RenovateAI

Lead Finder

Inspiration

We were inspired by a simple problem: contractors still waste huge amounts of time and money marketing to the wrong homes. Most home service businesses rely on broad mailers, generic lead lists, or word of mouth, even though there is a huge amount of public property, permit, and infrastructure data that could be used to target homes much more intelligently. We wanted to build a platform that turns messy public records into actionable lead intelligence for plumbers, basement remodelers, drainage contractors, landscapers, and other home service businesses.

What it Does

RenovateAI helps contractors identify which homes are most likely to need their services. Instead of marketing to everyone, users can filter and rank properties based on signals like address location, permit history, service-line context, assessed value, property characteristics, and trade-specific lead scores. The app includes a contractor-aware dashboard, a map-based lead explorer, detailed property pages, campaign tools, and an AI search layer that lets users describe their ideal lead in plain English.

How We Built it

We built RenovateAI as a Dockerized full-stack application with three main services: a PostgreSQL database, a FastAPI backend, and a React + Vite + Tailwind frontend. The frontend was rebuilt into a contractor-aware SaaS-style experience with tailored workflows for multiple trades. On the backend, we created property scoring logic, authentication, campaigns, notes, AI parsing, and admin tooling. We also integrated real-data enrichment by pulling from public sources like the US Census batch geocoder, the City of Evanston open data portal for permit history, and Cook County Assessor datasets for property characteristics and assessed values. The result is a platform that blends public data, scoring logic, and a much more polished contractor-focused UI.

Challenges We Ran Into

One of the biggest challenges was data integrity. Early on, the project relied too heavily on demo-style seeded data, which made the product feel less trustworthy. We had to rework the backend to replace fake addresses, improve geocoding, and make sure only Census-confirmed properties were being seeded. We also ran into issues with Docker setup, frontend architecture, stale environment variables, and Gemini model compatibility, since some older Gemini defaults no longer worked. On top of that, public municipal and assessor datasets are messy, inconsistent, and often change field names or dataset IDs, so we had to build multiple fallback paths just to make the data ingestion more reliable.

Accomplishments We’re Proud Of

We’re proud that RenovateAI evolved from a rough prototype into a real, usable contractor lead-intelligence platform. The frontend is now a polished, contractor-aware web app instead of a static demo page. We also pushed the project away from clearly fake demo data toward real Evanston addresses, Census geocoding, real permit ingestion, and Cook County property enrichment. Another accomplishment is that we kept the entire app Docker-runnable and demoable, while still supporting advanced features like AI lead search, campaign management, and property-level score explanations.

What We Learned

We learned that building around real-world data is much harder than building around a clean demo dataset, but it is also much more valuable. Public data is messy, incomplete, and inconsistent, so designing graceful fallbacks and clear provenance matters a lot. We also learned how important product polish is: the same backend logic feels dramatically more compelling when paired with an intuitive, contractor-specific UI. Finally, we learned that for a tool like this, trust is everything. If the addresses, data, or explanations feel fake, the entire product loses credibility.

What’s Next for RenovateAI

The next step is moving fully away from demo-style data and toward a production-grade ingestion pipeline. That means replacing remaining seeded or estimated fields with cleaner live-source-backed data, expanding beyond Evanston into more municipalities, and making provenance even more transparent at the property level. We also want to improve campaign execution, enrich service-specific scoring models, and make the AI assistant more tightly integrated into the lead-exploration workflow. Long term, RenovateAI would become a true contractor growth platform: not just identifying the best homes to target, but helping businesses turn those opportunities into real revenue