Setting Supplier Rates and Splits: A Field Guide for New Agency Owners
The TravelIQ team · 7 min read · Updated August 2026
Where commission comes from
Everything an agency pays anyone starts as supplier commission: the percentage of a booking's value the cruise line, tour operator, hotel, or resort pays the agency for producing the sale. Rates vary by supplier and product, but most of what a leisure agency sells pays somewhere in the ten-to-sixteen percent range — with real variation inside a single supplier's catalog, since the commissionable amount often excludes port fees, taxes, and airfare. Two ten-thousand-dollar bookings can earn meaningfully different commissions, and the difference lives in the fine print of what's commissionable.
Three practical habits follow. Know your actual rates — not the brochure headline but what each supplier pays you, at your volume, in writing, kept somewhere everyone who books can check rather than in the founder's inbox. Know the base — an advisor who quotes "16% of the trip" when the supplier pays 16% of the cruise fare has already created next quarter's reconciliation dispute. And treat rates as changeable, because they are: volume tiers, consortium affiliations, and annual renegotiations all move them, so date what you record. Every later number in this guide is arithmetic on top of these; get the inputs wrong and the fairest split in the industry is a fair share of a fiction.
Agency cut vs. advisor split
The split is the agency's single most-discussed number: of each commission dollar, what share goes to the advisor who sold the trip and what stays with the house. Across the host industry the spectrum runs roughly from 70/30 — common where the agency supplies real infrastructure, training, and lead flow — to 90/10 and above, where the host is closer to a compliance-and-credentials umbrella and the advisor brings everything else. Neither end is virtuous. They're different products at different prices.
The mistake new owners make is picking the number competitively — "the host down the road says 80/20, so we'll say 85/15" — without costing what the agency's cut has to fund: E&O coverage, the booking platform, marketing, training, the time someone spends fixing an advisor's debit-memo problem at 9pm. The house share isn't the owner's profit; it's the budget for everything advisors can't or shouldn't each buy alone, plus a margin. Price it like the product it is. An unsustainable split fails advisors slowly and then suddenly, when the support they signed for turns out to be unfunded.
Designing a split that scales
A flat split is right at the start — one number, explainable in a sentence, identical for everyone. Its fairness decays as the roster's production range widens: when your top producer books twenty times your newest recruit, a flat rate quietly overcharges the producer for support they no longer use, and the correction arrives as a one-off private deal that undermines the published number for everyone. One secret exception costs more trust than a whole tier ladder.
Tiers are the durable answer: the advisor's share steps up at published annual-production thresholds. Design them with restraint — two or three thresholds at round numbers everyone can recite, entry rate identical for all starters, and the step generous enough to feel like a promotion rather than a rounding error. Publish the ladder internally; its entire function is that the ambitious middle of your roster can see the next rung. Decide edge rules while they're hypothetical, not during a dispute: when does a tier take effect, does production reset each year, what happens in a down year. And write down how referrals and multi-advisor bookings split before the first one happens, because improvising that math on a live booking is how partnerships end.
Keeping the math honest
A split structure is only as good as advisors' belief that it's applied correctly, and belief tracks verifiability, not sincerity. The advisor should be able to trace any payout line back through the chain — booking, commission received, tier in force, split applied, their share to the cent — without asking the owner to explain it. When statements are hand-built at month-end, every discrepancy defaults to suspicion, because there's nothing to check against except the same person who made the error. When the trail is visible, a discrepancy is a lookup instead of a confrontation.
Two boundaries keep the system honest at scale. The arithmetic — applying tiers, resolving splits, carving referral shares — should be mechanical, computed from published rules, never re-derived by hand for individual bookings, because hand math is both where errors live and where favoritism is suspected even when absent. The judgments — what commission a booking should earn, whether an exception is warranted — belong to named humans and should look like judgments, not hide inside the math. And pay on a fixed rhythm where a settled statement stays settled: late supplier money rolls into the next cycle rather than reopening a period an advisor already reconciled against their own records. Advisors forgive the industry's slow suppliers. What they don't forgive is a payout system that behaves like weather.