Checklist

Data Governance for Learning: Controls Checklist

An actionable, fillable checklist to assess dataset readiness for learning and experiments. Covers classification, PII, consent/legal basis, anonymization, permitted uses, retention, DSAR risk, auditability, mitigations, and a lightweight approval workflow.

Interactive Tool

Data Governance for Learning: Controls Checklist

Purpose: Ensure datasets used for learning, experiments, or model training meet legal, ethical, and organizational standards while enabling safe reuse.

How to use: Complete this checklist before using a dataset for experiments. Be specific in fields describing anonymization, legal basis, permitted uses, retention, and mitigations. If you select that approval is required or request approval here, a steward or governance reviewer should confirm or update approval status.

Common name used by teams (include version or snapshot if relevant).
Person or team responsible for dataset lifecycle and access decisions.
Choose the appropriate organizational classification for the dataset.
e.g., transaction logs, customer records, sensor readings, clinician notes.
Any information that can identify a person directly or indirectly?
Summarize consent, contractual terms, or legal basis allowing reuse for learning/modeling. Reference applicable agreements or policies.
Select all that apply. If 'None' is selected, supply justification in 'Minimal anonymization steps'.
Describe applied techniques, thresholds, tools, scripts, k-anonymity or differential privacy parameters, hashing methods, or suppression rules. Include links to code or notebooks if available.
List allowed experiment types, teams, product areas, or model classes. Be specific to avoid overbroad interpretations.
How long will this dataset (and derived artifacts) be retained? Describe disposal or archival steps (e.g., 90 days after experiment end).
Estimate risk that individuals could exercise data rights or encounter re-identification. Consider identifiers and unique attributes.
Are access logs, use-of-data records, and notes of model training runs retained for auditing?
Does this dataset require review by privacy, legal, or an ethics board before use?
Select the existing approval state. Use workflow to request or confirm approval.
Name of the person (or committee) who granted approval, if applicable.
Date of approval (YYYY-MM-DD).
List required mitigations, restrictions, or conditions (e.g., limited access lists, notebook-only training, synthetic-only outputs, embargo periods).
Any other notes, links to registries, datasets, notebooks, or related policies.
If 'Yes', the dataset steward and governance reviewers should be notified to review this checklist and update approval status.
You can explore this tool now. Sign in or create an account to save your responses and return to them later.
Sign In to Save Create an Account
Make this tool part of your work

Save a personal copy, bring it to your team, or tailor the questions and workflow to fit what you are hungry to improve.

Member customization and team collaboration are coming soon.
Discuss Ask the Librarian

Discussion

Comments and conversation will live here.