Autotrader transforms the search experience with AI-powered categories
To help car buyers navigate an increasingly complex automotive market, Autotrader has launched a revolutionary new AI-driven search experience. The all-new ‘I’m looking for’ categories allow buyers to refine their journey using natural, real-world descriptions, making it easier than ever to locate their perfect match among the c. 423,000 vehicles advertised daily on the platform.
The new feature tackles a major industry shift: a flood of new brands has increased complexity, leaving almost half of UK buyers[1]beginning their journey without a specific make or model in mind, whilst one in four (24) [2] struggle to find the right information. Compounding the challenge, today's consumers increasingly express what they want in natural language, rather than relying on unfamiliar industry-defined terminology or the increasingly complex alphanumerical soupe of new model names. However, mapping these subjective queries to traditional filters is notoriously difficult; for example, defining a ‘sports car’ using standard descriptors like 'coupe or convertible' and '0-60 under 6 seconds' will surface everything from a Vauxhall Astra to a Rolls-Royce Phantom.
To solve this disconnect, Autotrader harnessed its decade-long investment in data science and AI to create 13 new search categories. Instead of relying on rigid keyword mapping, the platform uses Machine Learning Models and a dynamic data science-driven scoring system to accurately categorise vehicles in ways that naturally meet buyer expectations. This enables consumers to easily search for exactly what they need, even if they don’t know exactly what they want. By searching with broader ideas - whether that’s a 'first car' or a 'sports car' — the machine learning surfaces relevant cars they wouldn’t have necessarily found otherwise, while still allowing them to combine these intuitive categories with practical, traditional filters like maximum price and mileage.
With circa 84 million visits every month, this smart taxonomy is built on rigorous research of consumer search behaviour across the Autotrader platform, combined with additional Google search data, keyword insights, direct user testing, and Autotrader’s rich vehicle insights. The AI-powered approach also allowed for the creation of enthusiast-focused categories like 'OMG'. This specific feature has sparked massive engagement across automotive communities, with users on motoring forums citing it as an "excellent filter" and that they "love the OMG option".
For retailers, these AI-powered categories do more than just improve the browsing experience. By moving away from fixed filters, the platform’s detailed vehicle taxonomy and deep understanding of buyer behaviour guide consumers through market complexity directly to their perfect car. Crucially, this surfaces retailers less familiar stock to a wider audience of highly engaged, high-intent buyers who might not have discovered them organically.
With huge increase in choice entering the market, in terms of make, model, and fuel-type, the early stages of car buying can be overwhelming. We aren't just giving consumers new buttons to click; we’re leveraging our deep data infrastructure and AI capabilities to fundamentally change how vehicles are discovered. By intelligently matching stock to the natural language buyers actually use, we are delivering a significantly smarter, more enjoyable search experience that connects our retailer partners' stock with more informed and high-intent buyers.
Available online and on the app, ‘I’m looking for’ also allows car buyers to search by popular vehicle types (e.g. ‘Sports cars’, ‘Small cars’, and ‘Luxury cars’), as well as practical lifestyle (e.g. ‘Big boot’, ‘Family cars’, and ‘First cars’), and more economy focused categories (e.g. ‘Cheaper to run’, and ‘Cheaper insurance’).
[1] Autotrader data: Car Buyers Tracker April 2025, 2,000 respondents. 43% don’t know the exact make and model of the car they want at the start of their car buying journey.
[2] Autotrader data. Drivers Choice Awards 2025. 226,980 respondents.
Follow us on our social channels to keep up with the latest news, insights and company developments.
If you would like to be added to our News & views mailing list, or have any media-related enquiries, please contact our press team:
Looking to join the team?
Looking to make a sale?
Looking to make a purchase?
Sign up to our email alerts service:
Join our news & views mailing list
or submit media-related enquiries:
