Starting a mortgage brokerage in Canada

First, the question worth asking

A producing agent moving to their own brokerage is trading a share of their commission for ownership, and taking on a job — running a brokerage — that is not the job they are currently good at. That trade is often worth making. It is worth making deliberately.

The two motivations that work are wanting to keep more of what you already produce at real volume, and wanting to build something with other agents in it. The motivation that reliably disappoints is frustration with a current brokerage: you will discover that most of what irritates you is the industry rather than the firm, and now it is yours to fix.

Be clear about which you are doing, because it determines everything below — a brokerage built to house one producer and a brokerage built to recruit thirty are different businesses with different costs.

Licensing and the principal broker

Mortgage brokering is licensed provincially, and both the firm and the individuals need to be authorised. Every province requires the brokerage to have a designated individual — the principal broker, or the province's equivalent title — who is personally accountable for the brokerage's compliance with mortgage brokering law.

That accountability is real and personal. The principal broker is on the hook for supervising the agents, for the brokerage's policies, and for what happens on files they may never see. If you are opening a brokerage and appointing someone else to the role, understand that you are asking them to accept regulatory liability for your business.

Requirements, licence classes and terminology differ by province, and they change. Treat the regulator's current published requirements as the authority rather than anything you read second-hand, including this page.

Network or independent

This is the decision with the longest tail. Joining a network — a franchise or affiliation model — buys you brand, lender status from day one, some technology, and a route to volume-based compensation you could not access alone. It costs a share of revenue and some independence over how you operate.

Going independent keeps the economics and the control, and requires you to build lender relationships from a standing start with no volume history behind you. For a brokerage with genuine volume already, that is a solvable problem. For a new shop with one or two producers, it can be the difference between competitive compensation and not.

There is no general answer. There is an answer for your volume, your province and your lender mix, and it is worth modelling both rather than deciding on temperament.

• What lender status and compensation tier does the network actually give you on day one?
• What is the total cost — fee, split, technology, marketing levies — as a percentage of revenue at your projected volume?
• What are you contractually required to use, and what happens to your data if you leave?
• What is the term, and what does exiting look like — including who owns the client relationships?
• If you recruit, does the network's compensation model help or compete with yours?

The build-out, in the order it actually has to happen

Most of these run in parallel, but some genuinely block others — you cannot get lender status without the brokerage licence, and you cannot recruit credibly without knowing what you are paying.

1. 1

   Incorporate and set up banking

   The corporate entity, the business banking, and — where your province requires one for holding client funds — the trust account, with the controls the regulator expects around it. Get the accountant involved before rather than after.

2. 2

   Licence the brokerage and the principal broker

   The firm licence, the principal broker's authorisation, and any provincial requirements about premises, records location and public-facing name. Allow considerably more time than the published processing estimate.

3. 3

   Errors and omissions insurance

   Required for licensing in most provinces, and the requirements are specific about coverage limits and fraud provisions. Arrange it early — proof of coverage is usually a licensing prerequisite rather than a follow-up item.

4. 4

   Build the FINTRAC compliance program

   Since October 2024 this is not optional and not deferrable: compliance officer, written policies, risk assessment, training program, and the two-year effectiveness review scheduled. Build it before the first file, not after the first examination letter.

5. 5

   Get lender status

   Approach lenders with a realistic volume story and expect the first tier to be modest. This takes longer than anything else on the list and is the most common reason a launch date slips.

6. 6

   Decide compensation before you recruit

   Splits, desk fees, who pays for what, how and when commissions are paid out, and what happens to a deal in flight when an agent leaves. Every one of these will be argued about eventually. Write them down first.

7. 7

   Choose the operating system

   CRM, compliance records, commission calculation, payroll. Choosing this late means migrating in year two, which is the most avoidable expensive thing a new brokerage does to itself.

8. 8

   Then recruit

   With licence, lenders, compensation and systems in place you have something to recruit to. Without them you are asking agents to take a risk on your behalf, and the good ones will not.

The operating load nobody quotes for

The costs in a business plan are the ones with invoices. The costs that sink new brokerages are hours — specifically, the principal broker's hours, spent on work that does not originate a single deal.

Every one of these grows roughly linearly with agent count, which is why the transition from five agents to twenty is harder than the transition from zero to five. At five you can hold it in your head. At twenty you cannot, and the first sign of that is usually a compliance gap rather than a complaint.

This is the part where the systems decision from the previous section either pays for itself or costs you twice.

Cash flow, honestly

Mortgage commission is paid on funding, and funding lags origination by weeks or months. A new brokerage therefore spends money for a meaningful period before receiving any, and the period is longer than the founder's mental model of it — because the pipeline that exists at launch was built under the old brokerage's licence and may not travel with you.

Plan for the fixed costs to run for longer than feels necessary, and be deliberate about whether the founding producer keeps producing during the build. In most successful launches they do, which means the brokerage build happens in evenings — a real constraint worth acknowledging in the plan rather than discovering.

Recruit against real economics, too. Agents joining a new brokerage carry risk, and the offer that attracts good ones is a better system and honest numbers rather than the highest split. A split you cannot sustain gets revised, and revising a split is how brokerages lose the agents they most wanted.

• Model fixed costs running for longer than your expected first funding.
• Assume your existing pipeline does not travel with you until you have confirmed it can.
• Know your break-even in funded deals per month, not in dollars of revenue.
• Do not buy a split you cannot fund at scale; you will have to take it back.
• Keep a reserve for the first commission dispute, because there will be one.