INTERNAL TOOL

A flaky beta CRM stabilized and re-architected into a multi-tenant SaaS for brokerages.

MaxOps AI project hero
Industry
Real Estate
Timeline
~8 weeks
Stack
Softr · Airtable · Zapier · Make.com · CustomJS · Typeform · Calendly · Gmail / Outlook OAuth · Apify
01 · Problem

The problem

The beta platform had critical Softr ↔ Airtable sync issues — Kanban pipeline reverting, lead-status inconsistencies, and data overwrites — plus unreliable automations and an architecture that couldn't scale beyond a single account.

It wasn't production-ready.

02 · Solution

What we built

We stabilized the system and re-architected it into a scalable, multi-tenant SaaS platform for brokerages.

Agents, brokerage admins (tenants), and a platform admin each see only their authorized data on a shared infrastructure.

  • Fixed Softr ↔ Airtable synchronization: corrected field mappings and write-back logic, and separated automation-controlled fields from manual edits to stop overwrites and Kanban reversion.
  • Multi-tenant architecture letting many brokerages and agents share one infrastructure with role-based data isolation.
  • Secure OAuth email for agents (Gmail & Outlook) — send/receive without sharing credentials.
  • Calendly scheduling with webhook-triggered pipeline updates.
  • Round-robin lead assignment for even distribution across agents.
  • Lead-generation pipelines via scraping tools and Typeform feeding structured Airtable data, plus a documented future-state architecture.
MaxOps AI detail image 1
MaxOps AI detail image 2
03 · Process

How we built it

We restructured the Airtable relationships and Softr filters so each tenant sees only their own data, implemented role-based access control, and corrected the sync logic causing overwrites.

OAuth integrations connected agent inboxes securely; Calendly webhooks updated the CRM pipeline automatically; and Zapier / Make handled reliable triggers, Typeform-to-Airtable mapping, and round-robin assignment. A full system-architecture document mapped data flow from capture through nurture, scheduling, and conversion.

04 · Results

What changed after launch

Zero
Data-overwrite / Kanban-reversion errors
Multi-tenant
Many brokerages on shared infrastructure
Per-seat SaaS
Scalable revenue model unlocked
Let's build it

Want something like this?

Tell us what you're building. First conversation is on us. Fixed quote in 24 hours, no NDA needed.

Or browse pricing
Reply within 24 hoursNo NDA requiredHonest fit recommendation
More work

Other projects you might like