Multi-tenant CRM with realtime chat and M-Pesa billing for East Africa.
Edge360 consolidates 4–6 disconnected tools (spreadsheets, email, Slack, separate CRM/billing/support) into a single workspace for B2B service organizations. A multi-tenant platform pairing a React 18 SPA with a Node.js/Express + PostgreSQL backend, layering Socket.IO realtime chat and M-Pesa STK Push billing for the East African market.
- React 18
- Node.js · Express
- PostgreSQL
- Socket.IO
- M-Pesa Daraja
- JWT
- CryptoJS / bcrypt
How we framed it.
Multi-tenancy is enforced at the application layer through organization_id scoping — every prospect, relationship, message, and ticket is filtered by the JWT-resolved org context. A single Postgres database serves every tenant without per-tenant schemas or row-level security, keeping infrastructure simple while maintaining strict logical separation.
Inside the build.
Application-layer multi-tenant isolation
Every prospect, relationship, message, target, and ticket is scoped by organization_id. Single Postgres database, isolation enforced at the controller layer — inexpensive infrastructure, strict logical separation, no per-tenant schemas required.
Realtime chat with room targeting
Socket.IO mounted on the same Express HTTP server. (userId, prospectId | relationshipId) tuples key the in-memory presence map; sendMessage dispatches either to a room (broadcast minus sender) or directly to receiver_id (DM).
M-Pesa STK Push pipeline
Daraja OAuth token → STK push → callback persistence → client polling. Raw callback JSON is written to disk before processing — invaluable for debugging payment edge cases and reconciling disputed transactions in a market underserved by Stripe-centric SaaS billing.
Encrypted-transit authentication
Clients AES-encrypt the password before transit using a shared key; the server decrypts then bcrypt-hashes for storage. JWTs in localStorage, per-user single-active-session lock prevents silent account sharing.
Per-tenant pipeline configuration
Each organization stores its own milestones, happiness rating choices, lead sources, services, industries, and currency as Postgres text/jsonb arrays. Tenants customize the entire CRM vocabulary without schema migrations.
What shipped.
- Consolidated 4–6 disconnected tools into one workspace
- Multi-tenant CRM serving every customer from a single Postgres database
- M-Pesa STK Push subscription billing for the Kenyan / East African market
- Per-tenant vocabulary customization without backend deployments