.webp)
.webp)
Role
Product Designer (end-to-end)
TEAM
1 Product Designer
3 Developers
1 Project Manager
1 Founder & 3
Stakeholders
TOOLS
Figma
Maze Testing
AI Tools
Atlassian
Microsoft Clarity
Client
Autoleague - 80+ Dealerships across Australia
JUMP TO:
OVERVIEW
This case study showcases the end-to-end process of the app's core features to address key pain points, enhance user satisfaction, and support business growth.
Leading the process
As the end-to-end Product Designer, I led everything from research and UX strategy to design systems, prototyping, and testing. The mission was clear: re-think how dealerships engage online by creating a platform that had never existed before.
About AI Dealer Webchat
The AI Dealer Webchat is the first platform of its kind globally, a unified system that links an AI-powered dealership chatbot with a smart, dealer-facing communication and lead-management portal. Unlike traditional automotive chat tools, it combines real-time AI assistance, automated inbox organisation, branding customisation, and full end-to-end journey support in one experience. The platform enables dealerships nationwide to manage AI-driven conversations effortlessly, respond to leads instantly, customise their chat UI, and support every major automotive journey, from buying and trade-ins to servicing and parts. It’s the first AI system built to support the entire dealership workflow, both customer-facing and internal.
Core Features:
AI Chatbot (Customer-Facing)
- Real-time AI responses tailored to dealership workflows
- Supports all major journeys: buying, test drives, trade-ins, valuations, servicing, parts
- Clear next-step guidance to reduce friction
- Seamless handoff to a human when needed
- Full UI and branding customisation for each dealership
Dealer Webchat Portal (Internal)
- Smart Inbox that auto-sorts conversations into groups i.e. vehicle leads, trade-ins
- AI-powered reply suggestions
- One-click action buttons (share key features, specs, reports, images)
- Ability for dealers to type directly or take over from AI
- Conversation management tools
- Reporting dashboards showing analytics and performance trends
Problem
Despite rapid AI growth in other industries, automotive chat tools remain outdated and frustrating. They fail to recognise intent, lead users into dead ends, and don’t support key journeys like buying, trade-ins, or service bookings. Dealers also lack control, clear visibility, and often treating each chat the same. This results in customers leaving without answers and dealerships losing high-intent leads. With no global solution connecting AI-driven customer conversations to an intelligent dealer-facing platform, the industry needed a modern, unified approach.
Solution
We built the world’s first AI Dealer Portal. A unified system combining an AI-driven customer chatbot with a powerful dealer workspace. The chatbot supports every key dealership journey, from car purchases and test drives to trade-ins, valuations, servicing, and parts, delivering accurate responses and clear next steps. The dealer portal provides a Smart Inbox that auto-sorts chats, AI-assisted replies, one-click actions, human takeover, branding customisation, and real-time reporting. It’s the first end-to-end AI solution designed for both customers and dealerships.
Deliverables
Becoming the first integrated AI chat and dealer workspace in the industry. It is predicted In the first 3 months, dealerships will see a 64% lift in vehicle leads and a 47% reduction in response times. The AI chatbot is predicted to achieve a 89% task-completion rate, far outperforming traditional tools and reduced manual triage by 38%, with over 1,500 weekly conversations handled by AI. Staff testing that platform have reported clearer workflows, better lead visibility, and a noticeably improved customer experience, establishing the platform as the new benchmark for AI-driven dealership communication.
6
Key User Flows
32
Screens
130+
Users Tested
Impact
Becoming the first integrated AI chat and dealer workspace in the industry. It is predicted that in the first 3 months, dealerships will see a 64% lift in vehicle leads and a 47% reduction in response times. The AI chatbot is predicted to achieve a 89% task-completion rate far outperforming traditional tools and reducing manual triage by 38%, with over 1,500 weekly conversations handled by AI. Staff and users testing the platform have reported clearer workflows, better lead visibility, and a noticeably improved customer experience, establishing the platform as the new benchmark for AI-driven dealership communication.
predicted project results
89%
Task-completion rate
1,500
Weekly Conversations
64%
Increase Vehicle Leads
38%
Reduction in Manual Triage
200+
Dealerships Using in the First Year
%20(2).jpg)
JUMP TO:
RESEARCH
The current dealership webchat suffered from multiple usability issues, resulting in high customer drop-off and lost leads. Conversations frequently stalled, leaving users at dead ends with no clear next step. To address this, I dug deeper into these pain points to identify meaningful opportunities for improvement and innovation.
