Designing AI-powered search for millions
When Zillow, Homes.com, and Jitty launched AI search, Realtor.com had eight weeks to respond. I led a cross-platform sprint that rebuilt fragmented filters, enabled natural-language search, and laid the foundation for AI-powered discovery. Within two weeks of launch, traffic from search results to listings grew by 12%.
Company
Realtor.com
Year
2025
Role
Lead Product Designer
Timeline
8 weeks
01 CHALLENGE
Broken filters, frustrated users
Our filters hadn’t evolved in years. They were cluttered, inconsistent, and barely used. Analysis of 93 million monthly sessions showed that only 13% of users applied filters beyond location. Most scrolled through irrelevant results instead of refining their search.
The problem wasn’t the data. It was fear. People worried about filtering themselves out of a good home, so they avoided filters altogether.
Core issues:
Only 13% of users applied filters
40+ rarely used filters caused decision fatigue
Naming and interaction patterns varied across platforms
No way to express “soft” preferences like prefer a pool
Advanced filters used by less than 1% cluttered the UI
02 RESEARCH
Finding the signal in the noise
I analyzed filter usage across platforms and partnered with research to understand behavior. The insight was clear: users didn’t want more filters, they wanted smarter ones. People wanted to search the way they talked to a realtor—natural, conversational, and flexible.
Competitive Landscape
Across 13 competitors, including Zillow, Redfin, and Jitty, the best experiences favored clarity and progressive disclosure. None relied on large filter panels. Instead, they used lightweight prompts that adapted to intent.
Cross-Platform Audit
I documented 40+ filters across iOS, Android, and web, noting duplication and inconsistent hierarchy. Core filters like Property Type had 63% usage, while niche ones like Commute Distance were nearly unused but still surfaced equally.
03 SOLUTION
Building an AI-ready foundation
I brought product, engineering, research, and design systems together for a one-day workshop to align on goals. We defined seven guiding principles that shaped every decision.
Guiding Principles
Create focus and de-clutter
Prioritize what's most useful
Improve hierarchy of filter groups
Make filters easier to scan and understand
Use controls that match intent
Achieve platform parity
Keep patterns consistent
Unified Taxonomy
We rebuilt the structure from the ground up, consolidating categories and removing redundancies. Duplicates like Waterfront and Lake View merged into one. The new taxonomy allowed AI to map conversational queries like “three-bedroom home with a pool” directly to filters.
Redesigned Interface
We replaced checkboxes with chip-style controls for faster scanning and clearer states. Multi-select support for listing status improved flexibility. Simplified bed and bath filters reflected how people naturally search. Every change was designed with accessibility and scalability in mind.
Scalable Systems
I partnered with the design systems team to evolve shared components and extend the token library. Colors, spacing, and typography were standardized to ensure parity across iOS, Android, and web.
04 impact
A foundation for AI-powered discovery
In just eight weeks, we delivered a redesigned search foundation that scaled across platforms and powered Realtor.com’s first AI-driven search.
05 REFLECTION
From urgency to clarity
This project taught me how to turn urgency into clarity. Under pressure to ship fast, I focused on fixing the foundation instead of patching problems. The system we built now supports AI features that will continue to evolve for years to come.