Helen is Marketing Team Lead at Motion Trade with 4+ years in Web3 and crypto marketing. Before joining Motion Trade, she built and led the marketing function at CLS Global and managed social media campaigns for a portfolio of crypto clients at Ninja Promo. She specializes in turning complex trading products into clear stories.
The bid-ask spread is the distance between the highest price a buyer is currently willing to pay and the lowest price a seller is currently willing to accept. It is normally quoted in basis points as a percentage of the midpoint, so a token trading at a midpoint of one dollar with a best bid of 0.9995 and a best offer of 1.0005 has a ten basis point spread. The spread is the immediate round-trip cost of trading and a standard measure of market quality, but on its own it is easy to make look good: a single small order on each side can produce a tight spread over a book with no depth behind it. Spread and depth have to be read together.
Spread is the first liquidity metric most founders encounter and the one most often quoted back at them in a proposal. Understanding how it is calculated, what moves it and where it stops being informative is the difference between evaluating a market making arrangement and accepting a screenshot.
How it is calculated
Take the best bid and the best ask from the order book. The midpoint is their average. The absolute spread is the difference between them, and the relative spread, which is what people mean in practice, is that difference divided by the midpoint, expressed as a percentage or in basis points where one basis point is one hundredth of a percent.
Two refinements matter when reading reports. A point-in-time spread is a snapshot and can be cherry-picked, while a time-weighted average spread measured across a period is the meaningful version. And an effective spread, measured against the price at which trades actually executed, captures what traders really paid rather than what the screen displayed. When a desk reports spread performance, ask which of these it is reporting and over what window.
Why it exists at all
The spread is compensation for risk. Whoever quotes both sides is holding inventory between the moment they buy and the moment they sell, exposed to the price moving against them, and is also exposed to the possibility that the person trading against them knows something. The wider the uncertainty, the wider the quote has to be for the activity to make sense. This is why spread widens during volatility rather than despite it, and why an unusually tight quoted spread on a volatile, thin token should prompt a question rather than satisfaction.
What widens it
Volatility is the largest single factor, since inventory risk rises with the size of plausible price moves. Thin float and concentrated holdings are next, because a small number of holders capable of moving the price make quoting more dangerous. Venue quality matters, as fragmented or low-activity venues have fewer participants competing to narrow the quote. Fee structure plays a role, since maker and taker fees set a floor beneath which quoting is unprofitable. Unlock events widen spreads in advance as market participants anticipate supply. And the absence of a committed quoting arrangement widens it structurally, which is the situation described in what happens to a token with no market maker.
What a healthy spread looks like
There is no universal number, and anyone who offers one without qualifying it by token tier and venue is not being careful. What is defensible is the relative picture. Major assets on major venues trade at spreads measured in low single-digit basis points. Established mid-cap tokens on tier-one venues sit meaningfully wider. Newly listed small caps on mid-tier venues sit wider again, and the reasonable target is a spread that is stable and commercially usable rather than one that matches a blue-chip asset. A newly launched token quoting a blue-chip spread is usually doing so over almost no size.
The right question to put to a desk is not what spread it will hold but what spread it will hold at what depth for what percentage of the time. All three elements are needed for the commitment to mean anything, and the framing is covered in what should be in a market making agreement.
Why spread alone misleads
Place one small buy order just below the midpoint and one small sell order just above it and you have produced a very tight spread on a book that cannot absorb a single real trade. The spread metric is satisfied and the market is unusable, because the first genuine order consumes both of those quotes and executes against whatever sits behind them, which may be nothing.
This is why depth is the companion metric and arguably the more important of the two. Depth measures the cumulative size resting within a defined band around the midpoint, typically expressed as the notional value available within a given percentage of mid, and it is what determines whether a trade of real size executes near the displayed price. The measurement is set out in order book depth explained.
What the spread costs you in practice
For a trader, the spread is the immediate cost of a round trip. Buying at the offer and selling at the bid with no price movement in between loses the spread, so a fifty basis point spread costs half a percent before fees on a trade that went nowhere. Wide spreads therefore suppress the activity that would narrow them, which is the loop that traps illiquid tokens.
For a project, the cost is indirect and larger. Wide spreads deter the market participants a token needs, they make exchange listing reviews uncomfortable, they make the price chart look erratic for reasons unrelated to the project, and sustained poor spread quality is one of the metrics venues watch when deciding whether a listing continues to earn its place, as covered in why tokens get delisted.
How to read a spread report properly
Ask for time-weighted spread rather than snapshots, over the full reporting period rather than selected hours. Ask for it alongside depth at defined bands, so you can see whether the tight quote had anything behind it. Ask for the uptime percentage, because a commitment held ninety-nine percent of the time and one held seventy percent of the time are different products regardless of the headline number. And look specifically at the spread during the volatile hours, since that is when the service is being tested and when a thin arrangement stops quoting altogether.
A desk that reports all four without being asked is telling you something useful about how it operates. A desk that reports a single tight number is telling you something too.
