Sell your data to AI companies, on your terms
AI labs are paying companies for the records of how real work gets done. We value yours, check you have the right to license them, and broker the deal with buyers we have vetted.
- Free consultation. Paid only if a deal completes.
- We never take custody of your raw data.
- You approve every buyer and every term.
Data manifest
B2B software company, 140 staff
Ref 0417
- Records
- Support tickets with linked resolutions
- Volume
- 412,000 tickets, 9.4 million messages
- Coverage
- January 2016 to September 2026
- Systems
- Zendesk, Jira, Confluence
- Language
- English 94%, German 6%
Vetting record
- Authority to signChecked
- Ownership and chain of titleChecked
- Client contract restrictionsChecked
- Privacy notice and lawful basisChecked
- De-identification testIn review
Buyer interest
3 verified buyers reviewing a redacted sample
RIGHTS
CHECKED
For companies with records
Support tickets, case files, project archives, code history, call recordings. If your team has done skilled work for years, AI buyers may pay to license the record of it. Published payouts run from $20,000 to over $3 million per company.
- A free consultation, valuation and rights pre-check before you commit to anything
- Competing offers from vetted buyers, compared on the same terms
- Help with the manifest, exclusions and de-identification checks
For AI teams buying data
Private, rights-cleared data that cannot be scraped: workflow records, expert judgment, code with its history, evaluation sets. We find sellers, vet them against a written standard and hand you the evidence.
- A shortlist of sellers matched to your written requirement
- A vetting file per seller: rights, provenance, de-identification method
- Representative samples, not cherry-picked rows
Selling data: start where you are
1Exploring
Would anyone pay for our records?
2Weighing it up
Can we do it safely, and how?
3Ready to act
Who should we sell through?
What buyers say they pay
Figures each buyer publishes on its own site. Real closed deals are often lower, which is why competing offers matter.
| Buyer | Type | Published payout | Who it fits |
|---|---|---|---|
| micro1 | Training-data company | $100,000 to $500,000 (qualified partnerships), $500,000 to $3 million (large-scale or ongoing), more than $3 million (highly unique data). Agent trajectories listed at $1 million to $2 million a year | 30 or more full-time staff, mainly English records, US first, then the UK and Canada |
| Mercor | Training-data company | Not published on mercor.com/data. ddiy.co reported Mercor's referral page listing $20,000 to $1.5 million by industry (October 2026) | Not published on mercor.com/data. Target industries include legal, healthcare, finance and insurance, professional services, manufacturing, retail, real estate and logistics |
| Polyshares | Direct buyer | $100,000 to $2 million and above typical. Four published licenses range from $110,000 to $450,000 | No headcount minimum. Any company with real operating history, in any sector |
| Endless Data | Direct buyer | Up to $4 million (indicative estimate based on headcount, years of history and location) | Roughly 30 or more employees, several years of history, English documentation. US, UK and Canada prioritized |
| Handshake AI | Training-data company | $100,000 to more than $4 million (headline). Its calculator shows estimates up to $100,000 and $400,000 to $800,000 or more | Not published. The calculator asks for employees, tools used and years in business |
| Scale AI | Training-data company | $10,000 to more than $1 million per partnership (illustrative, scaling with cadence) | 40 or more employees preferred, assessed case by case |
How a deal works, start to finish
Buyers' published timelines put most deals at six to twelve weeks from first contact to payment. micro1, for example, allows itself 30 days to accept delivered data and 30 more to pay.
- 1
Free consultation
Thirty minutes on what you hold, who would buy it and what could block a sale. No data leaves your systems.
- 2
Valuation and rights pre-check
A written view of likely buyers, a realistic range and the rights issues to clear first.
- 3
Build the manifest
We help you document systems, record counts, dates and exclusions, and draw a representative sample.
- 4
Vetted buyers review
Verified buyers see a redacted sample under a mutual NDA. You approve each one.
- 5
Offers and license
We compare offers on the same terms, then you sign a license directly with your chosen buyer.
- 6
Delivery and payment
A fixed export of only the approved data, accepted against agreed tests, then paid.
Eight checks before any buyer sees your data
We do not take custody of your raw data. Buyers see redacted samples under a mutual NDA, and full data moves only after you sign a license directly with the buyer.
