Example of a data architecture policy

image

I’ve encountered many policy documents that list roles, responsibilities, and policy statements but often feel empty and ineffective. In a conversation with someone in corporate policy, I was guided towards a more effective format that is both simple and truly serves the intent of a policy: guiding decision-making.

Effective Policy Format:

  1. Context: Background of the policy.
  2. Decision: The agreed guideline to handle the situation in the context.
  3. Consequence: The ramifications of not adhering to the policy.
  4. Architecture Guidelines: The impact on the architecture.

I’ll present this format with a simple example below.

Context

  • Our organization handles various types of customer data, including sensitive personal information.
  • To comply with data protection regulations and ensure customer trust, we must implement strict data retention policies.

Decision

  • Customer data will be retained for a maximum of seven years from the date of the last transaction.
  • After this period, the data will be securely deleted from all storage systems.

Consequence

  • Failing to adhere to this policy will result in data being kept longer than necessary, potentially leading to non-compliance with data protection regulations, increased storage costs, and higher risk of data breaches.
  • Compliance audits will be conducted regularly, and any discrepancies will result in disciplinary actions and mandatory retraining for responsible employees.

Architecture Guidelines

  1. Data Identification: All datasets containing customer data must be classified.
  2. Storage Methods: Detail how and where all copies will be securely stored.
  3. Approval for Usage: The customer data steward must approve all consumption and copies of customer data.
  4. Retention Periods: The start and end dates of customer data must be clearly identified.
  5. Disposal Procedures: Procedures for erasing or destroying data at the end of its lifecycle.
  6. Data Formats: All customer data needs to be stored in consistent and approved formats.
  7. Backup & Archiving: Establish procedures for creating redundant copies and long-term storage.
  8. Access: Only approved personnel with the right roles and access clearance can access customer data.