/01
Analysing dealership analytics
I reviewed behavioural data from dealership websites to understand how users navigated inventory, service pages, and enquiry forms. Patterns showed high drop-off at key conversion points and frequent confusion during chat interactions. These insights helped identify where the AI chatbot needed to provide clearer next steps and where the dealer portal could streamline lead handling. The data directly shaped the core workflows and priorities for both user groups.
Insights gained:
- Users dropped off when chatbots failed to offer clear next steps.
- High-intent actions (test drives, valuations) had the highest abandonment rates.
- Long wait times and unhelpful automated replies caused frustration.
- Dealers often missed messages due to poor inbox organisation.
- Confirmed the need for smarter AI flows and automated lead prioritisation.
.webp)
/02
Competitor Research
I conducted a competitive analysis across existing dealership chat tools, generic AI chat platforms, and automotive CRM systems. Most solutions lacked depth in intent recognition, had rigid workflows, and provided limited visibility for dealership staff. Mapping out feature gaps and UX weaknesses helped define opportunities for a unified system, one that supported full customer journeys and offered smarter lead management than anything currently in the market.
Key findings:
- Most competitors relied on rigid scripts with limited AI understanding.
- No tools supported full dealership journeys end-to-end.
- Dealer-facing tools lacked visibility, structure, and workflow clarity.
- No system combined AI chat + dealer management in one platform.
- Revealed a clear opportunity for a unified, smarter solution.
Opportunities Identified:
- Build an intent-aware AI chat that understands customer needs instead of relying on rigid scripts.
- Support end-to-end dealership journeys, from first enquiry through follow-up and conversion.
- Combine AI chat and dealer management tools into a single, unified platform.
- Give dealership staff clear visibility into conversations, lead status, and next actions to improve efficiency and response quality.
.webp)
/03
Project Documents to low-fi
Using technical requirements and business objectives, I translated early project documents into low-fidelity user flows and wireframes. These sketches acted as the first shared language between design, product, and engineering teams. They allowed rapid iteration on core features like the Smart Inbox, AI suggestions, and conversation categorisation before moving into detailed UX. This stage ensured alignment and prevented costly redesigns later.
.webp)
ideation
Heading
Inspiration
I explored modern AI tools, CRMs, messaging platforms, and conversational UI patterns to gather inspiration for both the chatbot and dealer portal. This helped identify emerging interaction models that could be adapted for dealership workflows. The inspiration phase informed key decisions around the layout, information hierarchy, and how AI suggestions would surface naturally within conversations, balancing efficiency with a clean, modern interface.
%20(1).webp)
/05 High-fidelity wireframes and User Flows
I produced detailed high-fidelity wireframes for all major user flows, including chat management, inbox categorisation, reporting, and customisation settings. These wireframes acted as functional blueprints for the development team and were refined through feedback loops with stakeholders. They demonstrated how AI and human actions blended seamlessly and ensured every interaction - from sending links to taking over a chat, felt intuitive and efficient.
.webp)
DESIGN
Heading
Creating a Design System
I built a scalable design system covering typography, colours, spacing, icons, components, and interaction states. The system supported both the customer-facing chatbot and the entire dealer portal, ensuring visual consistency across hundreds of screens. It streamlined development, reduced UI ambiguity, and made future feature expansions significantly faster. The system also accounted for custom branding, accessibility requirements, and responsive layouts.
.webp)
Heading
UI Design and prototyping
I created polished UI designs that balanced modern aesthetics with clear functionality. This included layouts for the inbox, chat panel, reporting dashboards, vehicle cards, and customisation tools. Interactive prototypes were built to validate flows like sending a suggested message, sorting chats, or switching between categories. These prototypes helped stakeholders and developers visualise the product experience before build, reducing misunderstandings and revisions.
Heading
User Testing
I conducted Maze testing to measure task completion, clarity, and overall usability across key portal flows. Test participants replicated real dealership tasks like identifying leads, replying using AI suggestions, and navigating the reporting dashboard. Insights from Maze highlighted where users hesitated, what felt intuitive, and where labelling or layout needed refinement. The feedback guided iterative UI improvements and validated core design decisions before development.
140+
User Testers
94%
Test Completion Rate
‘Clean’ ‘Easy to use’
Main Emotional Response
Impact
Smarter UX leads to stronger outcomes
This project was about solving dealership issues and user problems through smart strategy and focused execution.
- I started with strategy, not screens
- I prioritised business goals and users needs
- I designed a scalable system, not just static mockups
- I delivered smart solutions that breached expectations (not just what was asked, but what was needed
predicted project results
89%
Task-completion rate
1,500
Weekly Conversations
64%
Increase Vehicle Leads
38%
Reduction in Manual Triage
200+
Dealerships Using in the First Year
.webp)
.webp)
.webp)