Revolutionizing Car Shopping with Generative AI

Project image

Challenge

  • The owner of the car marketplace wanted to improve service and reduce the search time for customers struggling to find the best car deal in a vast marketplace with multiple manual filters.

Results

  • A 25% increase in average customer screen time.
  • 15% more users made a purchase using smart search than normal search
  • AI search web and mobile apps attracted ~10,000 more users than legacy filter search platforms

Implementation Details

  • Our team developed a generative AI search engine with self-learning capabilities and integrated the developments into the client’s existing web and mobile platforms, extending them with the brand-new intelligent search feature. The AI-powered search efficiently screened through countless listings, considering factors like make, model, price, location, etc. The feedback loop allowed the AI to continuously learn from user interactions and improve its results over time. We further integrated the solution into web and mobile marketplaces for easy accessibility to a wide range of users.
  • To evaluate the effectiveness of the new intelligent search system in the real world, we deployed it along with the previous-generation classic search system using filters in A/B testing mode. In 3 months of observation, we’ve found that the AI-powered car deal finder remarkably overperformed the legacy system.

Industry

Service

Type

  • Case Study

Keywords

  • AI
  • LLM
  • GenAI
  • Generative AI Use Cases
  • Artificial Intelligence
Roadmap
Generative AI search engine development
8 weeks
Data integration
2 weeks
Web and mobile app integration, testing, and bugfix
4 weeks
Feedback loop and self-learning features development and integration
3 weeks
User acceptance and iterative refinement of corner cases
3 weeks
A/B deployment and observation
3 months (passive)
Final release

Sign up to receive the project description

    Want to talk?

    Michael Gurbych Director,
    Operations and Finance
    Book a meeting
    Roadmap
    Generative AI search engine development
    8 weeks
    Data integration
    2 weeks
    Web and mobile app integration, testing, and bugfix
    4 weeks
    Feedback loop and self-learning features development and integration
    3 weeks
    User acceptance and iterative refinement of corner cases
    3 weeks
    A/B deployment and observation
    3 months (passive)
    Final release