Recurring vs One-Time Affiliate Commissions: The Math That Decides Your Income
The two payment models produce completely different income curves. Here is the arithmetic, worked in the open, so you can run it on any pair of programs.
Published July 17, 2026
A recurring commission is a stream; a one-time bounty is a payment. On a $50/mo product paying 30% recurring, one referred customer earns $15 every month they stay, passing a $100 flat bounty during month 7 and reaching roughly $212 over two years at a typical 5% monthly churn. 64 of the 160 programs in our directory (40%) pay recurring. The flat bounty still wins in three specific situations covered below.
The two models, stated precisely
A one-time commission pays a fixed amount or percentage on the first qualifying transaction, once. A $100 bounty per hosting signup, $65 per Bluehost sale, $200 per Semrush subscription: after the payment clears, that customer never generates another cent for you regardless of how long they stay.
A recurring commission pays a percentage of every payment the referred customer makes, either for a capped window (12 or 24 months are the common caps) or for the customer's whole life. Frase pays 30% recurring uncapped; Murf AI pays 20% recurring capped at 24 months; Webflow pays 50% but only through the first year. The cap matters as much as the rate, and it is the first thing to check in the terms.
Your income under the two models behaves differently in kind, not just in size. One-time income resets to zero every month; you start again from nothing. Recurring income stacks: this month's referrals sit on top of every referral still paying from the months before. Same traffic, different curve shapes.
The break-even, worked step by step
Take a concrete pair, stated as an illustration so you can swap in real numbers: a product costing $50/mo where program A pays 30% recurring ($15 per customer per month) and program B pays a $100 flat bounty.
With no churn, cumulative earnings per referred customer look like this:
| Month | Recurring (30% of $50) | One-time ($100) | Recurring is |
|---|---|---|---|
| 1 | $15 | $100 | behind by $85 |
| 3 | $45 | $100 | behind by $55 |
| 6 | $90 | $100 | behind by $10 |
| 7 | $105 | $100 | ahead by $5 |
| 12 | $180 | $100 | ahead by $80 |
| 24 | $360 | $100 | ahead by $260 |
Break-even lands during month 7 ($15 × 7 = $105). Everything after that is money the flat deal never pays. But no churn is a fantasy, so the real question is what the stream is worth when customers cancel.
Adding churn: the honest version
If a fraction of customers cancels each month, your expected earnings from one referral are the monthly commission multiplied by the probability the customer is still subscribed. Summing that over time gives the expected cumulative payout. Same $15/mo example under four churn scenarios:
| Scenario | 12-month value | 24-month value | Beats the $100 bounty? |
|---|---|---|---|
| 0% monthly churn (nobody cancels) | $180 | $360 | Yes, by $260 over 2 years |
| 3% monthly churn (sticky B2B tool) | $153 | $259 | Yes, by $159 over 2 years |
| 5% monthly churn (typical prosumer SaaS) | $138 | $212 | Yes, by $112 over 2 years |
| 10% monthly churn (leaky consumer tool) | $108 | $138 | Yes, by $38 over 2 years |
Read the 10% row carefully, because it is the one that surprises people: even a leaky product beats the bounty here, just later and by less. The flat payment only wins this matchup when churn gets brutal, the bounty gets bigger, or your time horizon gets short. Which brings us to when one-time genuinely is the right call.
The three situations where one-time wins
- The bounty dwarfs the stream. Hosting pays $65 to $200 flat (see WP Engine) on products costing a few dollars a month. A 30% recurring deal on a $5/mo plan is $1.50/mo; the $200 bounty equals eleven years of that stream. When the flat payment exceeds one to two years of the recurring alternative, take the flat payment.
- The product churns fast. Recurring commissions on a product people quit after two months are one-time commissions with extra steps. If trial-heavy audiences and quick cancels define the category, the guaranteed number wins.
- You need cash flow now. A recurring stream pays most of its value in months 6 through 24. If you are reinvesting in content or ads this quarter, $100 today can be worth more than $212 spread over two years. This is a financing decision, not a math error.
The compounding effect across your whole pipeline
The per-customer math understates the real difference, because referrals arrive every month. Say you refer just 10 new customers a month, same $15/mo commission, 5% monthly churn. Under the bounty model you earn $1,000 every month, flat, forever the same. Under the recurring model, month 1 pays $150, but each new cohort stacks on the survivors of every previous cohort. By month 12 your base is around 95 paying customers and the monthly check is near $1,430; by month 24 the base plateaus near 180 customers and roughly $2,700 a month, nearly triple the bounty income, from identical referral volume.
