LeadOps Autopilot: Sourcing and Scoring US Clinic Leads from Public Data

Описание и применение

LeadOps Autopilot – система автоматизированного поиска и оценки клиник США, использующая публичные данные из NPPES и OpenStreetMap, оценивающая их по веб‑сайтам и выдающая прозрачные причины оценки, после чего клиенты вручную одобряют контакты.

Задача

Clinic lead lists are often stale, incomplete, or irrelevant because they only confirm existence without indicating operational fit for lead‑intake automation.

Результат

The system provides provenance‑rich, score‑based clinic leads with transparent reasoning, enabling cheap iteration, compliance by construction, and visible failure modes, while human approval gates outbound outreach.

Рабочий процесс

• Collect clinic data from CMS/NPPES and OpenStreetMap Overpass API
• Enrich with website discovery via Google Places API
• Fetch each clinic’s homepage and score based on fit and negative signals
• Generate pain hypothesis and draft outreach copy
• Segment leads by vertical and batch by fit tier
• Human reviews CSV and approves high‑fit rows before any outreach is sent

Доказательства

Most lead lists are bought, stale, and sprayed. Ours is none of the three — and the interesting part is not the scraping.

Источник

The scorer (<code>leadops.py</code>) fetches each clinic's homepage and scores what it finds there.

Источник

The MVP stops before sending.

Источник