How to Set Commission Rates That Partners Respect

Most commission rates are set in a meeting, by someone who remembers a number from another company. Twenty percent sounds fair, thirty sounds generous, and ten feels cheap. Then the program launches and nobody can say why it pays what it pays. Partners can usually tell.

A rate that partners respect is one that was built from your economics and then explained clearly. The steps below use round numbers you can swap for your own.

What does a customer have to be worth before you pay anyone?

Start with the number most program managers skip: gross margin per customer over a period you can defend. Revenue is the wrong base. If you sell software with an 80% gross margin and a customer pays $100 a month, you keep about $80 of it. If you sell a physical product with a 35% margin on a $100 order, you keep $35.

Now decide how much of that margin you are willing to spend to acquire the customer. Look at what you pay elsewhere. If paid ads cost you $300 to land a customer, and a partner who sends the same customer costs $150, the program is cheap even though $150 feels like a lot of commission. If your cost per customer from other channels is $60, the same $150 is an expensive mistake.

Write the constraint down as a sentence: “We will spend up to X to acquire a customer, and we want to recover it in Y months.” Everything after this follows from that sentence.

How do you turn that budget into a percentage or a flat amount?

Take the software example. A customer pays $100 a month, you keep $80, and you will spend up to four months of margin to acquire one. That is $320 of ceiling per customer.

You now have choices, and they are not equivalent:

• A flat $200 per new paying customer, paid once. Simple to explain, and the cost is fixed.
• 20% of payments for the first 12 months. A customer who stays the full year earns the partner $240 in total, and one who cancels in month three earns $60.
• 30% of payments for the first 6 months. Front-loaded, so the partner sees money sooner.

All three sit under the $320 ceiling. The percentage versions have a built-in protection: a customer who leaves early costs you less. The flat version pays the same whether the customer stays one month or three years, so it needs a different safeguard, which we cover in how refunds and clawbacks should work.

If you are deciding between a flat amount and a share of revenue, flat fee versus revenue share goes through when each one fits.

Why does the partner’s effort matter as much as your margin?

Your ceiling tells you what you can afford. It does not tell you what a good partner will accept. For that, think about how much work the partner does to produce one sale.

A newsletter author who drops a link into an issue they were writing anyway does very little extra work. A consultant who runs a demo for your product, helps the buyer through procurement and answers questions after the sale does a great deal. Paying both the same rate says something about whom you want.

A rough way to check yourself: imagine the partner’s hourly return. Say a reviewer spends 10 hours on a detailed comparison, and it brings you 4 customers a year. At $200 each, that is $800 for 10 hours of work plus whatever the content earns elsewhere. Whether that is attractive depends on the partner, but you can at least see it. If the figure looks thin, that is a more useful signal than the percentage itself.

Programs that depend on skilled partners, such as agencies or consultants, usually need to pay for the work of selling. Programs that rely on a wide audience of small referrers can pay less per sale and win on volume. Neither is better. They are different jobs, and choosing the program type is the first decision that shapes the rate.

Should the commission be recurring?

For a subscription product, this is the question partners ask first, and it deserves a straight answer.

Recurring commissions align the partner with retention. A partner who earns every month the customer stays has a reason to answer questions, send updates and refer people who will actually use the product. The cost is that your liability stays open, and so does your reporting.

A one-time payment is cleaner to model and cheaper to run, but the partner is paid at the moment of sale and has no further stake. If your churn is high in the first month, that is a real problem: you pay for customers who vanish.

The middle path is a limited recurring window, such as the first 6 or 12 months of payments. Partners understand it, you can cap total exposure, and the partner has a reason to care for a while. Whatever you choose, state the window in plain language on the program page. “20% of every payment for 12 months” is easier to trust than “lifetime commission” followed by exceptions in the terms.

What does a rate look like when you test it against real scenarios?

Before you publish, run the rate against three customers. This takes ten minutes and finds most problems.

