You are here:
Data Retention Guidelines
In Marketing Cloud Engagement, data retention refers to a process that periodically removes information from data extensions. It helps keep data extensions smaller and more efficient by retaining only the most useful or relevant data that marketers want to use.
How It Works
Each data retention job runs for 4—10 hours, and doesn’t remove data outside of its scheduled runs. A job removes as many qualified records as possible during its run time and then pauses at the end. The next job picks up where it left off. As a result, large batches of records can require multiple job runs to finish removal.
Here are a few other details to understand about data retention in Marketing Cloud Engagement.
- Records aren't always removed on the same day that they meet the retention criteria. A job runs one time per day. The job removes any leftover records from the last run and the records that met the criteria within the previous 24 hours. If a record reaches the limit on the same day, but after the job runs, it's flagged for removal during the next job.
- Data retention isn’t considered a critical function. If your org is running up against performance limits, retention jobs are delayed.
- Due to the occasional unpredictability of data retention runs, we don’t recommend using data retention policies for critical data management. For example, tools such as Send Filtering, Clear Data, and Contact Delete are better suited for preparing segments for message sending or conducting GDPR and consent-related tasks.
Best Practices
These best practices can help you manage your data more effectively.
- Not all data extensions need a data retention policy. For example, if you manage a primary customer list, it makes sense to keep it as is.
- For data extensions that need a data retention policy, configure your settings as soon as possible.
- For logging data extensions, which grow quickly, we recommend defining shorter retention periods.
- Define retention periods that are short but meaningful for your business needs. Use this rule: determine how long you can actually make use of that data, and then double it. Or, try a retention policy of 90 days, and monitor performance and user needs.
- If you need to retain data for many years, export the oldest data from your data extensions for safekeeping.