That plateau is worth understanding: at 5% monthly churn, a steady 10 referrals a month tops out around 200 retained customers (10 divided by 0.05), which is where new arrivals only replace cancellations. The recurring model does not compound forever, but it compounds for two years before flattening at a level the flat model never reaches.
Hybrids, caps, and second tiers
Hybrids pay something up front plus a smaller stream: ClickUp pays $25 per signup plus up to 25% recurring, and Cloudways offers either a bounty track or $30 plus 7% lifetime. Hybrids are underrated for new affiliates: cash flow now, compounding later.
Caps convert recurring into delayed one-time. A 50% rate capped at 12 months on a $50 product is worth at most $300 per customer, and at 5% churn about $286. Compare capped deals with uncapped ones by computing the capped ceiling, not the headline rate.
Second tiers stack a percentage of your recruited affiliates' earnings on top of your own. A recurring program with a second tier compounds on two axes at once. The mechanics get their own treatment on our two-tier page, and the income projector models both streams together; the earnings calculator handles the single-stream comparison.
Where recurring deals actually live
Recurring share by category, computed from the live directory:
| Category | Programs | Paying recurring |
|---|---|---|
| SaaS | 19 | 74% |
| SEO Tools | 13 | 62% |
| Email Marketing | 10 | 70% |
| AI Tools | 22 | 64% |
| Productivity | 7 | 71% |
| Hosting | 13 | 15% |
| Education | 8 | 25% |
| Finance | 13 | 8% |
The pattern follows merchant economics: subscription software with fat margins shares revenue for as long as you keep sending it; categories paid per action (finance) or with thin margins share once or not at all. The full rate ranges behind these categories are in our commission rates data.
The risk column nobody prices in
A recurring stream is a promise that the program keeps its terms. Programs cut rates, cap previously uncapped deals, and occasionally shut down, and when that happens your stream takes the hit even if your referrals stay subscribed. Whether existing referrals keep their old rate after a cut varies by program and is usually answered nowhere until it happens; asking the affiliate manager in writing before you commit real traffic is not paranoia, it is due diligence.
This is why the last variable in the recurring versus one-time decision is program stability, not arithmetic. We track every commission and term change we detect on the public changelog, and the documented history of programs cutting rates, including the most famous case, is written up in when affiliate programs cut commissions.
The portfolio answer: run both models on purpose
The framing of recurring versus one-time as a choice misleads slightly, because a healthy affiliate operation holds both, assigned to different jobs. Flat bounties are working capital: they pay this month for this month's work and fund whatever you are building. Recurring streams are the balance sheet: they accumulate into an income floor that keeps paying through a slow publishing month, an algorithm update, or a vacation.
A practical allocation for someone starting from zero: lead with one or two bounty-heavy categories to get cash flowing, and route every piece of evergreen content (the tutorials and comparisons that will still rank in two years) toward recurring programs, because content with a long life deserves a commission with a long life. Twelve months in, the recurring floor changes what risks you can afford everywhere else.
Auditing a recurring stream: the three numbers to watch monthly
- Active referred base. The count of your referred customers still paying. If it plateaus while referrals continue, your churn has caught up with your inflow, and the plateau level (monthly referrals divided by monthly churn rate) is your ceiling under current traffic.
- Cohort survival. Of the customers you referred six months ago, how many still pay? This is your real churn number, and it is the input that turns the tables above from illustration into forecast. Program dashboards that hide it are telling you something.
- Effective rate drift. Commission received divided by your referred customers' estimated spend. When it slips without an announcement, something changed: a cap engaged, a product repriced, or terms moved. That is your cue to reread the terms page and check the changelog.
Five questions for the affiliate manager before you rely on a stream
In writing, before the program earns a structural place in your income:
- Is the recurring commission capped in months, and does the cap apply per customer or per affiliate account?
- If rates change, do my existing referrals keep the rate they signed under?
- What events break attribution: plan changes, account transfers, checkout through support, currency or region switches?
- What is the reversal window, and what share of commissions historically reverses?
- If the program migrates networks, does my referred base migrate with it?
Clear written answers to all five are rarer than they should be, and their absence is information. A recurring stream is a multi-year bet on the counterparty's promises; programs that respect that answer plainly. The documented record of what happens when they do not is in when programs cut commissions.
Methodology
Directory statistics (program counts, recurring shares by category) are computed from the affiliatejob dataset of 160 programs at build time and update automatically with the data. The break-even and churn tables are worked arithmetic on a stated example ($50/mo product, 30% rate, $100 bounty), not measurements; swap in the numbers of any two real programs and the formulas hold. Named programs link to their live listings, where terms are re-verified on a rolling basis; see the methodology for how verification works.