Flow AI cuts realtor outreach from hours to seconds with n8n

How Flow AI built a voice‑driven AI outreach engine for real estate agents, replacing manual follow‑up and scattered tools with compliant, automated campaigns.

Back to Case Studies

Background

Flow AI Inc. is a real estate technology startup founded by Dharamjeet Patel, a former software engineer who transitioned into real estate and quickly discovered how outdated industry processes were slowing agents down.

On a typical day, agents were expected to spend three to five hours on follow up, including cold calling, mass emails, and manual texting of past leads. Rather than spending most of her day behind a desk, Dharamjeet focused on in person interactions with walk-in leads. This approach helped her to become the producing agent nationwide for one of the largest US homebuilders.

The insight was straightforward. Time is the real bottleneck in real estate sales. As Dharamjeet explains, “On a normal day, agents sit for three to five hours just doing follow up. Time is money, and most of them are losing both.”

Flow AI was created to solve this problem through Flow AI, a natural language, voice driven assistant that allows agents to launch large scale outreach campaigns and follow up with customers in seconds.

Initially, Dharamjeet stitched together Zapier, a free Twilio number, and Mailchimp for personal use. While the system was fragile and disconnected, even basic automation was enough to outperform peers still working manually. Turning this setup into a product required a scalable automation foundation.

That foundation became n8n.

Challenge

Flow AI set out to eliminate time consuming, non compliant outreach while preparing its infrastructure to operate as a commercial software product.

Agents were expected to contact hundreds of past leads every week while complying with strict US regulations such as 10DLC and CAN SPAM. In practice, inviting an entire database to an open house could take several days or require hiring a full time assistant at around 50,000 dollars per year.

According to Dharamjeet, the original stack of Zapier, Twilio, and Mailchimp was not suitable for a product business. “Tools were siloed, opt-in status was difficult to enforce consistently, and there was no strong isolation between customers or centralized logging for audits”.

Flow AI needed a single platform that could orchestrate multichannel workflows, isolate each customer’s data, integrate voice, messaging, and email, log every interaction for compliance, and scale from a few workflows to hundreds of customer projects.

Solution

After evaluating multiple options, Flow AI chose n8n Enterprise, self hosted, as the backbone for Flow AI.

The team started on the Community edition to validate flexibility, and upgraded to Enterprise to unlock unlimited projects and full control over hosting. This allowed Flow AI to own its infrastructure and automate at scale.

The architecture is built around a simple principle. Each customer runs in their own dedicated n8n project. This guarantees strict data isolation, prevents accidental cross sends, and allows custom workflows per customer without affecting others.

"Every customer of ours has their own project,” explains Dharamjeet. “We can customize workflows without impacting anyone else. Our entire company relies on n8n."

Flow AI sits on top of this foundation as a voice first experience. An agent can say, “I am hosting an open house next Saturday, invite everyone in my contacts,” and the system takes over. ElevenLabs converts speech into structured input and sends it to n8n via webhook.

flow-ai-n8n-nodes.png

Inside n8n, workflows query the Postgres database, verify opt-in status and compliance, generate personalized content using AI, apply safeguards for 10DLC and CAN SPAM, and route messages through SMS providers or Mailgun. Every step is logged, creating a complete audit trail. End to end execution typically completes in under 60 seconds.

As the system matured, Flow AI moved from large monolithic workflows to smaller modular workflows, improving reliability and maintainability.

Beyond the product itself, Flow AI also uses n8n for internal operations such as investor updates. Any repeatable communication task is handled through a workflow rather than manual effort.

_"n8n is a tool that is very easy to step into, but once you master it, it makes your product feel priceless. It gave us the flexibility to customize everything and still keep it fast and reliable." Joshua Weatherspoon, Lead Engineer, Flow AI Inc.

Impact

By moving its outreach engine onto n8n, Flow AI transformed both development speed and the day to day workflow for agents.

Campaigns that once took days now run in under a minute. Tasks that previously required a full time assistant can now be executed through a single voice command. Agents no longer need to manage CRMs or templates to stay compliant.

Compliance is built into every workflow, with full logs available for audits. Internally, Flow AI went from concept to a working product in roughly one month and onboarded its first users within a few months.

Most importantly, the product addresses the industry’s core constraint. As Dharamjeet summarizes, “We save hundreds of hours. Time is the biggest problem in real estate, and n8n is what allowed our users to be more present with walk-in clients while still following up with past leads.”

“If you asked me to send 500 emails and 500 text messages manually, that would take me a week or I would need to hire an assistant. Now I just talk to the AI agent and everyone gets the message within about 60 seconds.”

Dharamjeet Kaur Patel

Founder, Flow AI