# Data protection impact assessment (DPIA) · Glossary
A structured assessment of how a planned processing activity could affect individuals’ privacy and what measures will reduce those risks.
[Glossary](/glossary) · Data privacy and security

# Data protection impact assessment (DPIA)

     A structured assessment of how a planned processing activity could affect individuals’ privacy and what measures will reduce those risks. Under UK GDPR, a DPIA is required where processing is likely to result in a high risk to individuals.

## Why it matters

     New technology, large-scale profiling, automated decisions and the use of sensitive data frequently trigger the requirement for a DPIA, which means many AI projects need one. Completing it early shapes design decisions; completing it after build usually means rework.

     A good DPIA is a working document owned by the project, not a form signed at the end. It should describe the data flows, the necessity of each element, the risks to individuals and the specific controls that address them.

## In practice

     For example, a UK local authority planning to use a model to prioritise housing repair requests might complete a DPIA before development, leading it to exclude certain household characteristics, add human review for low-priority classifications and publish a plain-English notice to residents.

## Where Rodan fits

     Rodan supports DPIAs with the technical evidence they need, including data flow diagrams, model documentation and control design, as part of [AI and Decision Systems](https://rodan.io/what-we-build/ai-decision-systems) delivery.

## Related terms

- [Personally identifiable information (PII)](/glossary/personally-identifiable-information)

- [Automated decision-making](/glossary/automated-decision-making)

- [Data minimisation](/glossary/data-minimisation)

- [AI governance](/glossary#ai-governance)

- [Responsible AI](/glossary#responsible-ai)

- [Anonymisation](/glossary/anonymisation)
HTML: https://rodan.io/glossary/data-protection-impact-assessment