Devin Hartcastle
Measure the trader, not just the referral
A partner who sends 100 challenge buyers at a low cost can look better than one who sends 35, until refunds, repeat purchases and funded-trader outcomes are included. The unit of analysis should be a referred trader followed through the stages that matter to the business, not a click or a first checkout.
Track each partner cohort from first tracked visit to payment, refund or chargeback, challenge result, funded account and later purchase. Use consistent windows, such as 30 days for initial conversion and 60 or 90 days for repeat activity. Keep the original referral source attached to the trader record, even if the trader returns through a direct visit or a different campaign.
Separate the funnel into measurable events:
- Tracked visits and valid landing-page sessions.
- Challenge checkout starts, paid orders and failed payments.
- Refunds, chargebacks and any payment disputes.
- Challenge completion, funded-account activation and first payout.
- Repeat challenge purchases and net revenue by cohort.
Define each event before partners are onboarded. For example, count a funded account only when the account is activated under your rules, not when a trader passes a challenge. This stops teams comparing different definitions across reports.
Set attribution and payout rules before launch
Affiliate programmes usually reward a publisher for referred sales. An introducing broker, or IB, may operate a sub-affiliate network or receive a continuing share of eligible revenue. Prop firms should define exactly what the share applies to.
Challenge fees, discounts, refunds and repeat purchases can all change the payable amount. Do not base a payout on simulated trading volume unless the business model and applicable rules clearly support it.
For a small programme, use a referral ID, a partner-specific landing URL and a coupon code as a fallback. Store first-touch and last-touch sources separately. A 30-day last-click window is a common starting point, but report coupon-assisted orders and view-through claims separately rather than silently assigning them to the last click. Deduplicate by order ID and trader account. Set a written rule for cross-device referrals and direct returns.
The figures below are typical market ranges for planning, not audited results or promises. Results vary by offer, geography, traffic source and definitions.
| Metric | Typical market range | How to use it |
|---|---|---|
| Click-to-paid challenge conversion | 0.5% to 3% | Compare like-for-like traffic and landing pages, not raw partner totals. |
| Refunds and chargebacks as a share of paid orders | 2% to 8% | Review by partner and payment method; spikes can indicate weak traffic or unclear offer terms. |
| Paid challenge to funded account | 3% to 10% | Use the same qualification window and funded-account definition across cohorts. |
| Second challenge purchase within 60 days | 10% to 30% | Check whether repeat orders are genuine demand, not a consequence of confusing rules. |
| Revenue-share commission on eligible net receipts | 10% to 30% | Model the payout after discounts, refunds and payment costs. |
Every commission rule needs a matching event definition in the reporting system.
Judge partner quality on contribution margin
First-sale CPA is useful for pacing spend, but it is not a profitability measure. Calculate partner contribution as net receipts less refunds, payment processing, partner commission, support costs and any attributable acquisition incentives. If a trader makes another purchase, include it in the cohort value only after the purchase clears the refund window.
Then compare that value with the partner cost over a fixed period. A simple 60-day contribution-to-commission ratio helps identify partners whose cohorts pay back quickly.
Extend the view to 90 or 180 days where repeat purchasing is material, but do not mix immature cohorts with mature ones. A cohort acquired last week has not had the same opportunity to generate funded activity as one acquired three months ago.
Break results down by partner, source, country, device and offer. A partner may have strong conversion overall but poor performance in a particular market or on a discounted challenge. Watch operational signals alongside revenue: support contacts per buyer, payment disputes, rule-related complaints and the share of traders who reach funded status. These can reveal poor fit before a chargeback rate moves.
Use minimum sample sizes. A difference between 1 and 2 purchases is not reliable evidence that one partner is twice as effective. Flag results for review when there are enough paid orders to compare, and show both the rate and the count. Keep low-volume partners in a separate view rather than ranking them against established publishers.
Design commissions around measurable outcomes
Flat CPA makes partner costs predictable, but it can reward volume without regard to trader quality. Revenue share aligns payment with receipts, but can create disputes over discounts, refunds and repeat orders. Hybrid terms can balance the two: a lower initial CPA plus a share of eligible net receipts, or tiered rates after a partner reaches agreed quality and volume thresholds.
Put a hold period into the terms so commissions are approved after the relevant refund and dispute checks. State whether repeat purchases qualify, how long the attribution window lasts, which countries are excluded and how suspected self-referrals are handled. Apply the same terms consistently, with documented exceptions rather than private arrangements that cannot be reconciled.
Run one change at a time. If testing a higher commission, keep the landing page and offer stable for that cohort. Compare net contribution, not only order count. For a partner producing modest volume, use a capped pilot with a defined review date. A marketing workflow tool such as prop-firm.ai's $99 monthly plan can sit alongside the reporting process, but the event definitions and payout ledger still need an owner.
Run a monthly partner review
Give marketing, finance, support and risk teams the same partner-level report. Reconcile attributed orders with payment records before approving payouts. Review exceptions weekly when spend or sales volume is high, then make a fuller monthly decision about rates, caps and placements.
A useful review sequence is:
- Reconcile clicks, orders, refunds and approved commissions against source records.
- Compare mature cohorts on net contribution, funded-account activity and repeat purchases.
- Investigate outliers in disputes, support burden, geography or conversion before changing terms.
- Record the action, owner and review date for each partner: scale, hold, renegotiate or pause.
Keep a change log for attribution settings, coupon rules and commission terms. Without it, a reporting shift can look like a performance change. If you are standardising campaign briefs and reporting workflows, the prop-firm.ai marketing agent is one option to assess alongside your existing stack.
Stop doing these:
- Paying on unverified checkout events.
- Ranking partners by gross sales alone.
- Changing attribution rules mid-cohort without recording the change.
Frequently Asked Questions
What should a prop firm track in an affiliate programme?
Track referrals through paid challenge orders, refunds, chargebacks, funded-account activation and repeat purchases. Report rates alongside order counts, and use consistent cohort windows so partner results can be compared fairly.
How long should affiliate attribution last for a prop firm?
A 30-day last-click window is a common starting point, but the right period depends on the buying cycle and partner mix. Set the rule in writing, store first-touch data as well, and report coupon-assisted or view-through sales separately.
How should an IB commission differ from an affiliate commission?
An affiliate commission commonly rewards a referred sale, while an IB arrangement may include sub-affiliates or continuing revenue share. Define the eligible receipts, refund treatment, repeat-purchase terms and attribution rules for either model before launch.
