Designed India's first Odometer fraud check system to build trust for used car market

Designed India's first Odometer fraud check system to build trust for used car market

Designed India's first Odometer fraud check system to build trust for used car market

Designed India's first Odometer fraud check system to build trust for used car market

Designed India's first Odometer fraud check system to build trust for used car market

Designed India's first Odometer fraud check system to build trust for used car market

Duration

Duration

Duration

4 months (Jan 25 - April 25)

4 months (Jan 25 - April 25)

4 months (Jan 25 - April 25)

Platform

Platform

Platform

mWeb and Desktop

mWeb and Desktop

mWeb and Desktop

My role

My role

My role

Product designer (0→1)

Product designer (0→1)

Product designer (0→1)

Live website

Live website

Live website

Impact at a glance

Impact at a glance

Impact at a glance

Purchase Conversion

Purchase Conversion

Purchase Conversion

+46.9%

+46.9%

+46.9%

NPS

Score

NPS

Score

NPS

Score

+38%

+38%

+38%

Feature Adoption

Feature Adoption

Feature Adoption

12k

Checks/month

12k

Checks/month

12k

Checks/month

Support Tickets

Support Tickets

Support Tickets

-39%

-39%

-39%

Time on listing page

Time on listing page

Time on listing page

+53% Engagement

+53% Engagement

+53% Engagement

A strategic product design initiative that increased buyer confidence, drove cross-sell revenue by ₹8Cr annually, and positioned Cars24 as India's most transparent used car platform

A strategic product design initiative that increased buyer confidence, drove cross-sell revenue by ₹8Cr annually, and positioned Cars24 as India's most transparent used car platform

A strategic product design initiative that increased buyer confidence, drove cross-sell revenue by ₹8Cr annually, and positioned Cars24 as India's most transparent used car platform

Overview

Overview

Overview

My role

My role

My role

Led end-to-end design from research to launch and optimization

Led end-to-end design from research to launch and optimization

Led end-to-end design from research to launch and optimization

Responsibilities

Responsibilities

Responsibilities

Drove product strategy and vision for trust-building initiatives

Led user research with 24 buyers/sellers and analyzed 6 months of behavioral data

Designed and tested 5 visual concepts with 2,400+ users

Collaborated with data scientists to translate ML outputs into intuitive interfaces

Facilitated cross-functional workshops with PM, engineering, operations (12+ stakeholders)

Defined success metrics and analyzed post-launch performance

Drove product strategy and vision for trust-building initiatives

Led user research with 24 buyers/sellers and analyzed 6 months of behavioral data

Designed and tested 5 visual concepts with 2,400+ users

Collaborated with data scientists to translate ML outputs into intuitive interfaces

Facilitated cross-functional workshops with PM, engineering, operations (12+ stakeholders)

Defined success metrics and analyzed post-launch performance

Drove product strategy and vision for trust-building initiatives

Led user research with 24 buyers/sellers and analyzed 6 months of behavioral data

Designed and tested 5 visual concepts with 2,400+ users

Collaborated with data scientists to translate ML outputs into intuitive interfaces

Facilitated cross-functional workshops with PM, engineering, operations (12+ stakeholders)

Defined success metrics and analyzed post-launch performance

Project Context

Project Context

Project Context

The ₹2.5 trillion trust problem

The ₹2.5 trillion trust problem

The ₹2.5 trillion trust problem

India's used car market was valued at $32.05B in 2024, with 4.4M cars sold annually 25% more than new car sales. Yet 30-40% of used cars have tampered odometers, making fraud the 1st barrier to online purchases.

India's used car market was valued at $32.05B in 2024, with 4.4M cars sold annually 25% more than new car sales. Yet 30-40% of used cars have tampered odometers, making fraud the 1st barrier to online purchases.

India's used car market was valued at $32.05B in 2024, with 4.4M cars sold annually 25% more than new car sales. Yet 30-40% of used cars have tampered odometers, making fraud the 1st barrier to online purchases.

The business challenge:

The business challenge:

The business challenge:

58% bounce rate on listing pages

58% bounce rate on listing pages

58% bounce rate on listing pages

72% of buyers cited odometer fraud as top concern

72% of buyers cited odometer fraud as top concern

72% of buyers cited odometer fraud as top concern

45% abandoned purchases due to verification issues

45% abandoned purchases due to verification issues

45% abandoned purchases due to verification issues

28% of support tickets were post-purchase disputes about mileage

28% of support tickets were post-purchase disputes about mileage

28% of support tickets were post-purchase disputes about mileage

My hypothesis:

My hypothesis:

My hypothesis:

If we provide free, transparent odometer verification with visual confidence scoring, we can reduce buyer anxiety, increase conversion, and create natural demand for paid vehicle history reports.

