The Referral System / Q3 2025
A Multi-Partner Referral Platform With CRM Routing.
Referrals in, partners matched, CRM updated. All automated.
// The Problem
The client's referral intake was entirely manual. An external office would send a referral, someone would enter it into the CRM by hand, match the lead to the right service partner, send notifications, and track the outcome. Every step was a place where referrals got lost or delayed.
// The Build
I built a web-based referral platform running on a small set of AWS Lambda functions with a modern frontend. The workflow: a CRM automation triggers the initial referral, which hits the lead-intake Lambda. The lead receives an encrypted link via SMS and email.
The lead follows the link to a multi-step intake flow: short-lived identity verification (rate-limited to contain brute force), demographics review pre-filled from the CRM, agreement review that branches by product, and payment setup for collateral capture without charging up front.
Short-lived session tokens. Confirmation links expire after a week. API keys stay server-side through server-rendered actions and are never exposed to the browser. Every step creates or updates CRM records through isolated custom fields that prevent unintended downstream automations.
The pipeline is monitored: dead-letter queues, Lambda error alarms, and API availability checks.
// outcome
Referral routing fully automated. Manual intake process eliminated. CRM records created on submission.
auto
Partner matching
0
Manual routing steps
real-time
CRM sync
Multiple
Partner integrations
// NDA note
This project was completed under NDA. Full narrative with technical detail, trade-offs, and sanitized artifacts available under your own NDA. Contact directly to request the long form.
Got a similar problem? Let’s talk.