Loading...
FinchTrade

Crypto Market Making: How It Works, Strategies and Provider Evaluation

Understand crypto market-making strategies and evaluate providers through two-sided depth, quoting records and clear risk ownership.

By FinchTrade 10 min read
TL;DR Key takeaways
  • Market makers post bids and asks while managing inventory risk. A tight spread alone does not show how much traders can actually buy or sell.
  • Match the arrangement to the job: ongoing listing support, exchange-program participation and a large block trade require different evidence of competence.
  • Evaluate side-specific depth and qualifying two-sided uptime, then clarify who funds inventory, bears losses and controls withdrawals.

A narrow spread can make a trading pair look liquid. It says much less about whether a trader can sell a meaningful position, whether the bid survives sustained selling, or how quickly quotes return after a disruption.

That is the central issue in crypto market making: a maker commits capital to two-sided prices while managing the risk of trading at those prices. Its strategies determine how much liquidity other participants can actually use. For an exchange liquidity lead or institutional desk, evaluating a provider therefore means connecting its proposed terms to its behaviour—not just its headline volume.

How two-sided quotes become inventory risk

A crypto market maker posts bids to buy and asks to sell. Resting orders give incoming traders something to trade against; the maker must manage the inventory created when those orders fill. A taker order executes against available resting liquidity. These labels describe how an order executes, not simply who submits it: a market maker can take liquidity to hedge, and a limit order that crosses the book can execute as a taker.

Consider a hypothetical pair quoted in dollars. A maker bids $9.99 and offers $10.01, with 100 tokens available on each side. The midpoint is $10, so the $0.02 spread is 20 basis points. If it buys 100 tokens at its bid and later sells all 100 at its ask, it captures $2 of gross spread.

That is not $2 of guaranteed profit. Trading fees, hedge trades and financing can change the net result. More fundamentally, the second trade is not assured. If the maker buys those tokens and the reference price falls to $9.80 before it sells, the position has a $19 mark-to-market loss before costs.

The maker also faces adverse selection: an incoming trader may act on new information or a price change before the maker updates its quote. A bid that looked competitive a moment earlier can become an expensive purchase. The spread is intended to compensate for these risks and operating costs; it is not a risk-free margin.

How a market-making strategy adjusts

The consequential part of a crypto market making strategy is not choosing an initial spread. It is deciding how price, size and inventory should change after the next trade.

Dynamic spreads and inventory-sensitive quotes

A maker can widen its spread when price uncertainty increases, incoming flow becomes harder to hedge, or hedge execution becomes more expensive. It can quote more tightly when those risks ease. Spread changes must be read alongside size: keeping a tight spread while reducing the quantity available may offer less useful liquidity than widening modestly while retaining depth.

Inventory creates a second reason to adjust. After repeated sales into its bid, the maker owns more tokens. It can move its quotes downward relative to the reference price, reduce bid size, or offer more size on the ask. This inventory-sensitive quote skew discourages further selling into its bid and encourages buyers to take inventory off its hands.

Such asymmetry is not automatically poor performance. It can be sensible risk management. But a mandate requiring meaningful buying capacity cannot be satisfied by abundant ask-side liquidity alone. It also explains why market making is not price support: a risk-aware maker can lower its quotes rather than absorb unlimited selling at yesterday’s price.

Depth, hedging and cross-venue rebalancing

Laddered orders distribute liquidity across several price levels rather than concentrating everything at the best bid and ask. In our hypothetical example, 100 tokens at the best bid tell us little about the price a 250-token sale would obtain. Additional bids further down the book can absorb a larger sale, but its average execution price will depend on that deeper liquidity—not the headline spread alone.

Hedging lets a maker reduce directional exposure without waiting for an opposite customer trade. A maker that buys tokens on one venue might sell them on another, or offset price exposure with a suitable derivative. Neither route eliminates risk. A spot hedge needs accessible liquidity and available balances. A derivative hedge can introduce basis risk—the risk that hedge and underlying prices move differently—as well as funding costs and margin requirements. A leveraged hedge can also face liquidation.

Cross-venue rebalancing keeps usable inventory where orders execute. Buying tokens on one venue and selling them on another may offset price exposure, but it leaves more tokens on the first venue and fewer on the second. Transfers can involve costs, delays or restrictions, so the ability to pre-position capital matters. A strategy that looks balanced across all venues can still run out of spendable cash or tokens on the venue where liquidity is required.

These dependencies become especially important under stress. An API outage can prevent timely updates or cancellation, leaving the maker uncertain about live orders. A venue or custodian failure can make balances inaccessible or expose them to loss. If several makers depend on the same hedge market or funding source, that shared constraint can cause liquidity to retreat across the book at once. Tight quoting in calm conditions does not establish resilience to those failures.

Order books are different from AMM pools

In an automated market maker, or AMM, traders swap against pooled assets according to pricing rules rather than match with a firm’s resting limit orders. Liquidity providers contribute assets and earn fees, but swaps change their asset mix; fee income need not offset underperformance relative to simply holding those assets. In concentrated-liquidity designs, a position can stop supplying active liquidity when the price leaves its chosen range. Smart-contract risk adds another exposure. Successful pool management is therefore not, by itself, evidence of competence in continuous order-book quoting.

