Keller Williams Realty International (KWRI) · Case study
Listing Factory and Market Brief
I built two early prototypes to reduce repetitive agent work: Listing Factory for preparing a listing package and Market Brief for daily market insights and social content. I tested them with agents and supplied the builds and guides for technical review.
Corporate headquarters · Industry & Learning. Research, prototypes, and enablement for the Keller Williams network of agents, Market Centers, and regional teams.
The starting point
Preparing listing materials and turning market information into useful content required repeated manual work. I investigated how a small set of agent inputs could drive those outputs.
My responsibility
I designed and built both prototypes, refined them using Claude and Gemini, tested with agents referred by Jason Abrams, and prepared the use cases, setup instructions, and one-pagers for LABS.
Tools & methods
Google Sheets · Apps Script · Drive · Apple Shortcuts · Model APIs · Rates API
Listing Factory: automate the preparation package
I built an intake-driven workflow with Google Sheets, Apps Script, and Drive. Listing information triggered the creation of folders, copied and prefilled templates, a checklist, returned links, and an agent email.
The package included listing copy, social captions, an open house plan, and AI prompts. I added readiness checks and duplicate prevention, then refined the setup with Claude and Gemini and tested it with agents.
Market Brief: combine insights with ready-to-use content
I built a separate Apple Shortcuts prototype using inputs such as ZIP code and niche. It brought together local insights, national housing updates, and mortgage rates, then supplied social scripts, hooks, ideas, and copy-and-paste posts.
I tested direct rate retrieval through an API and explored versions using Claude, Gemini, and ChatGPT APIs to find information. Every agent referred to test Market Brief gave positive feedback and requested continued access.
Testing demand and preparing the handoff
After testing and refinement, I presented the use cases, repeatable setup, and one-pagers to LABS. The team requested the build materials and noted that it was pursuing similar work.
Market Brief still needed support to make account connections and Shortcut setup repeatable for wider distribution. I supplied the workflow and materials, but that distribution dependency remained unresolved.
Outcome
Working prototypes; wider distribution unresolved
Two working prototypes were tested with intended users and handed to LABS with supporting guidance. Market Brief established demand for continued access, but neither prototype advanced to a confirmed broad release.
Let’s connect
What could work better?
I’m interested in AI innovation and solutions roles where research becomes useful systems.