Last checked: 7 October 2026 (buyer statements quoted on this page)
Sometimes, if your first contract allows it. Exclusivity decides whether you can license the same records to a second buyer. Resale rights decide whether your first buyer can pass them on. Both sit in the license grant, and both are worth reading before you take a second offer.
They are separate clauses that point in opposite directions, and a single deal can contain both.
A promise by you not to license the defined data to anyone else, for a period or without end. It is a restriction on your future deals, not on the buyer.
The key words are "the defined data". A narrow definition blocks almost nothing beyond the delivered files. A broad one can block deals you have not imagined yet.
The buyer's right to pass your data, or a processed version of it, to other companies. Data companies that supply AI labs may need this right to do business at all.
The key questions are to whom, in what form, and under which limits on use and deletion.
Most exclusivity terms fall somewhere on this line. The further right, the more you give up.
You may license the same data to others. The buyer gets a copy, not a monopoly. This is the model that leaves room for a second or third deal on the same records.
For a set period you may not license the defined data to others. When it ends, you are free again, although the data may be older and less in demand by then.
The defined data is spoken for permanently. That is the largest concession in a data deal, so it belongs in the price conversation, not in the small print.
Two clauses with the same term can block very different amounts. The breadth of "the defined data" matters more than the label on the clause.
| If exclusivity covers… | It may block | Question to ask |
|---|---|---|
| Only the delivered copy | Licensing those exact files again | Can we license other date ranges from the same systems? |
| The same records | A fresh export of the same period to anyone else | Does a de-identified or summarized version count as the same records? |
| Similar data from the same systems | Future months of chat, tickets or email you have not created yet | Is new data created after signing outside the restriction? |
| Any data of the same type | Possibly every future deal involving that kind of record | Can this be narrowed to named systems and dates? |
Not every buyer is the final user. Read the grant to see how far your data can travel.
Some buyers are data companies that deliver datasets onward to AI labs. Mode, for example, states on its site that it buys "an agreed copy" and de-identifies before onward delivery (as published, checked 7 October 2026). So the useful question is usually not whether data moves downstream, but to whom, in what form and under what limits.
The questions on the right are generic. They apply to any buyer and are not claims about any named company.
A second offer often arrives after the first contract is signed. This fictional example shows where the conflict appears.
Example Co., a 40-person software firm, licenses its Slack and Jira history for 2019 to 2025 to Buyer A, with an 18-month exclusivity covering "similar data from the same systems".
Buyer B asks for two years of Jira tickets. Example Co. assumes this is fine because Buyer B wants a different format.
The 2024 and 2025 tickets sit inside Buyer A's scope. Because the definition says "similar data", tickets created after signing may be covered too. The format does not change that.
If a company signs the second deal anyway, the first contract decides what follows. Agreements may provide for damages, termination, refunds or clawbacks of fees already paid, and indemnity claims. Even a non-exclusive first deal can create problems: a confidentiality clause may cover the dataset, or you may have promised that you made no conflicting grants.
Look backwards too. An earlier client contract may give the client ownership of deliverables, and an earlier data license may already restrict the same records. The guide to indemnities and warranties explains how those promises are usually backed.
Exclusivity is something you give up, so it belongs in the price conversation, and it needs tracking once signed.
Practitioners advise sharing a manifest and samples rather than a full dataset, and getting more than one offer. Offers are only comparable if each one says whether it assumes exclusivity, for how long and over what data. A reasonable request is to ask each buyer to state its terms for both a non-exclusive and an exclusive license. If a buyer will only work on exclusive terms, that is useful information too: it tells you what the deal costs you in future options, and it gives you a reason to compare it with a second offer before you decide.
Published payout ranges do not separate the two. micro1's referral page lists "$100K-$2M+ for approved data packages", Mode lists "$100K-$5M" and Grepped "$20K-$5M" (as published, checked 7 October 2026). Those are ranges, not offers, and none of them tells you what exclusivity is worth. See how much AI companies pay for data and getting more than one offer.
The same clause weighs differently depending on what your records contain and how long your company will be around. These examples are fictional and labeled; they show where the questions come from.
A 50-person marketing agency's client contracts give clients ownership of deliverables and limit use of client material. Before granting anyone exclusivity, the agency needs to know which records are its own to license at all. A resale right granted to a buyer could also carry client material further than the agency ever intended.
