Nine sources scraped daily. Every signal classified at the neighborhood level — East Nashville vs. The Gulch vs. Germantown, not just "Nashville." Delivered to the people signing the leases.
Automated crawlers pull signals from 9 non-traditional sources — TikTok, Reddit, Google Reviews, Yelp, Indeed job postings, ClassPass capacity, LoopNet leases, Google Trends, and Discord. Every complaint, praise, and competitive mention — captured.
GPT-4o-mini runs each signal through a six-category taxonomy: pain points, satisfaction, competitive shifts, unmet demand, expansion signals, and market trends. No human reads 600 Reddit posts. The model does.
The Brief lands in your inbox with the recommendations on top, the supporting signals underneath, and source links to every quote. When something genuinely breaks mid-week — a closure, a competitive entry — we send a one-off Market Event email. Otherwise the cadence is predictable. You get the signal, not the alert fatigue.
Complaint videos, gym tours, brand mentions, viral sentiment
3 scraping methods — PRAW, Google CSE, Arctic Shift archive
Star ratings, review text, response patterns, velocity
Review trends, rating velocity, competitive positioning
Hiring surges, role types, expansion signals before press releases
Capacity data, class fill rates, pricing shifts
Commercial leases by corridor — who's expanding, who's vacating
Search interest over time, rising queries, demand signals
Community channels, insider sentiment, staff complaints