Privacy by design in practice
Privacy by design means decisions are made before data is collected. For each new feature we ask what personal data is needed, for what purpose, for how long, who can access it and what the least invasive way to achieve the goal is. Often a coarse location, a hashed identifier or an aggregate is enough. The review is designed to take minutes for low-risk changes and to escalate only the features that profile people, use sensitive categories or involve children, so it does not become a bottleneck.
We turn these questions into a lightweight review that fits sprint planning, with templates for product managers and checklists for engineers. Defaults matter: collection off unless needed, short retention, restricted internal access and clear user controls built into the first release, not retrofitted. When a change introduces a new vendor, new analytics or an AI model that sees personal data, the review also covers contracts, data location and whether users have been told clearly.
