How to Start a Prop Firm: The Complete 2026 Guide

A retail prop firm is an operating system around a promise: pay a fee, follow the rules, and a qualifying result can become a payout. The promise only works if the market, the evaluation, the technology, the funding, the risk rules and the legal position can all live with one another.

Treat the six decisions as a set. Changing the market can change which technology is even available. Changing the evaluation can change how much cash the payout reserve has to hold. Changing the jurisdiction can rule out a product that looked simple on a slide.

Forex, futures and crypto are not interchangeable skins on the same firm. Each one changes data cost, platform choice and the kind of trader who will pay for an evaluation. Decide this alongside the evaluation model and the jurisdiction, not after a platform contract is signed.

Futures market-data fees are a licensing question, not a flat “one charge per active account” rule. Exchanges and data vendors price subscriber arrangements differently. Some fees follow the account, some follow the user, and some follow the firm’s licence. Read the schedule for the exchange and the provider you will actually use before you put a per-account number in the model.

The evaluation is the product. A one-step challenge is shorter and usually more expensive. A two-step challenge adds a second set of rules and a longer path to a payout. An instant-funding offer skips the evaluation and starts the trader on a funded account, which moves the risk forward.

None of these is automatically safer for the firm. A stricter target can reduce how often traders reach a payout, and it can also raise refund requests and complaints if the rules are hard to understand. A looser target does the opposite. Model the fee, the expected time on the evaluation, and the payout terms together, then write the rules so a trader can apply them without a support ticket.

Simulated and live evaluations can share a rule sheet and still be different businesses. Say which one you are selling. If the account is simulated, say that the displayed balance is the score used for the payout calculation. If orders are routed to a live venue, say that too, and plan the market risk that comes with it.

Building the stack yourself means owning the trader area, the risk checks, the billing and the payout queue. That is real control, and it is also a software company sitting inside the prop firm. Licensing a platform means those pieces already exist and you configure them. A white-label offer is usually the fastest way to put a brand on a working flow, with less room to change the parts you do not control.

YourPropFirm is a prop-firm operating platform in this category. Its public product covers a firm dashboard and CRM, a risk rule engine, a white-label trader area, payments and KYC, and connections to trading platforms such as MetaTrader 4, MetaTrader 5, cTrader, DXtrade and Match-Trader, with further platforms through partner integrations. YourPropFirm and TraderWaves are both part of Quant Technology Group, and TraderWaves operates within the YourPropFirm ecosystem.

That relationship is why it is named here. It is not a claim that one vendor, including TraderWaves, supplies every part of a prop-firm stack. Payments, identity checks, market data, legal advice and the firm’s own reserve policy still sit with the operator. TraderWaves for business is a separate set of tools for communities, reviews and trading workflows, not a substitute for the evaluation, risk and payout systems a prop firm runs.

A licensed platform can shorten the software work. It does not shorten the legal, banking and operational work by the same amount. Those still have to be planned before launch.

Fee income and a capital reserve are two different answers to the same question: when a trader qualifies, who pays them?

A firm funded mainly by challenge fees can be viable if the fees, the pass behaviour, the refunds and the payout rules leave enough cash after costs. It can also run short if payouts, chargebacks or a cluster of winning accounts arrive before the next month of fees. Fee funding is not automatically unstable. It is a model that has to be watched against the cash it actually holds.

A firm that holds a capital reserve can absorb a bad month more comfortably, and it can still fail if the reserve is too small for the book, if costs are ignored, or if live trading losses and payouts arrive together. A reserve is not inherently resilient. Its size relative to the obligations is what matters.

Routing evaluations or funded accounts to a live venue does not, by itself, make the firm sustainable. Funding source, unit economics, reserve adequacy and trading outcomes all sit in the same result. A live book can hedge the firm’s exposure or add market risk the fee model never had. Write down which of those you intend before you describe the firm as “live” in the terms.

Two controls do most of the obvious work on an evaluation: how much the account can lose, and how large a single position can be. Daily loss and maximum drawdown are the first. A cap on lot size, contracts or margin is the second. Thinking of them as two dials is useful when you are drafting the rule sheet. It is not a complete risk model.

