Transforming user experience in cars-as-a-service industry through Strategic AI/ML Integration — a UX case study.

As I delve deeper into understanding the capabilities and limitations of Artificial Intelligence, I see an opportunity for AI/ML to improve an existing flow in the Automotive industry.

Karena E. I
UX Planet
Published in
6 min readDec 6, 2024

--

Image Credit: Karena E. I

Overview

This case study focuses on integrating AI/ML to improve user experience in the car-as-a-service automobile marketplace. I dive further into the following:

  • Introduction and the current challenge.
  • Research and Data Insight
  • Benefit of AI/ML integration
  • AI strategy
  • Design of the suggested flow
  • Building trust in the AI system

CaaS Platforms are essentially E-commerce for Automobiles

Cars-as-a-service (CaaS) evolved and keeps evolving to meet the need of individuals who need cars to convey them from one point to the other comfortably but cannot afford to buy new or even old cars. So they look for automobile rental firms to suit this need.

CaaS platforms function similarly to e-commerce marketplaces, but for automobiles. These digital marketplaces act as intermediaries between car dealers and buyers, offering a flexible and comprehensive car shopping experience and providing customers with an extensive selection of cars across different years, models, and makes.

The key advantages of these platforms include:

  • Diverse vehicle selection from multiple dealers
  • Easy comparison of different car models and prices
  • Transparent availability information
  • Flexible and Streamlined rental process designed to meet user needs

Current challenges for Businesses in CaaS

The rental marketplace faces the nuanced challenge of balancing individual expectations with practical availability, creating a complex matching problem that requires sophisticated solution strategies.

Prompt samples based on real Data on how customers source for cars in rental marketplaces.

Users approach car rentals with a wide range of preferences and constraints:

Budget Considerations: Some customers have strict financial limitations. Pricing can significantly influence rental decisions.

Car Preferences: Customers range from highly specific to completely flexible. Some users desire precise make and model specifications. Many seek particular features at price points not readily available in the market

User Motivational Spectrum: Some individuals have crystal-clear transportation requirements. Others are uncertain about their exact needs. Many users end up compromising, selecting from limited available options

In some cases, users may abandon their rental search if no suitable match exists.

Prompt samples based on real Data on how customers source for cars in rental marketplaces.

Data and Research Insight

  • Data shows that customers factor in budget and needs when sourcing for a car.
  • Customers are mostly flexible with their car preferences due to the nature of the market place.
  • Customers have the opportunity to easily compare different car models and prices.

By shortening the time to find a car and ensuring that recommended vehicles match user expectations or needs, rental rates could potentially increase, leading to more conversions.

AI/ML integration provides an avenue for value added innovation, a more personalized and enhanced user experience for customers.

Enter CarFY AI,

CarFY AI is an AI-powered system for streamlining car discovery in Car-as-a-Service platforms, providing a personalized and enhanced user experience for customers.

Image credit: Karena E. I

Key potential benefits include:

  • Reducing time to find a car by over 90%
  • Minimizing scrolling through the marketplace
  • Providing recommendations fine-tuned to users’ specific motivations for needing a car

AI Strategy

One critical component of building any AI product is having a strategy on how the product will be developed or making important decisions from the get-go to ensure that your AI meets responsible AI frameworks such as privacy protection of user data through the collection of data process, avoiding reinforcing biases in the training or fine-tuning process of your AI system.

Role of AI: Recommendations and Personalized content is one area where AI augmentation has proven effective in many industries. I intend to use it to improve the digital user experience of car sourcing in the CaaS automotive sector.

AI Vision: Intelligently match customers with their ideal car through a sophisticated, single-prompt recommendation engine.

Goals for AI/ML Integration

  1. Competitive Differentiation: Leveraging advanced AI/ML capabilities to create a distinctive market position that sets our solution apart from traditional automotive sales approaches.
  2. Hyper-Personalized User Experience: Utilizing intelligent algorithms to deliver precision-targeted vehicle recommendations and personalization to transform customer interactions from transactional to deeply contextual.
  3. Value Proposition Enhancement: Develop a sophisticated recommendation ecosystem demonstrating an exceptional understanding of customer needs, preferences, and potential future requirements.
  4. Sales Funnel Optimization: Implement data-driven AI strategies to streamline the customer journey, reduce friction points, and increase conversion rates through intelligent, adaptive engagement mechanisms.
Framework for Building AI systems

Machine Learning

Machine Learning (ML) is utilized for data-driven predictions. ML transforms raw data into meaningful, context-aware insights. This adaptive technology continuously refines its performance through expanded data exposure and sophisticated learning mechanisms enabling it to digest complex information, recognize intricate patterns, and uncover critical anomalies.

ML techniques have been utilized by organizations like Netflix and Spotify to revolutionize user experiences, delivering hyper-personalized recommendations that feel intuitively crafted for individual preferences.

For CarFY, our ML algorithms represent a dynamic intelligence recommendation system that doesn’t just suggest — it understands, anticipates, and personalizes with unprecedented accuracy. By processing vast datasets and accessing car catalog from diverse dealers, the algorithms can generate recommendations that range from platform-wide suggestions to granular, customer-specific suggestions.

The system’s strength lies in its ability to continuously learn, adapt, and evolve, ensuring that recommendations become progressively more precise and relevant.

This approach is anchored in robust ethical AI frameworks, balancing technological potential with responsible implementation.

Data Collection

ML uses data to continually learn and offer more accurate results based on the specific use case of your AI system. Defining what data is required and collected at the preliminary stage of AI product development is crucial as it influences the effectiveness of recommendations, and how we communicate to our users to build trust in our AI system.

User Data such as — age range, gender, motivations, budget are required to ensure effectiveness of AI recommendations.

Input-output mapping

For our AI to produce reliable output, the algorithm has to receive useful input.

Building trust in the AI system

To build trust in any AI system, users need to understand How the AI works and have flexible control of the AI output to accept, decline, ignore recommendations.

Mental Models: I’m utilizing existing mental models that customer are already used to like Claude, Chatgpt for the interface design. It’s simple and straight to the point.

Handling Erros

AI system are probabilistic in nature and are never 100% correct. To ensure that errors are handled correctly in a way that allows the user to recalibrate their expectation of the AI system, control it’s output or ignore it, the error feedback has to be clear and explanatory.

Errors are opportunities for building trust. There are multiple types of error but I’ll focus on user errors or user perceived errors since the stakes are not very high and I’m focused on user (customer) experience optimization.

Handling errors
Adding User controls

--

--

Published in UX Planet

UX Planet is a one-stop resource for everything related to user experience.

Written by Karena E. I

Written by Karena | AI Product designer sharing tidbits at the intersection of strategy, design and innovation.

Responses (3)