Take a standard plan at $100 a month with a 20% commission for 12 months.

1. The good customer. Stays 12 months or longer. The partner earns $240 and you spent $240 against $960 of margin (at 80%). Healthy.
2. The early cancel. Pays two months, cancels. Partner earned $40. You lost little. Healthy.
3. The refund. Pays once, asks for money back in week two. If you paid before the refund, you have a $20 problem. If you pay only after the refund window closes, you have none.

That third case is the one most programs forget. It is a policy about timing rather than about rate, and it protects you more than shaving a point off the percentage does.

Then test the upgrade: the customer starts on a $100 plan and moves to a $500 plan in month four. With a percentage, the partner’s earnings rise with the account. With a flat fee, they do not. Both are defensible; just know which one you picked.

When should different partners get different rates?

A single rate for everyone is the easiest thing to run, and you should start there unless you have a reason not to. The reasons to differ are real, though.

The most common is performance. Tiers let a partner who sends 20 customers a quarter earn a better rate than one who sends two, which rewards the people carrying the program without paying a premium for promises. Tiered commissions, with worked examples covers the arithmetic, including how to avoid a cliff that makes partners game the threshold.

The second is product. If a premium plan has a higher margin than your entry plan, a higher rate on the premium plan is consistent with the economics, and it pushes partners toward the customers you want.

The third is the partner type. A reseller who handles the sale and support is a different relationship from someone who posts a link. If that is your situation, run it as a separate program with its own terms rather than special-casing individuals. In Affiliate Tabs, a company can run several programs side by side, each with its own offers, terms and partners, and each with triggers that say what event pays what amount; per-account rate overrides are not part of the product, so separate programs are the intended answer. See the product page for how programs and triggers fit together.

Be careful with secret deals. A partner who learns that someone else got a better rate for the same work will not complain loudly. They will just stop sending leads.

What do partners look at besides the percentage?

Experienced partners read the whole offer, and the rate is one line of it. The other lines often matter more.

• The attribution window. A 30-day window and a 90-day window can differ more in real earnings than 5 points of commission.
• The approval delay. If a commission sits pending for months, the rate is partly fiction.
• How refunds are handled. Partners know products get refunded. They want to know it will not surprise them a year later.
• How and when they are paid. A monthly cycle that is described accurately and kept is worth more than a vague promise of fast payment.
• Whether the product converts. A generous rate on a product that does not sell is worth nothing.

The last one is uncomfortable. Sometimes the right move is not to raise the commission but to fix the landing page, the pricing page or the trial. Partners send traffic to your site, and your site decides what it is worth.

A last check on any rate: say it aloud to a partner you respect. If you find yourself adding apologies or exceptions, the number is not done.

How do you change a rate without losing trust?

At some point you will want to change a rate, up or down. Increases are easy. Decreases are where programs get hurt.

If you must lower a rate, give notice with a date, apply it to sales after that date, and let partners know what they have earned stays earned. Better still, avoid the cut by building the original rate with some space in it, and use time-limited bonuses for pushes you do not want to commit to permanently. A bonus that expires on schedule is not a cut. A rate that quietly drops is.

It also helps to review rates on a calendar rather than in a panic. Twice a year, compare cost per customer through the program with your other channels, look at which partners produce customers who stay, and decide whether the numbers still support what you pay. Keep the review short and keep the notes, so that next time the rate has a reason behind it.

What is a sensible way to start?

Pick a rate that your numbers can bear with room to spare, write it up without exceptions, and recruit a small group of partners you can talk to. After three months, you will know more about what the right rate is than any amount of benchmarking could tell you, because you will have real customers, real churn and real conversations with the people who sell for you.

If you set it too high, you can hold it and grow into it. If you set it too low and lose strong partners in the first quarter, the cost is harder to measure but just as real. When in doubt, lean toward a rate that makes your best prospective partner say yes quickly, then use the contract terms and the structure of the payout to manage your risk.