Data readiness for AI: start with the documents you already have
Most organizations do not need a data transformation to begin. They need to know which information is reliable, who owns it and who may use it.
By the Halden editorial team
A smaller problem
Begin with one workflow, not the whole estate.
Data readiness can sound like a multi-year programme. It does not have to be. For a first use case you need only the information that workflow depends on: the policies, templates, records and examples that people already use.
Take stock
List the sources and judge them honestly.
For each source, note where it lives, who owns it, how current it is and who may see it. You will quickly find duplicates, out-of-date files and gaps that people work around by asking a colleague.
- Where does the information live?
- Who owns it and keeps it current?
- Is it accurate and complete enough to rely on?
- Who is allowed to use it, and for what?
Draw the lines
Decide what must stay out.
Agree which information should never be entered into AI tools: personal data, client confidential material, unreleased financials and similar. Write the rules in plain language and put them where people work.
Clear boundaries make it easier to say yes to everything else.
Improve what matters
Fix the sources people rely on most.
Rather than cleaning everything, tidy the handful of sources a pilot needs: archive duplicates, update the key documents and assign owners. Treat it as part of the workflow, with a review date.
Good information habits built around one workflow become the pattern for the next.
A next step
Turn this into a decision for your organization.
Explore how AI Readiness works with leaders and teams, or start with a short reflection on where you are today.