Company Data License Terms: Exclusivity, Term and Permitted Use

This page is general information, not legal advice. It does not describe any particular contract. Ask your own counsel.

Before you agree to anything with a data buyer, know which kind of deal you are in and which terms carry the value and the risk. Whether a deal is called a license or a sale, ask which rights are granted or transferred, which rights the company retains, and whether it may continue to use or license the data. For company data the label matters less than the terms: what the buyer may do, for how long, whether anyone else can do it, and what happens at the end.

This page goes through those terms one at a time, with the question to ask on each. It is general information, not legal advice.

Four terms that decide a data deal’s value and risk

Whether a deal is called a license or a sale, ask which rights are granted or transferred and which rights the company retains. What the buyer can actually do is set by the contract, so read the terms, not the label.
Four terms decide most of the value and the risk: exclusivity, term, permitted use, and what happens to the data at the end.
If a dataset holds personal data, settle which privacy law applies, each party’s role, the lawful basis, the notice and contract limits, and any transfer safeguards before you agree a price.
AI data buyers pay companies to license their work records, like email, chat and documents. One application checks the company against every buyer we work with. No files and no company records are needed to start.

Disclosure: OpenTrain is an independent referrer. If a buyer we match you with credits your company's referral, that buyer may pay OpenTrain a referral reward. You pay OpenTrain nothing. We never receive or hold the data your company licenses. We keep the company and contact details you give us, and we send them to a buyer only when you ask us to.

See Which Buyers Fit

Takes about two minutes. You pay OpenTrain nothing.

Licensing versus selling company data

License. Ask what permission the company gives, to whom, for what use and for how long, and which rights it keeps. Terms usually set the purpose, the length of time, who may use the data, and what must happen at the end.
Sale. Ask which rights pass to the buyer, which rights the company retains, and whether the company may continue to use or license the data afterward. The contract decides, and it may give some rights back.

Ask counsel what rights the company has in each dataset before you pick a structure.

The terms of a data license, one by one

Terms that decide the value and the risk of a company data deal, with what to ask. This is not a contract template.

Term	What it means for the company	What to ask	Question for your counsel
Exclusive or non-exclusive	An exclusive grant may bar the company and other licensees for the agreed scope. Check the field, territory, term, reserved rights and who may use the data. Non-exclusive: the company can license the same data again.	Is exclusivity asked for? For what purpose, where, and for how long? Would it stop the company using its own data?	What does exclusivity limit for us, and what would we have to give up?
Term and renewal	How long the buyer may use the data. A fixed period, a period that renews, or no end date. What happens when it ends.	Is the term fixed? Who can renew, and on what notice? Can either side end early, and for what reasons?	What can end this deal early, and what survives the end?
Permitted use	The purpose the buyer may use the data for. Do not assume a use is allowed unless the contract's grant and other terms cover it. Have counsel read the clause. For AI training, whether it covers training, fine-tuning, testing, and use in a product.	Is the purpose named in plain words? Does it cover models trained on the data? Are other uses ruled out?	Does the permitted use match what our privacy notices and customer contracts allow?
Scope of the data	Which datasets, which versions, and whether later updates are included.	Exactly which data is covered? Is later data included, and at what price?	Do we have the right to license every dataset listed?
Onward sharing and sublicensing	Whether the buyer may pass the data on, sublicense it, or give it to contractors.	Who else may see or use the data? Under what duties?	What controls do we need on anyone the buyer passes the data to?
Models and outputs	What happens to a model trained on the data and to its outputs.	Who owns a model trained on the data? Does anything come back to the company?	Do we want any limit or any right over results, and is it realistic?
Personal data and compliance	Who decides how personal data is used. The ICO's UK GDPR code describes separate or joint controllers as determining the purposes and means of processing personal data. Ask counsel which role applies to each party. For personal data, ask counsel whether the EU GDPR or the UK GDPR applies, which role each party has, and what lawful basis, transparency, security, contract and transfer safeguards the deal needs.	Does any dataset hold personal data? Which law applies? What notices or consents back the transfer?	What is the lawful basis for sharing, and do we need a data-sharing agreement?
Deletion and return	What the buyer must delete or hand back when the deal ends, and how that is shown.	Is deletion required at the end? Does it cover copies and trained models? How is it confirmed?	What deletion duties can we realistically enforce?
Warranties and disclaimers	What the company promises about the data: that it has the right to license it, that it is lawful, that it is accurate.	What would the company be asked to promise? Which promises can it make truthfully?	Which warranties are safe for us to give, and what is our exposure if one is wrong?
Liability and indemnity	Who pays if a claim arises from the data, and any cap on it.	Is there a cap? Who bears the cost of a privacy or ownership claim?	What is our maximum exposure, and can it be capped?
Audit and reporting	Whether the buyer must report how it uses the data and the company may check.	How does the company know the data is used as agreed?	What audit rights are reasonable and enforceable?
Price and payment structure	A fixed fee, recurring fees, or a share of what the buyer earns. Which is used changes who carries the risk.	How is value set and when is it paid? What happens to payments if the deal ends early?	How are payments and taxes treated?