Match the arrangement to the job

“Liquidity” can describe several different requirements. Before comparing providers, establish which one you are buying or undertaking.

  • Token-listing liquidity support: a continuing quoting mandate for a particular pair. The relevant questions concern two-sided size, continuity and inventory capacity, especially where trading history and hedge options are limited. The objective is a more usable market, not a defended token price.
  • An exchange market-maker program: a venue’s participation framework, potentially linking quoting thresholds to fee tiers, rebates or other incentives. Joining a crypto market maker program means operating a strategy within those rules; it is not the same as hiring a firm to supply liquidity. Eligibility also does not demonstrate performance beyond the program’s particular requirements.
  • Institutional block execution: a transaction-specific need involving a substantial trade size, an executable price and settlement terms. A record of tight small quotes does not prove that a provider can commit capital to a large block. Conversely, successful block execution does not establish an ability to maintain an order book throughout the trading day.

Buying crypto solutions for market makers is a different decision again. An internal desk may need market data, venue connectivity, execution software and position or loss controls—not outsourced trading. Those tools do not necessarily supply capital or assume inventory risk.

Whether hiring crypto market making services, joining a program or equipping an internal desk, clarify who operates the quoting engine, funds inventory, executes hedges and responds to incidents. Connectivity and market-data arrangements must support the intended venues; risk controls must cover live orders and positions, not merely whether the software is running.

What the quoting record should tell you

A provider can report a tight spread and high volume while offering very little buying capacity. The practical test is whether those attractive terms represent liquidity at the size you need. For a continuing quoting mandate, read three measures together:

  • Quoted spread: the gap between the best bid and ask, expressed in basis points of their midpoint—20 basis points in the hypothetical example above.
  • Depth: the amount available on each side within an agreed price band around a defined reference price, expressed in tokens or quote-currency value. Measure bids and asks separately: abundant offers do not help a holder trying to sell into an empty bid.
  • Two-sided quoting uptime: the share of eligible measured time when both sides meet the agreed spread and size conditions. It measures qualifying quotes, not simply whether a trading system is online.

Look for evidence from comparable assets, venue conditions and trade sizes, across both ordinary and stressed periods. A highly active asset with accessible hedges offers limited reassurance for a thin listing without them. Also establish whose liquidity is being reported: a good venue-wide spread or deep order book cannot automatically be credited to one maker.

The reporting window can be just as misleading. A whole-period average may conceal a weak bid during heavy selling, while infrequent snapshots can miss brief withdrawals altogether. Observation frequency, measurement windows and exclusions therefore need to be explicit. Venue halts or planned maintenance may be excluded from contractual uptime, but that time should remain visible: traders still experienced unavailable liquidity.

Timestamped quotes and venue order records help distinguish prices the maker intended to post from orders the venue actually accepted. Fills confirm completed trades, not uninterrupted availability. A screenshot captures one instant, and volume alone does not show the capacity available between trades. Even an accepted quote is not an execution guarantee: it may be filled or cancelled before your order arrives.

Stress-period reports are especially revealing. Does bid depth shrink while the spread still looks attractive? When do qualifying quotes disappear, and how long does meaningful two-sided capacity take to return? Inventory limits, a lost hedge route and a venue outage can produce similar symptoms; explaining the cause may require position records and incident reporting. A maker need not maintain unchanged quotes through every shock. What matters is whether its retreat follows agreed risk limits, what liquidity remains, and whether recovery restores both sides—not merely a running system.

Make the remaining risk explicit

A credible quoting record still leaves commercial responsibilities to settle. Establish who funds and owns inventory, who bears trading and hedging losses, and who pays funding costs. Identify the custodian or venue holding each balance, who has withdrawal authority, and what happens if funds become inaccessible. Your team still needs to set acceptable treasury access and counterparty-exposure limits.

Clarify the provider’s role in each service. A principal trades for its own account; an agent executes on your behalf. Neither label alone establishes total cost or safety. Understand the retainer, trading spread, explicit fees, venue incentives and inventory-loan economics that apply. Ask how those incentives affect routing and quoting decisions. Compensation tied only to volume can reward activity without rewarding durable depth.

The agreement should connect quoting obligations to reporting and escalation, including when the maker may reduce size or withdraw quotes, who must be notified, and what recovery requires. Termination terms should address open orders, remaining hedges, inventory valuation and the timing and form of inventory return. A promise of immediate return is of limited use if the assets are locked on an unavailable venue.

Reject proposals built on artificial volume or guaranteed token prices. Fabricated activity provides no evidence of usable liquidity, while market making cannot remove market-wide selling pressure or unlimited inventory risk. Those promises obscure the service you actually need: genuine two-sided trading capacity within understood limits.

Conclusion

The right arrangement connects a coherent strategy to a verifiable quoting record and clear risk ownership. Judge the liquidity available on each side, at the size you need, through ordinary and stressed periods. A tight spread matters only when there is meaningful capacity behind it.

For requesting more information about how we can help reach out to us. We're here to help and answer any questions you may have.

Contact us!

Power your growth with seamless crypto liquidity

A single gateway to liquidity with competitive prices, fast settlements, and lightning-fast issue resolution

Get started