Market Maker KPI
Market maker KPIs are the numeric obligations a market making agreement places on the firm: how tight the quoted spread must be, how much size must sit on each side of the book, and what proportion of the time those quotes must be live. They are what separates an enforceable agreement from a statement of intent.
TradFi parallel: Like the quoting obligations a designated market maker accepts on a traditional exchange, measured continuously and reported against.
Key Takeaways
- 01Spread, depth and uptime are the three KPIs that carry most of the weight in a market making agreement
- 02They must hold simultaneously, a tight spread with no size behind it satisfies one target and delivers nothing
- 03Depth is meaningless without a stated distance from the mid price
- 04Uptime is meaningless without a stated sampling frequency and a cap on the longest single outage
- 05Every KPI needs a venue list, because obligations met on one exchange say nothing about the others
- 06A measurement clause naming the data source and dispute process is what makes the numbers enforceable
How It Works
Three KPIs carry most of the weight. Spread caps the distance between the best bid and the best ask, usually in basis points. Depth sets the minimum notional resting on each side within a stated distance from the mid price. Uptime sets the proportion of the measurement period during which the spread and depth obligations are both satisfied simultaneously.
The three interact, which is why quoting any one of them alone is misleading. A very tight spread with almost no size behind it is easy to maintain and close to useless to anyone trying to trade. Deep size at a wide spread is expensive to cross. High uptime on a loose spread target means little. This is why a serious agreement states all three together and requires them to hold at the same time, rather than reporting each in isolation.
The definitional details are where disputes originate. Depth needs a distance from mid to be meaningful, since size sitting two percent away is not the same as size at half a percent. Uptime needs a sampling frequency, because a target measured hourly and one measured per second describe very different services. Every KPI needs a venue list, since obligations met on one exchange say nothing about the others. Agreements that leave these unstated are not enforceable, however precise the headline numbers look.
Real World Examples
Tight spread, empty book
A firm maintains a 20 basis point spread with 500 US dollars on each side. The spread KPI is met continuously and no participant of any size can trade. Pairing every spread target with a minimum depth requirement at a stated distance from mid is what prevents this.
Uptime that hides a long outage
A 95 percent monthly uptime target permits roughly 36 continuous hours offline. If that window lands during a volatility event, the token had no quotes when it needed them most. Capping the longest permitted single outage alongside the percentage closes the gap.
KPIs met on one venue only
An agreement states targets without naming venues. The firm meets them on the deepest exchange and quotes thinly elsewhere. Since the obligations never specified coverage, nothing was breached. Venue lists are part of the KPI, not a separate concern.
Frequently Asked Questions
What is a reasonable spread target?
It depends heavily on the token's volatility, market capitalisation and venue. A target that is reasonable for a large liquid asset is unachievable for a thin one, and a firm that accepts an unrealistic target will simply breach it. The useful question is not what number is standard but whether the number is achievable and measurable on the named venues.
Should KPIs differ by venue?
Usually yes. A tier one exchange and a smaller venue have different liquidity profiles, and applying one target across both either overcommits the firm on the smaller venue or undercommits it on the larger. Per venue targets are more work to write and considerably more meaningful.
How often should performance be measured?
Continuously if the data allows, and reported at a frequency the issuer will actually review. The important thing is that the sampling frequency is written into the agreement, because a target measured per second and the same target measured hourly describe materially different obligations.
What happens when a KPI is missed?
Whatever the agreement says, which in many agreements is nothing. A remedy clause is what gives the targets force: a fee reduction, a cure period, a shortened notice period, or a right to terminate after sustained underperformance. Without one, a missed KPI has no consequence.
Related Terms
Track on Tokenomist
Supply-side analysis for educational purposes. Not financial advice. Verify assumption and precision labels on the relevant token page.