A market data license is a contract about use, not about access to numbers. Delayed data can be free to view on a screen while machine use of the same data is charged. A fee can apply again when you turn the data into a new figure or hand it to someone else. Exchanges write these distinctions down. Nasdaq’s European markets data policy, version 1.5.4 of April 2026, is a clear public example. It separates display use, non-display use, derived data and redistribution. Each is priced according to what the customer does.
This article walks through those four uses with that policy as the model. The categories come from one exchange’s published policy. A crypto data vendor writes its own terms, so read the definitions in the contract in front of you rather than assuming they match.
Display use: a person at a screen
Nasdaq defines display data as market data provided or used through a monitor or screen that is human readable. A trader looking at a quote, an analyst reading a chart or a dashboard that shows a price to its user all fall here.
The policy is specific about what stays inside display use. An application that updates a portfolio and shows the result on a screen is not treated as non-display. Neither is a calculation such as a volume-weighted average price when its output goes to a display alone. That second example carries a warning in the policy text, though: a derived license may still apply.
Non-display use: a machine reads it
Non-display use is everything that is not display use. The policy describes it as automated access or use by a machine, without a natural person using a display. For its Nordic and Baltic markets Nasdaq splits it into two categories:
- Category 1 covers non-trading calculation and verification, with portfolio valuation, administration, risk calculations, analysis and fund management as examples.
- Category 2 covers trading-based activity, including order routing and fully automated trading. It includes Category 1 as well.
Two details matter to engineering teams. For trading infrastructure such as order and execution management systems, the policy makes non-display use fee-liable whether the system runs in the cloud, in a data center, across an enterprise or on one person’s desktop. The unit of count for non-display is the customer. A firm unsure about a use case can submit it to its Nasdaq representative for a determination, so the vendor settles unclear cases.
Derived data: a new figure from licensed ones
Derived data is a figure computed from licensed data. Nasdaq treats it in two ways:
- A single-security figure that contains price or quote data will generally be fee liable at the underlying product rates.
- A multi-security figure that contains price data, such as a mutual fund valuation, will generally not be liable at the underlying rates. The customer pays a derived data fee instead.
The policy also describes a license for new original works. It gives the customer the right to use real-time data to derive new works based on multiple securities for unlimited distribution to controlled products. The condition is that the new works cannot be reverse engineered back to the underlying data.
That condition shows how broad the definition of the data is. Nasdaq’s definition of “Information” reaches processed data from which the information, or a substitute for it, can be identified, recalculated or re-engineered. Under that definition, a figure computed from licensed data can still fall within the license, which is why derived data carries its own terms.
Redistribution: giving it to someone else
The policy defines redistribution as providing a controlled or uncontrolled product. It defines “external” as distribution outside the customer’s own entity. A data feed is an uncontrolled transmission in which the customer does not control both the entitlements and the display. APIs and similar solutions are generally data feeds. Distributors that use APIs need pre-approval.
For a protocol, wallet or fintech team the practical question is simple. Does your product show licensed prices to your own users? Does it send them through your own API? Does it feed them into a contract that other people rely on? Each of those can take the data beyond internal use, so check how the contract treats it. A contract that covers internal use may say nothing about these cases.
The policy adds a second axis: who is using the data. A professional customer is one that uses market data for a regulated financial activity or to provide a service for third parties, or one that meets its size thresholds.
Where the contract sits
In Nasdaq’s structure the customer signs a market data license agreement with Nasdaq and is invoiced under it. The policies sit beneath that agreement. Nasdaq says it will update them periodically, so a use that was clear last year deserves a fresh read.
Why a crypto holder needs a documented price
Licensing governs how you may use a price. Accounting governs why a treasury or finance team needs one at all. In December 2023 the Financial Accounting Standards Board issued Accounting Standards Update 2023-08, which created Subtopic 350-60 for crypto assets.
Under the update, an entity measures in-scope crypto assets at fair value in the statement of financial position each reporting period. Gains and losses go through net income. Before it, these assets were treated as indefinite-lived intangibles at cost less impairment, where increases in value were not recognized. The update applies to all entities for fiscal years beginning after December 15, 2024, including interim periods.
An asset is in scope when it meets all six criteria in the update. It must be an intangible asset with no enforceable rights to underlying goods, services or other assets. It must be created on or reside on a distributed ledger based on blockchain or similar technology. It must be secured through cryptography and be fungible. It must not be created or issued by the reporting entity or its related parties. The update does not say which tokens qualify, so that question is settled asset by asset with the entity’s accountants.
The update also does not mention data vendors. The link is indirect. Fair value is measured under the existing fair value guidance in Topic 820, which the Board judged sufficient for crypto assets. Among the topics the Board listed there is the use of quoted prices provided by third parties. A finance team that holds in-scope assets needs a price for each reporting date. The disclosures ask for items such as the fair value and number of units of each significant holding. A third-party price with a stated method and a timestamp is something a team can document for that purpose.
Why institutions pay for data they could compute
Any team with enough engineers can read pools and exchanges to compute a price. Buying one instead is a question of accountability more than of arithmetic.
A number a team computes itself is a number that team has to defend: which sources went in, which moment it describes, why an outlier was set aside, what changed since last quarter. A vendor figure comes with a stated method and a timestamp that the vendor documents. When a risk committee, a counterparty or a reviewer asks where a price came from, the question has somewhere to go. That reproducibility is what makes a figure usable in a contract, a collateral calculation or a reporting package.
Display, non-display, derived use and redistribution each describe a different way of relying on the vendor’s figure, so each can carry its own terms.
Questions to ask before signing
These questions work for any vendor, ML Data included. The longer buyer checklist is in How to Evaluate a Crypto Data Provider.
- Which uses does the plan cover: display, non-display, derived or redistribution?
- Can prices be shown to your own users in a product, or are they for internal use?
- Are calculated results, such as a portfolio value or a collateral ratio, covered?
- If your system is automated, does the license count it? Does it matter where it runs?
- Who counts as the customer: one entity, a group or each application?
- What happens to your right to use the data when the plan ends?
- How are unclear cases decided? Can you get the decision in writing?
Where ML Data fits
ML Data is a data provider for DEX-native prices: it measures on-chain market state and serves it over an API. The method is documented on the methodology page. DEX-Native Price Feeds, Explained describes how an on-chain price is measured. The Trust Center states that ML Data is operated by Matrixedlabs Technologies - FZCO, whose information security management system is certified to ISO/IEC 27001:2022. Why ISO 27001 Matters for Web3 Data explains what that kind of certificate covers.
Plans are listed on the pricing page. If your use needs larger or custom terms, the Enterprise page describes how those engagements work. The contact page is where to start the conversation. Bring the questions above and your use case.
