Use this recommended sequence when a commercial team discovers a role mentioning content acquisition, licensing, evaluation or external datasets.
- Record the source precisely. Save employer, exact title, canonical URL, retrieval date, displayed location, any posting date, and whether the application route is present. Keep a short factual extract separate from your interpretation.
- Identify the object being bought. An archive licence, expert-written task, data annotation service and commercial software product are different opportunities. Do not treat all of them as “AI data demand.”
- Find the closest content cohort. Describe subject coverage, modality, date range, update frequency and rights evidence. Avoid beginning with a total file count that combines usable and unresolved material.
- State a falsifiable fit hypothesis. For example: “Our licensed chemical reference tables could address this stated scientific-data focus.” The hypothesis must remain provisional until a buyer confirms the need.
- Ask through the appropriate business route. Use a published partnerships or procurement route, or a permissioned introduction. A recruitment application is not a content sales channel, and an employee's existence does not imply authority to sign a licence.
- Request a qualification response before preparing extensively. Establish the intended use, evaluation criteria, counterparty, procurement route, budget process and next decision. A polite meeting without these facts is still exploratory.
- Update the evidence stage. Record “job signal,” “qualified discussion,” “written requirement,” “evaluation,” “signed agreement” and “cash received” separately. Never upgrade a lead because a supplier puts a familiar logo in a presentation.
A useful commercial report should show failed hypotheses as well as promising ones. If a role concerns commissioned human feedback, closing an irrelevant archive lead is progress: it prevents spending on the wrong preparation programme.