A startup closing this year licenses its Slack and email archive. Exclusivity matters less to a company that will not exist next year, but resale terms matter more: once it is gone, nobody is left to enforce use limits. Forbes, Fast Company and Gizmodo covered this closure market in 2026, and Troveo cites roughly $10,000 to $100,000 per archive deal and about $5,000 per code repository.
A 90-person software firm licenses its Git history. Parts of its repositories include open-source code under licenses with their own terms, plus contractor code. An exclusive grant over "all repositories" could reach code the firm does not exclusively control, and a resale right could pass that code on without the license terms that travel with it.
The three have one thing in common: the hard part is not the exclusivity clause itself but what sits inside the data. Client material, open-source code and the archive of a company that will soon not exist each change who can enforce a limit and who can object to it. Sort the records first, then negotiate the grant. A clause that looks generous on a clean dataset can be risky on a mixed one.
A red flag means ask why and negotiate. A green flag means the term is easier to live with later. These are generic patterns, not descriptions of any buyer's contract.
| Topic | Red flag: ask why | Green flag |
|---|---|---|
| Term | No end date | A fixed term that starts at a defined event |
| Exclusive data | "Similar data" or "any data of the same type" | The delivered copy, or named systems and dates |
| Field of use | All buyers, all purposes | Limited to one field or one type of buyer |
| Early end | None, even if the buyer breaches or stops paying | Ends on breach, late payment or non-use |
| Resale | To anyone, in any form | Named recipients or types; de-identified copy only |
| Downstream limits | Not mentioned | Use limits and deletion duties flow down to recipients |
| Visibility | You never learn who received the data | You are told, or can ask, who received it |
| Price | Exclusivity added without discussion | Priced separately from a non-exclusive option |
Generic contract risks every seller should check. They are not claims about any named company.
Most exclusivity problems are not deliberate. They come from assumptions about what the first contract covers.
A summarized, de-identified or reformatted version may still count as the same data under the definition.
If exclusivity starts at acceptance, a slow review stretches the whole period.
Even samples shared with a second buyer may conflict with exclusivity or with confidentiality terms.
Without a register, a new manager can sign a conflicting deal without knowing the first one exists.
Confidentiality, warranties and use limits still apply to a non-exclusive license.
Once data moves on, your practical control ends at whatever the flow-down terms say.
It is a promise by you not to license the defined data to anyone else, for a set period or without end. How the defined data is worded decides how much the promise blocks: only the delivered copy, the same records, or similar data you create later.
Only if neither contract stops you. A non-exclusive license is designed to leave room for a second deal; an exclusive one may not. Other clauses can matter too, such as confidentiality or a promise that you have made no conflicting grants. Have your lawyer compare the first contract with the second request before you agree to anything.
That depends on the grant in your contract. Some buyers are data companies that deliver data onward to AI labs; Mode, for example, states on its site that it de-identifies before onward delivery (as published, checked 7 October 2026). Ask whether resale or sublicensing is allowed, to whom, in what form, and under which limits.
Terms vary by deal, and buyers do not publish a market standard. What is clear is the size of the concession: no future license of that data to anyone, with no end date. Ask what the buyer pays for it, whether a time limit would work, and whether it ends if the buyer breaches or stops using the data.
Only if the definition reaches it. Read whether exclusivity covers the delivered copy, the same records, or similar data from the same systems. A broad definition can block future deals for months of records you have not created yet.
The contract decides. Agreements may provide for damages, termination, refunds or clawbacks of fees, and indemnity claims. Ask your lawyer which remedies your draft contains before you sign, not after a second offer arrives.
Check the definition and the confidentiality clause first. Sharing samples of data covered by exclusivity, even for evaluation, may conflict with the contract. Ask your lawyer before any second conversation goes beyond general terms.
Not by itself, in most cases, but some NDAs and letters of intent add a no-shop or exclusivity period during negotiation. Read every pre-contract document for those terms before you sign it, and ask how long any such period lasts.
Applying to several programs is how you get comparable terms. Exclusivity in a data license starts with a signed agreement, but read any NDA or letter of intent for exclusivity or no-shop terms before you sign it.
Independent site. Some links are referral links: if your company signs with a buyer through them, the buyer may pay us a fee. You are not charged, and we never see your data.