> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/feature-management-experimentation/use-fme/feature-management/monitoring-and-analysis/export-data.md).

# Export data

The data export functionality within the Data hub gives development teams the ability to query and export all impression and event data. With data export, development teams can:

* Create new data exports and add query criteria to refine your exports
* View previously generated data exports

### Create data export <a href="#create-data-export" id="create-data-export"></a>

You can export both impression or event data by following below steps:

1. From the navigation bar, click **Data hub** and select the **Data export** tab.
2. Click the **Create** button for the data type you want to export.
3. Select a time range to query across.
4. Optionally click **Add query criteria** to filter the data you are exporting across a variety of dimensions.
5. Click the **Create** button.

Impressions and events are available for the last 90 days. The data export result is empty if the feature flag version is not within the selected time range.

{% hint style="info" %}
Depending on the query criteria and the number of events or impressions, exports can take up to 5 minutes to generate.
{% endhint %}

### Accessing data exports <a href="#accessing-data-exports" id="accessing-data-exports"></a>

Once you create your data export, you can access your generated exports to download via a CSV. In this table, you can delete any generated exports.

### Data export limits <a href="#data-export-limits" id="data-export-limits"></a>

The following limits apply to data exports:

* 5 data export requests per account per day, which can be increased
* 10,000,000 records per impression data export
* 1,000,000 records per event data export
* 7 day data export retention window. Reports are available for 7 days before they are deleted and no longer accessible

Contact [support@split.io](emailto:support@split.io) if you have any questions or concerns regarding your data export limits.

### Troubleshooting <a href="#troubleshooting" id="troubleshooting"></a>

#### Opening downloaded `.gz` Files on MacOS <a href="#opening-downloaded-gz-files-on-macos" id="opening-downloaded-gz-files-on-macos"></a>

If you are using MacOS and have downloaded a data export file from the Data hub that ends with `.gz`, you might find that the file does not open by default in your user interface. This is because some MacOS versions do not support `.gz` files natively.

To successfully open and extract these files, the recommended approach is to install the **The Unarchiver** app from the Mac App Store, which reliably handles `.gz` files.

Alternatively, you can use the Terminal application on your Mac to unzip the file manually with the following command:

```bash
gzip -d [file name]
```

#### Events not appearing in Harness FME <a href="#events-not-appearing-in-harness-fme" id="events-not-appearing-in-harness-fme"></a>

When [posting events](https://docs.split.io/reference/events-overview) using the Admin REST API, you may receive a successful `202` response code but not see the event in Split’s Data Hub. This can occur if any of the following issues are present:

* The Epoch timestamp is invalid. Ensure the timestamp is in milliseconds, not seconds.
* The traffic type is incorrect. Confirm that the traffic type used in the call exists in your Split organization.
* The API key is incorrect. Use the SDK API Key when posting events, not the Admin API Key.