Designing commission structures that protect margin
The right commission comes from your own gross margin, not a competitor's headline rate. AI can compute your maximum affordable commission and lay out the tiers; your affiliate platform (Refersion, Tapfiliate, UpPromote) is what enforces the rate, cookie window, and payout cap.
Fastest path: one prompt, end to end
🤖 AI prompt — paste into ChatGPT / Claude
You are an affiliate program economist. Compute a commission structure from MY unit economics ONLY.
My numbers:
- Average order value (AOV): [$__]
- COGS per order (or gross margin %): [__]
- Buying cycle: [impulse / considered / long B2B or high-ticket]
- New-customer LTV if known: [$__ or blank]
Do this:
1. Compute gross-margin dollars per order. State it.
2. Compute my MAX affordable commission as a % of the SALE such that I still keep a healthy share of margin (target: commission takes no more than ~one-third of margin dollars). Show the arithmetic.
3. Recommend percentage vs. flat rate: percentage if order values vary widely, flat if high-ticket or thin-margin. Give the specific number.
4. Propose a 3-rung monthly tier ladder (reset monthly by sales count or GMV), and a higher new-customer rate than the returning rate.
5. Recommend a cookie window matched to my buying cycle (impulse ~7-30 days; considered/high-ticket 60-90 days).
Output ONE markdown table | Decision | My number | Reasoning |.
Do not guess or estimate any number. Use only the figures I gave you; if margin/COGS is blank, leave the commission cells blank and tell me exactly what to provide. Do not invent an LTV. If you can't compute a value from my inputs, say so rather than filling it in.
Or do it in 5 steps
- Find your max affordable commission from margin. Model commission as a share of gross-margin dollars, not revenue. A safe ceiling is roughly one-third of your margin per order — beyond that the program stops being accretive.
- Pick percentage or flat. Percentage scales with cart size, so it suits fashion and gifting where order values swing; a flat per-order fee protects margin on high-ticket or thin-margin items. Some programs run a tiered or hybrid blend of the two. SaaS defaults to a recurring 20–30% of the subscription (Tapfiliate 2026 benchmarks).
- Add tiers that reward your best partners. A ladder — say 10% on the first 10 sales/month, 15% on the next 40, 20% beyond — keeps top affiliates from splitting effort with a competitor. Reset monthly by GMV or count, not lifetime volume, so the incentive is to keep performing.
- Set the cookie window to your buying cycle. Attribution windows default to about 30 days in most platforms (Tapfiliate 2026); privacy limits like Apple's ITP and ATT push some programs shorter. High-ticket and software often run 60–90 days to match a longer decision cycle.
- Pay more for new customers than returning ones. A repeat buyer the affiliate didn't influence shouldn't earn full commission. Reserve your top tier for partners you can show are incremental. Done: your terms name the rate, tier resets, cookie window, and per-order cap.
Worked example: max affordable commission
Say your AOV is $120 and COGS is $66, so gross margin is $54/order (45%).
| Step | Figure |
|---|
| Gross margin dollars/order | $54 |
| One-third-of-margin ceiling | $18 |
| $18 as a % of the $120 sale | 15% |
| So max affordable commission | ~15% of sale (or a flat ~$15) |
| New-customer rate | 15% |
| Returning-customer rate | 8–10% (they cost you less to keep) |
If COGS rose to $84 (margin $36), the one-third ceiling drops to $12 → about 10% of sale. Same product, tighter margin, lower rate — the margin sets the number, not habit.
Recalculate whenever COGS or ad costs change.