Before someone walks into a showroom, they've already decided.
EV
When the Aera launched, there was nothing for a prospective buyer to compare it to. The only way to form an opinion was to see it — and the only place to see it was a dealership.
The brief wasn't "improve the buying flow." It was earlier than that: give people a reason to want to show up. The digital experience had to carry the full weight of first impression — and then hand off to the showroom with intent already formed.
Objectives
360° visualizer to increase conversion to test ride and bookings. & Improve SEO through social sharing and inbound links.
Product scope
1 model, 2 variants, 7 colors. Prices visible per variant. No accessories in Phase 1.
End states
Book a test ride or Book a visit — both driven by user choice, not a hard funnel.
MY ROLE
End-to-end UX/UI — research, IA, interaction spec, visual design, handoff.
The System
8 configurators.
One consistent pattern.
I benchmarked the best vehicle configurators available — ranging from EVs to superbikes — looking specifically for how each handled the moment between discovery and decision.
Ultraviolette
Closest Indian EV comp. Strong visual hierarchy, weak share flow.
Tesla
Masterclass in scope restraint. Price-first, no noise.
BMW · Mercedes
Premium interaction standard. Floating hotspots for feature discovery.
Bajaj Pulsar
Mass market reference. Mobile-first patterns, Indian buyer behavior.
Ducati
Sound as experience. Rev button as emotional hook.
TVS · Triumph
Partnership product, similar buyer profile. Navigation and variant patterns.
01
The vehicle owns the screen always.
Every benchmark that held attention made the product the protagonist. Controls, labels, CTAs lived at the periphery. The moment UI chrome competed with the object, users stopped looking at the bike.
02
Variant and color are different decisions, they need different UI.
Variant is a specification choice that changes what you're buying. Color is personal expression. Collapsing both into the same selector type makes the interface feel like a form, not a showroom.
03
Feature discovery needs to feel like exploration, not documentation
BMW and Mercedes use floating hotspots — users click to zoom and learn about specific components. This pattern keeps the bike dominant while delivering spec depth on demand.
04
Sound and environment unlock a different layer of desire
Ducati's rev button wasn't a gimmick — it measurably increased time-on-page. Switching environments (day/night, city/highway) lets the user project themselves into ownership before they've made any commitment
User journey
Six steps from awareness
to walking in the door.
I mapped the full journey from first contact to showroom — then designed the visualizer to own the middle of it. The tool's job isn't to close the sale. It's to arrive at the dealership with the decision already made.
Awareness
Digital & physical ads.
Exploration
A guide introduces the tool
Configuration
Switches variants, tries colors, activates lights
Evaluation
Sees the cost and EMI split for the chosen variant
Save & share
Shares via templated image
One screen.
Every element earned its place.
Tight scope amplifies the weight of every call. With one model and two variants, nothing could hide behind feature density. The test for each element was the same: does this help someone fall in love with the bike?
CASE 1
The motorcycle owns the screen
Controls at the viewport edge. Bike holds 70%+ of the frame unconditionally. Switching colors had to feel like repainting — not submitting a form.
Core · 02
Flat architecture — intentionally
No wizard, no progress bar. With two variants, a step-by-step flow would be condescending. Single screen. All choices surface-level. Complexity lives in the system, not the interface.
Interaction · 03
Instant repaint — no transition
Color switching replaces the image with no animation, no fade. The absence of transition is the design choice — the change feels physical, not digital.
Interaction · 04
Floating hotspots for USPs
Tappable points on the bike — headlamps, powertrain, VIC, cooling. Each zooms in and surfaces specs + media in a backplated drawer. Navigate without zooming out.
Interaction · 05
Environment and lights
Tappable points on the bike — headlamps, powertrain, VIC, cooling. Each zooms in and surfaces specs + media in a backplated drawer. Navigate without zooming out.
Interaction · 06
Rev the bike
Engine sound with gear shifts. Background music fades when activated. Sound as emotional hook — borrowed from Ducati's playbook, applied to an EV context.
Share · 07
Templated share image
One tap generates a branded image — chosen view, Matter logo, configurator link. Designed to drive inbound from social. Download also available without email gates.
Conversion · 08
Two CTAs — user's choice
Both Book a test ride and Book a showroom visit available. No single path forced. Data showed pushing bike booking before sufficient exploration caused drop-off.
Built to ship Phase 1.
Designed for Phase 3.
Every architectural decision in Phase 1 was made with what comes after it in mind. Nothing would need to be ripped out — only extended.
Every cell in the grid can be viewed in full context. At any point, the pathologist can jump from the cropped single-cell patch to the full-slide 100x field of view — seeing the cell in relation to its neighbours, the surrounding red cell morphology, and the overall smear distribution. They can pan and zoom freely within the field. This replicates the muscle memory of moving a physical slide stage.
The image tiling was optimised so that the transition from cell view to full FOV felt instantaneous — even with 400MB+ slides. Perceived latency here would have broken the clinical rhythm entirely.
Summary & Report
From review to
signed clinical document.
Once the pathologist has reviewed cells across all tabs — WBC, RBC, Platelets, Malaria — they arrive at the Summary screen. This is where the case comes together: a consolidated four-column view showing differential counts, RBC morphology grading, platelet morphology, and a free-text Impressions panel on the right.
Impressions — structured + free text
The Impressions panel on the right supports both structured templates and free-text clinical narrative. The pathologist can choose a template (e.g. "Normocytic normochromic; no significant anisopoikilocytosis") or write their own interpretation. This hybrid approach matched how pathologists actually work — templated for routine cases, free-form for complex ones.
Approve or reject — a deliberate final gate
The final action is an explicit binary: Approve or Reject. Not a soft "done" or "submit" — a clinical sign-off that carries liability. The interface treats this moment with appropriate weight. The approved report generates a PDF and, where integrated, pushes to the Lab Information System (LIS) automatically.
Scale & Scope
Designed for one application.
Built to expand.
The AS76 launched with peripheral blood smear as its primary application — but the platform was architected from the start to support additional slide types through software upgrades. During my time on the project, bone marrow and tissue applications were also shipped, extending the same review interface paradigm to more complex diagnostic contexts.
UP TO
12
Slides processed per hour
UP TO
200
WBCs Pre Classified per case
UP TO
2,000+
RBCs graded to ICSH standard
Design Principles we set
Remote review
Cloud-agnostic architecture allows pathologists to sign off on cases from any location — democratising specialist access for rural and underserved labs.
On-device AI
All AI inference runs locally. No internet dependency, no patient data leaving the lab — critical for compliance and trust.
LIS integration
Seamless integration with existing Lab Information Systems means approved reports flow directly into the lab's existing workflows.
Outcome