EXPERIENCE
Estimate Pro for Businesses
Enterprise platform for Automotive Dealerships showcasing requests on home page & Insights. The enterprise platform allows mechanics to communicate directly with their clients i.e vehicle owners.
CORE EXPERIENCE
Diagnose remotely with Mechanics

Information exchange between vehicle owners and mechanics enhancing easy data sharing & dialogue between them. The Premium service allows vehicle owners (B2C) to quickly connect with mechanics helping to resolve urgent issues.
data visualization
Dashboards for B2B


To bring relevant info to Mechanics in a simple and organized manner, I designed dashboards - thereby reducing cognitive load, saving time and increasing efficiency.
DESIGN ELEMENTS
Designing for delight
Micro-design system and character design for micro interaction. I designed the various design elements for the project like micro-interactions, characters, color palette etc.








