Explore what your records might make possible, what to prepare, and what to review before sharing anything.
Explore whether support histories, software documentation, and operating procedures could support a licensing discussion. Start with a company overview, without exporting records.
Read the guide
How to sell or license company records to AI companies: what buyers want, what public deals have paid, the legal checks and how a first conversation works.
Read the guide
What AI companies have paid for company data, from a reported $5,000 per book to $10 million for Spirit Airlines’ records, and what sets the price for yours.
Read the guide
Review a data-licensing proposal before granting access: scope, recipients, permissions, costs, payment conditions, and what happens when the agreement ends.
Read the guide
Often yes, if the company holds the rights. What privacy law, contracts, copyright and the 2026 Spirit Airlines data sale mean for licensing company records.
Read the guide
AI developers, industry AI companies and data programs buy company data. What public deals from Reddit to Spirit Airlines show, and how to reach a buyer.
Read the guide
What a data licensing agreement covers, clause by clause: scope, permitted use, AI training and models, personal information, payment, term and deletion.
Read the guide
How a B2B company makes money from its data: improve operations, add data to what you sell, or license it. Examples by industry and what licensing requires.
Read the guide
A sourced tracker of public AI data licensing deals, including the rare sales of company records such as emails, chats and Slack archives. Updated October 2026.
Read the guide
Which company records AI developers buy, what each type has sold for in public reports, the main legal risk for each, and how much preparation it takes.
Read the guide
Yes, and buyers are paying. What Slack, Teams and email archives have sold for, what makes one valuable, and the employee privacy checks to run first.
Read the guide
What private code bases have sold for, why commit history and linked issues matter more than lines of code, and the ownership and license checks to run first.
Read the guide
What call audio is worth to AI developers, why "recorded for quality and training" consent may not cover it, and the voice, payment card and privacy checks to run.
Read the guide
Why work orders, technician notes and maintenance logs interest AI developers, what makes a service history valuable, and what to remove before licensing it.
Read the guide
An issue-spotting framework for advisors whose clients are asked to sell or license company data to AI developers, including wind-downs and bankruptcy sales.
Read the guide
How company data licenses are priced: one-time fees, annual or per-delivery payments, usage-based terms and revenue share, with the trade-offs of each for the seller.
Read the guideThe difference between exclusive and non-exclusive company data licenses, how exclusivity can be narrowed by field, term or use, and the questions to ask before granting it.
Read the guideA practical overview of de-identifying company records before licensing: removing direct identifiers, handling free text, re-identification risk and documenting the method.
Read the guideHow privacy laws such as the GDPR and California’s CCPA affect licensing company records for AI training: personal data, lawful basis, notices, sale rules and practical steps.
Read the guideHow HIPAA affects licensing healthcare company records for AI, the difference between PHI and operational data, de-identification standards and the questions to settle first.
Read the guideThe difference between licensing company records for AI model training and for retrieval (RAG), how rights, deletion and payment differ, and how to define use in the agreement.
Read the guideClosing a business? Whether its Slack, email, tickets and code can be sold to AI companies, what they fetch, and the privacy and contract checks first.
Read the guideHow Mercor's company data program works: tools it connects, target industries, published payouts of $20K to $1.5M, payment timing and eligibility.
Read the guideHow micro1 pays for company data: $100K to $2M+ tiers, the 30-employee and English requirements, payment timing and which countries it prioritizes.
Read the guideHow Handshake AI buys company data: the $100K to $4M+ range, 20-employee and 3-year minimums, eligible countries and payment terms.
Read the guideTuring's Project Lazarus pays up to $1M for a company's full operating record, or $10K to $100K+ for a codebase. Who qualifies and how fast it pays.
Read the guideCompare micro1, Mercor, Handshake AI, Turing and Scale AI: published payouts, size minimums, countries, language and payment timing.
Read the guideCan companies in the UK, Canada, Australia, India or the Philippines sell data to US AI buyers? Who accepts you, pricing, tax and privacy steps.
Read the guideYes, AI companies pay businesses for records. How to verify a buyer, the red flags to watch for and what to ask before sharing anything.
