Run a first investigation you can explain afterward.
Use a target you own, are authorized to assess, or that is an intentionally public example. The goal is a reviewable evidence trail, not the largest possible result set.
1. Write down the authorization and question
Record why you may investigate the target and what decision the research is meant to support. Avoid unrelated people or identifiers.
2. Start narrow
Choose the smallest relevant database fields or OSINT modules. Broad collection adds noise and can make false associations more likely.
3. Review coverage before conclusions
Note which providers returned evidence, which returned no match, and which were unavailable. A missing result is not proof that an account or exposure does not exist.
4. Inspect candidate matches
Compare timestamps, aliases, locations, URLs, and other context. Treat each match as a candidate until independent evidence supports the link.
5. Export the reviewable result
Filter to the evidence relevant to the question, export structured JSON or CSV if needed, and keep source context with the finding.
6. Write down what is still unresolved
Record the contradictions, coverage gaps, and unanswered questions next to the evidence. If the workflow is one you will repeat, the worked API examples in the product show how to run the same collection through /api/v1.