Data retention

Ada’s data retention approach follows industry practices designed to meet operational, privacy, and compliance needs. We retain the resources you make available to the AI Agent and end user data only as long as necessary to provide our services. End user data is automatically deleted after two years, while other categories of data may be retained longer, in accordance with our retention schedules.


Why we store data

We retain customer data only as long as needed to deliver services effectively and securely. Data storage allows customers to:

  • Review and analyze conversations with their end users
  • Keep platform content (e.g., knowledge base articles) available without needing to re-upload
  • Maintain authorized user accounts and access securely

Once the data is no longer required for these purposes, it is automatically deleted or has its personal details removed.


Deleting data on demand

In addition to automatic retention rules, customers can delete user data at their own pace using the Data Compliance API. This provides flexibility to:

  • Remove specific end user data whenever needed
  • Meet organizational retention requirements
  • Reduce the amount of sensitive information stored in the system

Example: If an end user requests their data be erased, the API can be used to find and remove it right away—without waiting for the default retention period.


Best practices

  • Delete data on demand: Leverage the Data Compliance API to delete data on demand when users request erasure (e.g., under GDPR/CCPA) or when your own policies require stricter retention.
  • Export before deletion: If you need to keep records for longer than the retention schedules allow (e.g., for compliance or auditing), export transcripts or other data before they expire using the Data Export API.
  • Use redactions where appropriate: Configure redaction settings to remove sensitive details (like Social Security numbers or other PII) from transcripts before they are stored. This reduces the volume of personal information retained.
  • Review access to exported data regularly: Make sure only authorized users with a business need can access exported data, especially sensitive information like transcripts and metadata.
  • Stay aligned with your internal policies: Your organization may have its own rules for how long data must be kept, based on laws, industry standards, or internal governance. Compare our retention schedules with your requirements to ensure data isn’t stored longer than allowed or deleted sooner than needed.

Where to find details

For complete information, see our Trust Center: Customer Data Retention Policy.


FAQs

Q: Can Ada keep customer transcripts for more than two years?
A: Not if they contain personal data. Transcripts are automatically deleted after two years unless personal details are removed. Customers can always export transcripts before deletion for record-keeping purposes.

Q: Do LLM providers keep any customer data?
A: No. Ada applies strict contractual and technical safeguards to ensure that data shared with LLM providers is not persistently stored. Data is processed only in memory and retained only for the duration of the request, following Zero Data Retention (ZDR) principles.