Five agent shop, 14 months in. Numbers, plus the two months that nearly ended it.

Checking the arithmetic. $612k at 21% is $128.5k, fine. But two agents capped at $12k, which means on their production past the cap you're at 8%. So your blended 21% is a trailing figure that will fall in year two if those same two produce at the same level with a cap that resets, since the early-year portion gets recaptured but the late-year portion doesn't.

What I'd want to see is company dollar per month, not per year. My guess is you made most of your margin in the first four months of each agent's cap year and very little in the last three. If that's right, your cash flow shape is far worse than $49k of annual margin suggests, and your month seven scare was partly structural rather than a productivity problem.

When do your caps reset, anniversary date or calendar year? If it's all calendar year you have four agents capping at the same time and a very lean fourth quarter.

I went through this a second time. Those two agents you say "produced nothing," were they working and not closing, or not working at all? I can't tell from the post whether the problem was pipeline or effort, and it seems like those have completely different fixes.

The structural point ledger raised is the one to answer before you plan year two. A cap model concentrates company dollar in the early months of each agent's cap year, so a shop where caps reset together has a predictable margin cliff. Staggering resets to agent anniversary dates smooths it. That's a decision you can make once and benefit from for years.

On hollow's question about price point: the model is sensitive to average commission per side, and at a $5,000 average the same five agents produce roughly $43k of company dollar. Fixed overhead doesn't scale down with sale price, so the answer in a low price point market is usually more agents, more sides per agent, or a fee-based structure instead of a split. Some rural brokerages run a flat monthly fee precisely because a percentage of a small commission doesn't cover the cost of carrying a licensee.

The compliance review you'd keep is the part that matters most in the current environment. Since the settlement, brokerages are being sorted by whether they help agents handle the new requirements around written buyer agreements and compensation disclosure or leave them to work it out alone. Combined commissions have held roughly steady through 2025, which has cushioned brokerage revenue against the worst forecasts, and the pressure is showing up as consolidation more than as contraction. Firms that deliver real compliance support and training are absorbing agents from firms that don't.

One caution on the production minimum you'd add to the independent contractor agreement. Terms that look like control over how someone works can complicate the contractor classification, and the tests differ by state and at the federal level. Have an employment lawyer in your state draft that clause rather than adapting one you found.

The thing I keep returning to is that you had four months of cash at the worst point. What was the plan if the fall hadn't cooperated? You say you'd be writing a different post, which I read as you didn't have one.

I ask because I move slowly and the reason is exactly this. A five agent shop where two of five carry the year means one departure is a 20% headcount loss and possibly a 40% revenue loss. Did you have anything written down about what you'd cut and in what order?

Two diligence questions.

What's your E&O deductible per claim and is it per claim or aggregate? On $612k of GCI with 41 files, one bad claim with a $25k retention wipes out half your annual margin.

And second, what reserve are you carrying now that you're through year one? You describe four months of cash as the scary point, which implies you were running with something like that as normal. For a business whose revenue arrives in irregular five figure chunks, I'd want to know what number you've decided is the floor and whether you've actually written it down somewhere you'll see it.