1Authority to sign
The person signing can bind the company, or the trustee, liquidator or board can act for a closed company.
2Ownership and chain of title
The company created or acquired the records and holds the rights to license them, including IP assignments from staff and contractors.
3Contract restrictions
Client contracts, NDAs, vendor terms and data processing agreements do not prohibit the proposed use.
4Privacy notice and lawful basis
The privacy notice in force when the data was collected, and the applicable law, allow the disclosure, or the personal data is excluded.
5Third-party content
Licensed content, open-source code and material owned by others is identified and either cleared or removed.
6De-identification method
Direct and indirect identifiers are removed by a documented method, and a re-identification test is run on a sample.
7Representative sample
The sample is drawn at random across the archive, not hand-picked, so a buyer can judge the whole package.
8Manifest accuracy
Record counts, date ranges, systems and languages in the data manifest match what is actually delivered.
Buying data: start where you are
1Exploring
What data would actually help our model?
2Weighing it up
How do we buy it without inheriting risk?
3Ready to act
Who can source it for us?
Sell data by industry
What buyers want from your sector, what it can be worth and what you cannot license.
- Accounting and tax firms
- Auto dealerships
- Call centers and BPOs
- Construction companies
- Consulting and advisory firms
- Cybersecurity companies
- Dental practices
- Design and web agencies
- Engineering and architecture firms
- Financial services firms
- Game studios
- Government contractors
- Imaging and diagnostics centers
- Insurance companies and brokers
- Law firms
- Logistics and freight companies
- Manufacturers
- Market research firms
- Marketing and SEO agencies
- Media companies and video archives
- Medical practices and clinics
- Publishers and education companies
- Real estate firms
- Recruitment agencies
- Retail and ecommerce businesses
- Software companies
- Telecom and utility companies
- Translation and localization agencies
- Universities and research institutions
- Veterinary practices
Buy data by type
What each kind of training data is used for, how it is priced and what to check before you sign.
- Workplace communications data
- Customer support ticket data
- Source code datasets
- Agent trajectory data
- Evaluation datasets
- RLHF and preference data
- Expert domain data
- Speech and audio data
- Video training data
- Image datasets
- Multilingual text data
- Document and form data
- Call recording data
- CRM and sales data
- Egocentric and robotics video data
- Game and simulation data
- 3D and CAD data
- Low-resource language data
- Survey and market research data
Guides to selling and buying AI data
All 63 guidesSelling data
How to sell your company's data to AI companiesA step-by-step guide to licensing company records to AI buyers: who buys, what sells, the rights checks, published payouts, timings and how to avoid weak deals.
Pricing and value
How much is your data worth to AI companies?Published AI buyer payouts, real closed deals and the factors that move price, from years of history to exclusivity. Work out a realistic range for your data.
Selling data
What data do AI companies buy?The company records AI labs and training-data firms buy in 2026, from workplace chat and support tickets to code history and agent traces, and what buyers skip.
Selling data
AI data licensing explained: license, sale or revenue shareHow AI data deals work: one-off licenses, perpetual exclusive deals, revenue shares and outright sales, with buyer examples and the rights each grants.
Legal and compliance
AI data licensing agreement checklistA clause-by-clause checklist for sellers reviewing an AI data license: scope, derived assets, de-identification, payment, deletion, warranties and liability.
Selling data
Exclusive vs non-exclusive data licensesHow exclusive and non-exclusive AI data licenses differ, what each pays, which buyers use which, and middle options such as field-of-use and time-limited terms.
Questions sellers ask first
Can I sell my company's data to AI companies?
What kinds of company data do AI buyers want?
How much is my company's data worth to AI companies?
Do I need a minimum number of staff to sell data?
Should I offer an exclusive or non-exclusive license?
Do I keep my data after licensing it?
Can a closed or bankrupt company sell its data?
Will my clients find out that I licensed data?
For data owners
Ready to see what buyers will offer?
Book a consultation and we will scope your data, explain our fee in writing and set out the steps to competing offers from vetted buyers.
For AI teams
Ready to source?
Book a consultation to brief us. We shortlist vetted sellers, each with a vetting file covering rights evidence, provenance and a representative sample.