Source: https://docs.amadalis.ai/privacy/exports/

# Visitor data exports, schedules and retention

Open **Privacy & Data** > **Exports** in the site admin to download recorded site [datasets](https://docs.amadalis.ai/reference/glossary/#dataset), review export runs and set a [retention](https://docs.amadalis.ai/reference/glossary/#retention) period.

Checked against the product’s code on 7 November 2026.

Open **Privacy & Data** > **Exports**. A site dataset export is separate from exporting one visitor’s record or an EmDash whole site archive.

Only an Admin can export data, download its files, schedule exports or set retention. Being signed in alone does not allow a download.

## Export a range

Available datasets include events, decisions, goals, audience transitions, audiences, profiles, consent and [daily rollups](https://docs.amadalis.ai/reference/glossary/#rollup).

1.  Choose **Export a range**.
2.  Fill **From (inclusive)**.
3.  Fill **To (inclusive)**.
4.  Select the intended **Datasets**.
5.  Continue to the approval step.
6.  Read **Export this site’s data for download?**, including its range and datasets.
7.  Choose **Approve & export** only for the intended scope.
8.  Inspect that run in **Run history**.
9.  Read its status, row count and dataset scope.
10.  Open **Downloads** on that run.
11.  When the run has completed and files are listed, download one dataset, or choose **Download everything (.zip)** for the whole run as one ZIP file.

The exported records use compressed files with one data record per line in JSON format, which software can read. A queued run is not a completed download. Read its status, row count and dataset scope before using it in another system.

## Schedule and use the data

Where scheduled exports are enabled for the site, **Export schedule** offers daily, selected weekday or monthly runs. Amadalis must enable scheduled exports for the account where it is unavailable.

1.  Under **Export schedule**, choose **Set a schedule** or **Change schedule**.
2.  Choose **How often**.
3.  For selected weekdays, choose **On these days**.
4.  For a monthly run, choose **On day**.
5.  Set **At**.
6.  Review the displayed timezone.
7.  Review its destination information.
8.  Choose **Save schedule**.
9.  Read the approval.
10.  Approve only the intended schedule.
11.  Read its saved summary.

**[Warehouse ingestion](https://docs.amadalis.ai/reference/glossary/#warehouse-ingestion)** supplies instructions for loading exports into a data analysis system. DuckDB is a database tool that can inspect the files directly. The recipe for Google’s BigQuery analysis service uses an intermediate transfer. It does not read the platform’s file storage directly. These instructions do not configure the connection for you.

## Set retention

Visitor events are kept with no limit until you set a retention period. Once you set one, events older than it are deleted on the site’s regular scheduled run, but only after an archive export has covered them.

Setting a period does not mean eligible records have already been deleted or that a particular scheduled run succeeded.

1.  Review the event classes under **Retention**.
2.  Review their days.
3.  Choose **Save retention**.
4.  Read the approval.
5.  Approve only the intended change.
6.  Read the deletion run status.
7.  Check which events an archive export has already covered.

## Your next step

Check the completed run’s dates, datasets and downloaded files. Use [Visitor Data](https://docs.amadalis.ai/privacy/visitor-data/) when your next task concerns one browser record rather than a site dataset.