Data retention policy

This document outlines the data retention policies for Index Exchange (Index)’s Reporting API An Index product that provides aggregated reporting data that can be delivered through email, Amazon S3 bucket, or pulled through the API. The Reporting API allows customers to create their own customized reports and schedule their automated delivery. and Reports in the Index UI The Index UI at app.indexexchange.com that allows you to manage integration settings, such as inventory, campaign, and deal settings. products.

Retention periods

Understanding the retention periods outlined below will help you access the data you need for reporting and analysis.

Hourly aggregated data

  • Most reporting data that is aggregated at an hourly level has a retention period of 100 days. This means that for the majority of reporting metrics, hourly breakdowns are available for reports covering a time frame within the last 100 days.

  • For report time frames older than 100 days, hourly aggregated data is generally not available. In these cases, daily or monthly aggregation can be used instead.

Note: An exception to the above exists for the Topics API. Hourly aggregated reporting data for the Topics API has a shorter retention period of 14 days. If you require hourly data for the Topics API, ensure your reports cover a time frame within the last 14 days.

Daily aggregated data

  • Reporting data aggregated on a daily level has a retention period of three years. For some reports, daily breakdowns will be available from the report creation date because the datasets were created less than three years ago.

Monthly aggregated data

  • There is no set retention period for reporting data that is aggregated at a monthly level. Monthly aggregated data is available as far back as Index has collected data. Monthly aggregation provides a high-level view of long-term historical data.

Factors affecting data availability

The policies above can be impacted by the following factors:

  • The date that Index started collecting a particular metric or breakout.

  • The unlikely event of upstream data loss or other data interruptions.

  • Product direction changes that result in the addition, removal, or alteration of certain data points.

If you have any further questions, reach out to your Index Representative.