> ## Content Index
> Fetch the complete content index at: https://www.backyardbroker.blog/llms.txt
> Use this file to discover other available public pages before exploring further.

# How TMG The Mortgage Group Became Canada's Largest Independent Brokerage by Letting Agents Leave
- URL: https://www.backyardbroker.blog/how-tmg-the-mortgage-group-became-canadas-largest-independent-brokerage-by-letting-agents-leave/
- Published: 2026-08-23T03:30:21.000Z
- Updated: 2026-08-23T03:30:26.000Z
- Author: Grant Thomas
- Tags: #partner-article

TMG The Mortgage Group now employs over 1,700 mortgage professionals across Canada, making it the country's largest independent brokerage.

That model emerged from economic realism in a market where margins were being ground down from multiple directions.

## The Trap Most Brokerages Build

Traditional brokerage structures treat agent churn as an existential threat. The response is typically contractual: long notice periods, non-competes, database ownership clauses, and a thicket of fees designed to extract revenue even when production drops. The logic is defensible on paper. Agents are the revenue engine, so protecting against their departure protects the business.

The structural flaw is that protection becomes capture. When leaving is expensive and procedurally punitive, staying becomes the path of least resistance even when the fit has deteriorated. The brokerage wins short-term retention at the cost of long-term trust. The agents who remain are the ones least able to leave--the ones facing the highest switching costs, not the ones most committed to the platform.

Grant and Debbie Thomas co-founded TMG in July 1990 in Vancouver with four brokers and a different premise: if the brokerage is valuable, agents will stay because the platform serves them, not because exit is costly.

> "There must be a better way to look after agents," says Grant Thomas, CEO of TMG The Mortgage Group.

That framing led directly to the structural choices that define TMG today. Short notice periods. Agent-owned databases. Revenue from splits only, not layered fees. The brokerage makes money when the agent closes deals, and the relationship ends cleanly when either party decides it should.

## Why Exit-ability Creates Loyalty

The counterintuitive effect is that making it easy to leave increases the likelihood agents will stay. When an agent knows they can walk in 60 days and take their book with them, the decision to stay is an active endorsement of the platform's value, re-made every quarter. That continuous re-evaluation forces the brokerage to remain useful. The brokerage cannot rely on contractual lock-in or switching costs--it must deliver tangible service.

Contrast this with high-friction exit structures, where agents stay because switching costs are prohibitive. The brokerage still retains headcount, but the agents are no longer evaluating whether the relationship serves them. They are evaluating whether the cost of leaving exceeds the cost of staying. That is a different calculation, and it produces different behaviour. Resentment compounds. Service standards drift. The relationship becomes transactional in the worst sense.

TMG's structure skips that failure mode entirely. The 1,700 professionals operating under the banner are there because the split, the compliance filings, the brand neutrality, or the back-office support justifies the relationship. When it stops justifying the relationship, the agent leaves, and both parties move on.

## Margin Discipline and the Agent-First Response

The mortgage brokerage industry has faced sustained margin pressure for years, driven by regulatory costs, technology investment requirements, and competition from bank-owned channels. One common response is fee-loading: administrative charges, desk fees, marketing levies, compliance surcharges. These fees shift risk from the brokerage to the agent and create a revenue floor that persists even when origination volume drops.

TMG went the other direction. The company operates on commission splits only, with no layered fees. That structure aligns the brokerage's revenue directly with agent production. The brokerage earns nothing when the agent struggles, unlike fee-based models where administrative charges arrive regardless of whether the agent is thriving.

> "We've built a model where every dollar we earn is directly tied to agent success," says Grant Thomas. "That keeps us focused on what actually helps brokers close business."

The trade-off is clear. A split-only model requires scale to remain profitable at the corporate level. Without the volume that 1,700 agents generate, the lean structure would not cover the compliance, governance, and back-office costs required to manage independent contractors across multiple provincial jurisdictions. The model works because TMG reached critical mass, but reaching that mass required decades of patient growth without the capital cushion that fee revenue provides. The upside is permanent alignment. The brokerage earns only when agents earn, so every operational decision runs through the same filter: does this make it easier for brokers to get deals done?

## The 35-Year Horizon

Grant Thomas served as Chairman and CEO before bringing in Mark Kerzner as CEO in 2009 to lead the next phase of expansion. The timeline is worth noting. Twenty years to build the foundation, then a leadership transition to scale further. No blitz, no exit, no acquisition by a bank-owned consolidator.

The Thomas family's approach defies the venture-capital playbook that dominates current entrepreneurial advice. There was no pitch deck, no Series A, no growth-at-all-costs mandate. The business compounded quietly, broker by broker, market by market, funded by its own cash flow and governed by the principle that the brokerage exists to serve the agents, not the other way around.

That philosophy has produced a durable structure. TMG remains independent, privately held, and large enough that the agent-first model works at population scale. Over decades, the split-only structure and agent ownership of client files outperform fee-heavy models run on short time horizons.