Exclusive or non-exclusive

What it means for the company: An exclusive grant may bar the company and other licensees for the agreed scope. Check the field, territory, term, reserved rights and who may use the data. Non-exclusive: the company can license the same data again.
What to ask: Is exclusivity asked for? For what purpose, where, and for how long? Would it stop the company using its own data?
Question for your counsel: What does exclusivity limit for us, and what would we have to give up?

Term and renewal

What it means for the company: How long the buyer may use the data. A fixed period, a period that renews, or no end date. What happens when it ends.
What to ask: Is the term fixed? Who can renew, and on what notice? Can either side end early, and for what reasons?
Question for your counsel: What can end this deal early, and what survives the end?

Permitted use

What it means for the company: The purpose the buyer may use the data for. Do not assume a use is allowed unless the contract's grant and other terms cover it. Have counsel read the clause. For AI training, whether it covers training, fine-tuning, testing, and use in a product.
What to ask: Is the purpose named in plain words? Does it cover models trained on the data? Are other uses ruled out?
Question for your counsel: Does the permitted use match what our privacy notices and customer contracts allow?

Scope of the data

What it means for the company: Which datasets, which versions, and whether later updates are included.
What to ask: Exactly which data is covered? Is later data included, and at what price?
Question for your counsel: Do we have the right to license every dataset listed?

Onward sharing and sublicensing

What it means for the company: Whether the buyer may pass the data on, sublicense it, or give it to contractors.
What to ask: Who else may see or use the data? Under what duties?
Question for your counsel: What controls do we need on anyone the buyer passes the data to?

Models and outputs

What it means for the company: What happens to a model trained on the data and to its outputs.
What to ask: Who owns a model trained on the data? Does anything come back to the company?
Question for your counsel: Do we want any limit or any right over results, and is it realistic?

Personal data and compliance

What it means for the company: Who decides how personal data is used. The ICO's UK GDPR code describes separate or joint controllers as determining the purposes and means of processing personal data. Ask counsel which role applies to each party. For personal data, ask counsel whether the EU GDPR or the UK GDPR applies, which role each party has, and what lawful basis, transparency, security, contract and transfer safeguards the deal needs.
What to ask: Does any dataset hold personal data? Which law applies? What notices or consents back the transfer?
Question for your counsel: What is the lawful basis for sharing, and do we need a data-sharing agreement?

Deletion and return

What it means for the company: What the buyer must delete or hand back when the deal ends, and how that is shown.
What to ask: Is deletion required at the end? Does it cover copies and trained models? How is it confirmed?
Question for your counsel: What deletion duties can we realistically enforce?

Warranties and disclaimers

What it means for the company: What the company promises about the data: that it has the right to license it, that it is lawful, that it is accurate.
What to ask: What would the company be asked to promise? Which promises can it make truthfully?
Question for your counsel: Which warranties are safe for us to give, and what is our exposure if one is wrong?

Liability and indemnity

What it means for the company: Who pays if a claim arises from the data, and any cap on it.
What to ask: Is there a cap? Who bears the cost of a privacy or ownership claim?
Question for your counsel: What is our maximum exposure, and can it be capped?

Audit and reporting

What it means for the company: Whether the buyer must report how it uses the data and the company may check.
What to ask: How does the company know the data is used as agreed?
Question for your counsel: What audit rights are reasonable and enforceable?

Price and payment structure

What it means for the company: A fixed fee, recurring fees, or a share of what the buyer earns. Which is used changes who carries the risk.
What to ask: How is value set and when is it paid? What happens to payments if the deal ends early?
Question for your counsel: How are payments and taxes treated?

Public examples of data license terms

Public data licenses show how the same terms are written down in different ways. These are public license examples, not any buyer’s contract terms or a negotiated contract for your transaction. They make the choices visible.

CDLA-Permissive-2.0 (Linux Foundation). Section 1.1 says: “A Data Recipient may use, modify, and share the Data made available by Data Provider(s) under this agreement if that Data Recipient follows the terms of this agreement.” Section 1.2 adds that the agreement does not restrict use of any portion of the data that is in the public domain or may be used under another legal exception or limitation.
CDLA-Sharing-1.0 (Linux Foundation). Section 2.1 grants “a worldwide, non-exclusive, irrevocable (except as provided in Section 5) right to: (a) Use Data; and (b) Publish Data”, subject to the conditions in its Section 3. The license also defines “Results”, the outcomes or outputs from computational use of the data, as a term of its own (section 1.11).
Open Data Commons licenses. A family of licenses for data and databases. The PDDL places the data in the public domain, ODC-By is the attribution license, and ODbL adds share-alike to attribution.

None of these is a template for your deal. They are examples of how a term such as permitted use or onward sharing can be set.