Other controls change the outcome as well. A profit target, a minimum number of trading days, a consistency rule, a news restriction, a weekend rule, and a copy-trading check can all change who reaches a payout and how the book concentrates. A time limit or a consistency rule can affect risk. They change behaviour, they change which accounts stay active, and they change when losses show up. They are not a substitute for a loss limit, and they are not irrelevant to it either.

When an account fails, the useful record is the rule that was breached: daily loss, maximum drawdown, position size, or another published condition. Those categories describe the outcome. They do not, on their own, tell you why the trader did it. A drawdown breach can be a planned strategy, a gap, a sizing error or a tilt. The classification does not prove which one.

Keep breach reporting in those operational categories and review it on your own accounts. Do not treat a vendor anecdote, or a single leaderboard, as a measured failure study.

What a firm is allowed to offer depends on the jurisdiction, the product and the execution model. A simulated evaluation sold to retail traders is not automatically “unregulated” everywhere, and a live account is not automatically a licensed investment service in every country. The same challenge page can be a consumer contract in one place and a regulated activity in another.

Before you publish terms, get advice on at least these points for the countries you will actually serve:

• Whether the evaluation, the funded account, or both, are regulated activities where the firm operates and where the trader lives.
• How the contract describes simulated balances, live routing and payouts, so the words match the plumbing.
• Consumer, advertising and refund rules that apply to the fee.
• Identity checks, sanctions screening and payment-provider requirements.
• Tax treatment of the fee and of trader payouts.

This article cannot settle those questions. A lawyer who knows the product and the countries can.

The reserve is the cash the firm expects to need when traders qualify. A displayed account size is a poor stand-in for that number. In a simulated programme the account size is the score. The cash leaving the firm is the payout, after the profit share, and only for accounts that actually receive one.

An illustrative way to size a typical month is:

Expected monthly payouts ≈ active funded accounts × probability of an account receiving a payout that month × average gross payout-eligible profit per receiving account × trader profit share

• Active funded accounts are funded accounts still open during the month, not every challenge that was ever sold.
• Probability of an account receiving a payout that month is the share of those accounts you expect will actually be paid, not the share that merely pass an evaluation.
• Average gross payout-eligible profit per receiving account is the profit the rules count toward a payout, averaged only across accounts that receive one. It is not the displayed account balance.
• Trader profit share is the fraction of that profit the trader is paid. The rest, if the programme keeps it, stays with the firm.

Work an example only as an illustration. Fifty active funded accounts, a 20% chance that any one of them is paid this month, $3,000 of average payout-eligible profit on those paid accounts, and an 80% trader share:

50 × 0.20 × $3,000 × 0.80 = $24,000

That $24,000 is an expected payout figure for the month. It is not a reserve policy. A reserve also has to cover unpaid obligations already approved, operating costs, refunds, chargebacks, delays getting cash out of a payment provider, and months when payouts cluster instead of arriving at the average rate. Simulated balances do not fund any of that. Live routing can add trading losses on top of the payout.

Teams that have done this before often move through the work below. The timings are illustrations, not a schedule you should publish to investors. Market, evaluation model and jurisdiction are worked as one block before the technology contract. A closed beta is part of the sequence because the software, the calculations, the payments and the operations need a real run, not only a read-through of the rules.

1. Fix the product and the legal frame together. Market, evaluation model and jurisdiction, with written answers on what is simulated, what is live, and who the firm can accept.
2. Choose the technology against that product. Build, license or white-label, including payments, KYC and the platforms the market actually requires.
3. Write the rules and the reserve. Loss limits, position limits, the other controls you will enforce, and a cash plan that is larger than one average month of payouts.
4. Run a closed beta. Use it to test account creation, rule calculations, billing, refunds, payout approval and support, as well as whether a trader can understand the rules.
5. Open slowly. A public launch with a cap on new accounts is easier to unwind than a full marketing push on day one.

Skipping the beta to “save a few weeks” leaves those failures for paying customers. There is no reliable public ranking of which mistake closes firms most often. Undercapitalised payouts, unclear rules, a technology choice that cannot do what the terms promise, and a legal position that does not match the product are all enough to stop a launch. Plan for each of them.