If we provide free, transparent odometer verification with visual confidence scoring, we can reduce buyer anxiety, increase conversion, and create natural demand for paid vehicle history reports.

If we provide free, transparent odometer verification with visual confidence scoring, we can reduce buyer anxiety, increase conversion, and create natural demand for paid vehicle history reports.

This wasn't just about fraud detection; it was about building systematic trust that would differentiate Cars24 in a commoditized market.

This wasn't just about fraud detection; it was about building systematic trust that would differentiate Cars24 in a commoditized market.

This wasn't just about fraud detection; it was about building systematic trust that would differentiate Cars24 in a commoditized market.

Problem Discovery

Problem Discovery

Problem Discovery

I began with deep user research to understand why odometer fraud was such a barrier:

I began with deep user research to understand why odometer fraud was such a barrier:

I began with deep user research to understand why odometer fraud was such a barrier:

Research methods:

Research methods:

Research methods:

24 user interviews (15 buyers, 9 sellers) across Bangalore, Delhi, Mumbai

Behavioral analysis of 6 months of funnel data identifying drop-off patterns

Field research shadowing 5 inspection agents to understand ground realities

Competitive analysis of CarDekho, Spinny, OLX Autos verification features

Referring support ticket mining revealing; 2,400 monthly disputes about vehicle condition

24 user interviews (15 buyers, 9 sellers) across Bangalore, Delhi, Mumbai

Behavioral analysis of 6 months of funnel data identifying drop-off patterns

Field research shadowing 5 inspection agents to understand ground realities

Competitive analysis of CarDekho, Spinny, OLX Autos verification features

Referring support ticket mining revealing; 2,400 monthly disputes about vehicle condition

24 user interviews (15 buyers, 9 sellers) across Bangalore, Delhi, Mumbai

Behavioral analysis of 6 months of funnel data identifying drop-off patterns

Field research shadowing 5 inspection agents to understand ground realities

Competitive analysis of CarDekho, Spinny, OLX Autos verification features

Referring support ticket mining revealing; 2,400 monthly disputes about vehicle condition

Key insights:

Key insights:

Key insights:

Confidence through visualization, not data dumps

Context is everything

The cross-sell sweet spot

Transparency builds trust

Confidence through visualization, not data dumps

Context is everything

The cross-sell sweet spot

Transparency builds trust

Confidence through visualization, not data dumps

Context is everything

The cross-sell sweet spot

Transparency builds trust

Design Process

Design Process

Design Process

Problem framing workshop

Problem framing workshop

Problem framing workshop

How might we make algorithmic decisions feel trustworthy to non-technical users?

How might we make algorithmic decisions feel trustworthy to non-technical users?

How might we make algorithmic decisions feel trustworthy to non-technical users?

Information Architecture

Information Architecture

Information Architecture

Vehicle Details → Odometer Check (free) → Service History (paid) → Checkout

Vehicle Details → Odometer Check (free) → Service History (paid) → Checkout

Vehicle Details → Odometer Check (free) → Service History (paid) → Checkout

Visual design iterations

Visual design iterations

Visual design iterations

Visual Design

Visual Design

Visual Design

Content strategy

Content strategy

Content strategy

Every piece of copy was intentionally designed to build trust without creating alarm

Every piece of copy was intentionally designed to build trust without creating alarm

Every piece of copy was intentionally designed to build trust without creating alarm

Prototyping and testing

Prototyping and testing

Prototyping and testing

The final design

The final design

The final design

Overview

Overview

Overview

Landing page

Landing page

Landing page

Vehicle details page

Vehicle details page

Vehicle details page

Result page

Result page

Result page

Design Impact

Design Impact

Design Impact

Impact at a glance

Impact at a glance

Impact at a glance

Metric

Metric

Metric

Before

Before

Before

After

After

After

Change

Change

Change

Purchase conversion

Purchase conversion

Purchase conversion

3.2

3.2

3.2

4.7

4.7

4.7

+46.9%

+46.9%

+46.9%

NPS Score

NPS Score

NPS Score

42

42

42

58

58

58

+38%

+38%

+38%

Feature Adoption

Feature Adoption

Feature Adoption

N/A

N/A

N/A

67.8

67.8

67.8

12k checks/month

12k checks/month

12k checks/month

Support tickets

Support tickets

Support tickets

2400/m

2400/m

2400/m

1908/m

1908/m

1908/m

-39%

-39%

-39%

Time on listing page

Time on listing page

Time on listing page

2m 45s

2m 45s

2m 45s

4m 12s

4m 12s

4m 12s

+53% engagement

+53% engagement

+53% engagement

Thank you

Thank you

Thank you

let's connect &

let's connect &

Discuss in detail

Discuss in detail