When the data includes personal data

If the data holds personal data about customers, staff or others, the deal also has to meet the privacy rules that apply to that data. The UK Information Commissioner’s Office publishes “Data sharing: a code of practice”. It says the code “focuses on the sharing of personal data between controllers, ie where separate or joint controllers determine the purposes and means of the processing of personal data, as defined in UK GDPR Article 4(7).” The ICO’s data sharing code is guidance, and the ICO says it is under review after the Data (Use and Access) Act. Ask counsel which law applies, whether the parties are separate controllers, joint controllers or controller and processor, and what lawful basis, transparency, contract and transfer safeguards the deal needs.

What to ask every buyer about its terms

Each buyer sets its own terms, and you license your data straight to the buyer. Ask every buyer the same questions, so you can set the answers side by side:

Is the license exclusive or non-exclusive? For what field, territory and length of time?
How long does the license run, who can renew it, and what happens when it ends?
What exactly may you use the data for? Does that cover training, fine-tuning, testing and use in a product?
Who removes or transforms personal information, when, and who can see raw records?
What do you delete or return when the deal ends, including copies, and how is that confirmed?
How is the price set, and when is it paid?

Read each buyer’s terms with your counsel before you agree to anything.

What licensing your company's records means

Licensing means an AI data buyer pays the company to use copies of its work records, under a signed license that says what is shared and how it is used. OpenTrain shows you which buyers fit, and the license is between your company and the buyer.

How much a company can earn

"We're paying out $200k-2M+ for the data your business already generates"

What one AI data buyer says on its own site, October 2026.

That buyer says its payments typically range from $150K to over $2M, based on the quality, uniqueness and value of the data.

These are the buyer's own estimates, not a promise. The buyer sets any offer after it reviews your company, and it may offer less or nothing. Sources

Which records AI data buyers want

Everyday work records, like the ones below.

Email
Team chat
Documents and wikis
Customer and sales records
Support tickets
Code and project history
Step-by-step process guides

Keep personal files, customer payment data and patient records out of any license.

How personal information is handled

Every buyer we work with says on its own site that it removes or transforms personal information before your data is shared with others.
Before you sign, ask each buyer to show you in writing, with a sample, what personal information it removes or transforms, when it does so, and who can see your records before then.
OpenTrain does not remove personal information for you and cannot guarantee how a buyer does it.

Buyer A

What is removed or transformed
When
Who can see your records

ApproveApproved

Your records
Buyer explains what is removed or transformed
You review
You approve

Three steps to license your company's records

Answer a few questions about your company. Your industry, how many people work there, your role and where its records are kept. About two minutes.
See which buyers fit. We check your answers against each buyer's requirements and show you every buyer that fits, right away.
Apply, then decide. The buyer reviews your company and may make an offer. You choose whether to sign.

Each buyer makes its own decision about every company and may decline or may not contact you.

Sign a license before you share any records

First, ask each buyer to put its security and privacy terms in writing, and to confirm that nothing leaves your systems until you approve it.
Then sign a license that says what is shared, how it is used and what you are paid.
Share data only after you sign, and only what the license names. OpenTrain does not audit or guarantee any buyer's security.

Your systems

Email
Team chat
Documents and wikis
Customer and sales records
Support tickets

LicenseBuyer A

What is shared
How it is used
What you are paid

SignSigned

ApproveApproved

Buyer A

Documents and wikis
Support tickets

You license your data straight to the buyer.We never receive, hold or sell that data.

How OpenTrain gets paid

OpenTrain costs your company nothing. If a buyer credits your referral, the buyer may pay us a referral reward. You license your data straight to the buyer. We never receive, hold or sell that data. We keep the company and contact details you give us, and we send them to a buyer only when you ask us to.

Common questions about data license terms

Is a license or a sale better?

It depends on the data, the rights the company holds and what it wants to keep. Ask counsel.

What is a typical term or price?

We do not state one. We have no dated source for a typical term length or a typical price. Any range a buyer publishes is not a promise for your company.

Does the same data have to go to only one buyer?

Only if the company agrees to exclusivity. See the first row of the table.

Questions to ask your own counsel

What rights does the company hold in each dataset, and can it license or sell all of them?
Is a license or a sale the better structure for us, and what do we keep in each?
What does exclusivity limit for us, and what would we have to give up?
What can end the deal early, and what survives the end?
Does the permitted use match our privacy notices, our customer contracts and the purpose the data was collected for?
What is the lawful basis for sharing any personal data, is each party a controller, joint controller or processor, and does that need an Article 26 arrangement or an Article 28 contract?
What must the buyer delete or hand back at the end, including copies, backups and anything derived from the data, and how is it shown?
Which warranties are safe for us to give, and can our exposure be capped?
Does the data include special-category, biometric or children's data, or will it leave the country, and does that need extra safeguards or a DPIA?
Does the EU GDPR or the UK GDPR apply to this deal, and why?

See which buyers fit your company

Answer a few questions. You'll see each buyer that fits right away.