Crypto market makers charge in two fundamentally different ways. Under a retainer you pay cash for defined quoting obligations. Under loan + call option you pay little or no cash and instead lend tokens and grant options over them. The cost gets paid in upside, and it never shows up on an invoice.
Founders almost always compare quotes on monthly fee. That comparison is close to meaningless, because the two structures don't put the cost in the same place.
Loan + call option | Retainer (+ profit share) | |
|---|---|---|
Cash cost | Low or zero | Monthly, ongoing |
Token cost | Tokens lent + options granted | Inventory only, no options |
Upside dilution | Yes - can be substantial | None |
Cost visibility | Hidden until price moves | Fully visible upfront |
What the provider profits from | Price movement | Spread capture and fees |
Alignment with a deep, stable book | Weak | Stronger, if KPIs are defined |
Auditability | Hard | Moderate - depends on reporting |
Runway required | Minimal | Real monthly commitment |
Best fit | Cash-poor, token-rich, high price conviction | Funded projects that can define and measure targets |
A note on how this was written: what follows on loan + call option reflects how the structure is publicly documented and commonly discussed in this market, not deal-specific information from AnyPartners' own network. There are real situations, covered in full below, where it's the right structure to choose.
How loan + call option actually works
The mechanics, stated precisely, because this is where most misunderstandings start:
You lend the market maker a quantity of tokens for a fixed term - commonly 12 months.
They use those tokens as inventory to quote both sides of the book on agreed venues.
In exchange, they receive call options over some or all of the loaned tokens, usually at several tiered strike prices above the price at signing.
Cash fee is zero or a small setup charge.
At expiry they either exercise - buying tokens at the strike and paying you cash - or return the tokens.
The pitch is "we only win if your token wins." True, as far as it goes. It leaves out the part that costs you money.
What you actually paid. You sold the market maker the right to buy your tokens cheaply if the price rises. That right has a real, calculable value on the day you sign. Your token performs, they exercise, and the gap between strike and market price is money that would otherwise have sat in your treasury. That's the fee. It's just denominated in upside and settled later, which is exactly why it never shows up next to a dollar sign.
Four places the incentives split from yours. Option value tracks volatility, not liquidity quality. You want a deep, stable book; an option holder wants a big move, and those pull in different directions more often than the pitch admits. The provider can hedge elsewhere too: perpetuals, other venues, correlated assets. Whatever their real net exposure to your token is, you generally can't see it from the outside.
Then there's the part that shows up late. If price never approaches the strikes, the options drift toward worthless, and so does the provider's reason to keep the book tight, right when you need it most. And expiry is a supply event either way. Tokens get exercised or returned. Something happens to your float on a known date, so plan for it before signing, not in the final month.
Serious firms offer this structure. Nothing above makes it a bad-faith deal. It prices differently than founders assume, and the risk sits somewhere the monthly-fee comparison simply doesn't look.
How to price a token loan so you can compare it to a retainer
This is the calculation almost nobody runs, and it's the only way to compare the two structures honestly. You are granting options; options have a value; that value is your fee.
The method. Start with the notional: tokens loaned, times price at signing. For each strike tier, write down the strike, the quantity, the term. Then the input that actually drives the answer: your token's annualised volatility, or, pre-launch, the closest comparable tokens by cap band and sector. Price each tranche with a standard options model (Black-Scholes is fine for a first pass), sum them, divide by the term in months. What comes out the other end is a dollar figure you can put next to a retainer quote.
Illustrative only - this is a worked method, not a market figure. Say you lend 2,000,000 tokens at $0.05 ($100,000 notional), 12-month term, options split across strikes at +30%, +60% and +100%. Run those numbers through a standard model with volatility assumptions typical of early-stage tokens, and the summed value of the three tranches isn't small relative to the notional. Often a meaningful fraction of it. Sometimes more, depending almost entirely on the volatility you assume. Don't anchor on that description. Run it with your real numbers - that output belongs next to your retainer quotes, not this paragraph.
Two practical notes. Volatility is the sensitive input, so run a range rather than a single figure (low, expected, high) and look at the spread of outcomes. And if the notional is meaningful relative to your treasury, have the grant priced by someone who does this professionally. The cost of an hour of a quant's time is trivial next to the size of what's typically being given away here.
How retainer and profit share actually work
A retainer is the boring option, and boring is the point. Fixed monthly fee, defined obligations: a target spread, depth at a stated distance from mid, an uptime percentage, on named venues. Cost is knowable. No upside gets surrendered. The provider gets paid for maintaining the book, not for the price doing anything in particular.
Its one real weakness: a retainer with no measurable KPIs is a subscription to nothing. Say "competitive spreads" instead of a number at a stated depth with an uptime figure, and you've got no basis to evaluate delivery and no grounds to terminate. The advantage only exists if you actually use it.
Profit share flips the alignment. The provider keeps a share of trading profit, mostly spread capture, which ties their income to genuine market-making activity: more two-sided flow at tighter spreads earns more. Two real weaknesses come with it, and skipping them would make this section a sales pitch instead of an analysis. Verification is the first: profit share means trusting reported P&L unless your agreement gives you audit rights or venue-level data, so ask for them. The second is sharper. Wider spreads mean more capture per trade, and without an enforced maximum-spread ceiling, profit share can quietly reward the exact behaviour you're paying it to prevent. Pair the two, always.
Retainer plus profit share, done right, beats either alone. Done without defined KPIs, it's just two undefined structures stacked on top of each other. The structure doesn't do the work. The terms do.
Who this actually fits - narrower than the pitch suggests
Real cases exist, and pretending otherwise would be dishonest. But look at what has to be true at the same time, not just any single item on the list.
You need genuinely no cash. Not "tight." Not "would rather not spend it." A runway that can't absorb a monthly retainer for twelve months. You need real token reserves, large enough to lend without starving your own treasury. And you need enough conviction in your own price trajectory that giving away upside at a set of strikes feels like a fair trade, not a bet you wouldn't otherwise take. Most early-stage projects clear one of those, maybe two. Clearing all three at once is the actual bar - higher than "we don't have much cash right now," which describes a lot of teams and isn't by itself a reason to choose this over a retainer.
Where it does hold up: strikes set high enough that they only pay out on an outcome you'd genuinely call a win, in which case the cost only bites when things went well anyway. Or it's simply the only offer on the table. Smaller raises sometimes don't get a retainer quote at all, and a loan structure beats no market making.
If you do go this route, negotiate the service, not just the option terms. Shorter term. Higher strikes. Smaller loaned quantity. A defined return or exercise process at expiry. And quoting KPIs written in regardless of how the deal is priced - that's the one people skip. Zero cash fee isn't a reason to accept undefined obligations. Founders negotiate the option terms hard and leave the actual service vague. That's the single most common mistake in these deals.
The comparison that matters
The real question isn't which structure is objectively better. It's which one fits your runway, your reserves, your float, and your ability to measure delivery.
Work through it in this order. Can you sustain a retainer for twelve months without touching product runway? If not, loan structures are probably your realistic option. Spend your energy optimising their terms instead of agonising over the choice. Can you state a target spread and depth? If not, fix that first; neither structure protects a buyer who can't define delivery. Price the option grant using the method above and compare the monthly-equivalent figure to your retainer quotes. Plenty of founders find the "free" deal isn't. Ask both provider types the same KPI questions; how they respond tells you more than the pricing model does. And never sign twelve months as a first engagement without a defined review point. Providers confident in their own delivery accept review clauses without much friction.
If a provider won't write a measurable spread and depth commitment into the agreement, the structure has already stopped being the thing that matters.