What qualifies
What makes company records a potential fit for AI data licensing
Updated / Published by DATAWIN
Short answer
There is no single threshold. Fit depends on the combination of your industry, team size, how many years of records you have actually retained, the systems those records live in, whether records show work from problem to outcome, the primary language, and whether your company has the right to license them. Each program weighs these differently.
The factors that shape fit
| Factor | Why it matters | What we ask |
|---|---|---|
| Team size | Some programs publish minimum or preferred headcounts; definitions differ (full-time, office-based, W-2, total). | Current full-time count, plus U.S. desk and W-2 counts where relevant. |
| Retained history | Longer, continuous records can show how work evolved. | Years of records you still hold, separately from years in operation. |
| Systems | Records in connected tools tend to carry more context. | Which tools your team uses and whether records connect across them. |
| Record content | Records that show problem, decision, and outcome are more informative. | Record types and how often they show a task through to its result. |
| Language | Many programs currently prioritize English-language records. | Primary record language, and majority language if mixed. |
| Rights and authority | Without the right to license, commercial fit does not matter. | Ownership, known restrictions, and who can authorize a discussion. |
Does company age determine eligibility?
Not directly. Operating age and retained-record history are different things. A company that has operated for ten years may have kept only one year of usable records after a system migration. Conversely, a recently formed entity may hold older records after an acquisition. We ask about both separately and never infer one from the other.
Can a small company license its operational data?
Sometimes. Some programs publish headcount minimums, but others describe themselves as case-by-case or publish lower thresholds for specialist sectors. Smaller specialist firms, such as some law or accounting practices, or companies with an owned media library, may fit routes that larger generalists do not. We do not apply a universal headcount cutoff, and small or uncertain applicants can always submit.
Why thresholds vary between programs
Each licensing channel serves different end uses and sets its own criteria. Published wording also varies: "preferred" is not the same as "required," and descriptive phrases like "mature operations" are not numeric cutoffs. We keep that distinction visible in our review and mark vague criteria for confirmation rather than inventing numbers. See the methodology for details.
Records that tend to be useful
- Internal documentation, SOPs, and knowledge bases
- Support tickets and their resolutions
- Project and task histories
- CRM and sales workflows
- Operational communications
- Private software repositories with history and review context
For more depth on what affects value, read what affects business data value.
Frequently asked questions
Is there a minimum number of employees?
Not universally. Some programs publish minimums or preferences; others do not. We review each route separately.
What if I am not sure about some answers?
Choose "Not sure." Unknown answers are treated as unknown, never as a pass or a fail, and our team may follow up.
Does a high-quality profile guarantee acceptance?
No. Final eligibility always depends on the buyer’s own review and agreement on terms.