AI SEARCH
PROJECT OVERVIEW
The AI Search initiative at Dell was designed to drive both immediate business and customer value while building insights for long-term improvement. By integrating generative AI and LLM capabilities, the goal was to transform search into a responsive, conversational experience that informs, engages, and guides users through complex purchase decisions.
The goal is to provide more than accurate results, but to simplify interactions, maintains transparency, and incorporates continuous feedback.
Throughout the project, we ensured that the design reflects Dell’s brand voice, remains consistent and familiar across devices and platforms, and is accessible to all users, creating a seamless, intuitive experience that meets modern digital expectations.
MY ROLE
I co-led the UX design direction for this project, collaborating with product management, engineering, and data science teams to align user needs with AI capabilities.
My responsibilities included shaping the user journey, defining interaction patterns, and ensuring the interface delivers clarity and confidence even when AI-driven responses are involved.
I helped to bridge the gap between generative AI functionality and user-centered design principles in search toward a solution that balances innovation with usability, enabling users to discover, compare, and select products efficiently while feeling supported throughout the journey.
THE CHALLENGE
Dell’s existing search system presented multiple challenges. Users often encountered static, linear flows that pushed them through disjointed pages, creating confusion and frustration. Vague or imprecise queries led to incomplete or inaccurate results, and customers struggled to match product terminology with what was displayed.
They needed help identifying products for specific scenarios, understanding features and benefits, and narrowing down options in a way that felt manageable. Without smart recommendations or guidance on compatibility, users often experienced decision fatigue, abandoned searches, or missed opportunities to explore add-ons and bundles.
The challenge was not simply a visual redesign but a fundamental rethink of search, turning it into an adaptive, intelligent experience that anticipates user intent, supports exploration, and reduces friction across the purchase journey.
RESEARCH
To inform our approach, we conducted extensive competitive analysis across e-commerce platforms like Amazon and Walmart, as well as emerging conversational AI tools such as Perplexity, Kayak, and Pinterest.
We explored how dynamic responses, personalization, and intent recognition can optimize the search experience. By examining session-based learning, user profile integration, and behavioral cues like click-through and dwell time, we identified strategies for refining recommendations in real-time.
We also considered multimodal interactions - including voice and image-based search - to address evolving user expectations. This research highlighted the importance of building search experiences that are iterative, adaptive, and context-aware, allowing AI to dynamically guide users from initial inquiry to confident purchasing decisions while continuously improving through A/B testing and analytics.
KEY INSIGHTS
Our analysis revealed that leading e-commerce platforms are increasingly leveraging AI to make search more personalized, context-aware, and visually rich.
Users respond best when search understands natural language queries, incorporates past behavior and preferences, and provides clear reasoning for recommendations. Visual search, conversational interfaces, and persistent page-aware prompting have become critical for discovery, comparison, and decision-making.
For example, tools like Amazon Lens and Pinterest Lens demonstrate the impact of image-based search and personalization on engagement and conversion, while conversational assistants like Kayak’s “Ask Kayak” showcase how real-time filtering and guidance improve search efficiency. In an enterprise context like Dell, providing clarity, guidance, and transparency is essential to reduce friction in complex journeys.
These insights shaped a search experience that balances generative AI capabilities with structured product data, delivering a system that feels intelligent, trustworthy, and intuitive.
FRAMING
The core goal was to make search journeys efficient, accurate, and enjoyable while addressing every step of the buying process. Our mission was to create helpful, proactive guidance that feels approachable, always solution-oriented, and enhances the overall shopping experience.
By detecting user intent, funnel stage, and context based on queries and behavior, the system dynamically adapts responses to guide users toward relevant products and information. This approach ensures a smooth, informed journey from exploration to purchase, fostering confidence and building a meaningful relationship between the customer and the brand.
Create a foundation that improves the current experience and creates learning for future iterations.
DESIGN STRATEGY
This design strategy is a work in progress and contains confidential elements that cannot be shared publicly. While certain details remain private, the approach has been carefully developed to focus on clarity, efficiency, and adaptability. Interaction patterns were crafted to feel familiar yet streamlined, guiding users toward key actions such as filtering, comparing, and selecting products without creating cognitive overload or confusion.
Emphasis was placed on designing flows that anticipate user intent, reduce friction, and provide visual and functional cues that support decision-making. Even in its current confidential state, the strategy demonstrates a user-centered philosophy that balances AI-driven capabilities with clear, accessible design principles, laying the groundwork for an intelligent and seamless search experience.