Read the guideTroveo, Nyne, Harbor, Aribase, FileYield, Main Street Wealth and direct buyer programs compared: how each is paid, what it does for you and who it fits.
Read the guideOriginal data: published payouts, size minimums, payment timing and referral fees across 8 company-data programs and all 24 Mercor listings. Free CSV download.
Read the guideDirect answers on selling company data to AI companies: what it pays, which records sell, whether Slack, email and support data are legal to license.
Read the guidePractical guidance from Origin Data Partners, with primary sources for the rights questions.
Explore more examplesWhich AI data buyers accept your company, published payouts in your currency, and the local rules for your industry, across 10 countries and 12 industries.
Choose a starting point. Explore an example, then describe the history your own company retains.
Start with one repeatable project type. Which decision, check, and outcome records remain connected?
Explore implementation historySeparate a test description from the fixture used to run it. Check which product versions and outcomes remain traceable.
Explore test and QA historyDescribe closed reliability reviews at company level. Keep security-sensitive details and raw logs out of an initial conversation.
Explore incident reviewsIdentify the internal investigation and resolution notes separately from customer messages and attachments.
Explore support historyBegin with one product area. Note which versions, authorship questions, and links to decisions remain available.
Explore documentationFollow one of six illustrative scenarios about a software team's own work. The industry examples are not an eligibility list.
Find an industry exampleThese are ways to recognize records, not a licensing assessment. Keep the underlying files inside your company.
Licensing is one possibility to investigate. Compare it with other assets you already have, then review the work, rights and payment terms a real opportunity would require.
Compare revenue opportunitiesNew income can help only if it exceeds the preparation, review and delivery costs. First distinguish a margin improvement from a capacity saving or a possible future payment.
Review the margin questionsSupport investigations, engineering decisions, project histories, and operating procedures are categories to describe. Having those records does not establish an offer.
Recognize your record categoriesBegin with the resolution history: investigation, decision, fix, and outcome. Consider internal notes, customer messages, and attachments separately.
Explore a support-history exampleDescribe specifications, design decisions, issue histories, and runbooks. Keep documentation and source code separate when framing a possible scope.
Explore a documentation exampleThere is no standard price to infer from file count or company age. If an offer emerges, compare the proposed use, company effort, restrictions, and payment conditions.
Review value and offer questionsYes. An initial Origin conversation starts with company details and broad record categories. Keep the underlying files, customer messages, and credentials in your systems.
Map one workflow without an exportIdentify the recipient, proposed use, scope, permissions, handling, and terms. Approval of an introduction and permission to access records are separate decisions.
Use the review questionsOrigin helps organize the records you describe, open questions, and a possible next step in a Company Licensing Brief. You can begin without a prepared dataset.
See an illustrative company briefThese guides use illustrative examples. They help you frame a first conversation, not determine eligibility or value.
Startups and companies closing, selling assets or in a wind-down, with Slack, email, code and ticket archives.
Explore this business typeManaged service providers with years of ticket history, runbooks and resolution notes.
Explore this business typeHVAC, plumbing, electrical, and equipment service companies with years of job and diagnostic history.
Explore this business typePlants with procedures, quality investigations, and maintenance records.
Explore this business typeFirms with internal procedures, review checklists, and training material, kept apart from client records.
Explore this business typeCarriers, brokers, and warehouses with exception handling and operating procedures.
Explore this business typeBlank worksheets and illustrative examples. No email required. Keep completed notes inside your company.
Document meanings, formats, missing values and the team who can verify them.
Open the free CSV builder and example Download blank dictionary · CSVOrganize categories, retained history, responsible teams, and known gaps.
See the inventory example Download blank inventory · CSVWrite down the company question, record categories, and points still to resolve.
See an illustrative company brief Download blank brief · CSVAssign a reviewer to each open question before granting access.
See the decision log Download review checklist · CSVRecord your position on scope, AI training, deletion and payment before a draft arrives.
See the clause checklist Download agreement checklist · CSVCompare the requested use, company effort, obligations, and payment conditions.
See the offer worksheet Download blank comparison · CSVA six-system comparison of documented content, scope and history limits. Includes the complete source table.
Read the export research noteAI companies pay $20K to $5M for company records. Six questions, no files.