An S3 snapshot exposure source lets you upload exposure data in batch by delivering gzip-compressed CSV files to an MX8 Labs-managed S3 bucket. This is the right choice when you collect exposure data in your own systems and want to send it server-side - especially if you need to hash IP addresses before they leave your environment.
Before you begin
You'll need the following information ready:
- A source name for this exposure source.
- A retention period (in days) for how long records should be kept.
- Your Hash Salt (if you're salting hashed IPs).
- The column names in your CSV files that correspond to MX8 Labs' required fields.
- Your IAM Role ARN if you want MX8 Labs to grant your AWS role upload permissions (this maps to the External ARN field).
Step 1 - Open the creation dialog
Navigate to your exposure sources and click Create Exposure Source. Select the S3 snapshot card.

Step 2 - Configure basic settings
Fill in the top-level fields:
- Source Name - A descriptive label for this source. This cannot be changed after creation.
- Retention Days - The minimum number of full days to retain S3 snapshots and upload versions. Enter at least 1 day to cover your campaign's lookback window. The expiry clock starts from server ingestion time and uses UTC daily boundaries.
- Enabled - On by default. Toggle off to configure without activating.
Step 3 - Add dimensions
Dimensions work the same way as they do for pixel sources - they're additional attributes attached to each exposure record, used as aggregation buckets, and every source needs at least one. Type a dimension name into the Add Dimension field and press Enter or click + Add Dimension.
Step 4 - Configure field mappings
This is where you tell MX8 Labs how your CSV columns map to the fields it needs. Your column names don't need to match MX8 Labs' - you just need to specify which of your columns corresponds to each field.
Required fields:
- Hash Salt - The salt value used when hashing IP addresses. If you don't use a salt, leave this empty.
- UID Field Mapping - The column name in your CSV that contains the user identifier (e.g.,
user_id).
Optional fields:
- Exposed IP Field Mapping - The column containing plaintext IP addresses. Provide this or Hashed IP Field Mapping.
- Hashed IP Field Mapping - The column containing hashed IP addresses. Provide this or Exposed IP Field Mapping.
- Brand Field Mapping - The column containing the brand name associated with each exposure. You must provide a brand mapping or at least one dimension mapping - the form won't save with neither.
- External ARN - Your IAM Role ARN. MX8 Labs uses this to grant your AWS role permission to upload files into your dedicated S3 prefix.

Step 5 - Configure dimension mappings
If you added dimensions in Step 3, scroll to the Dimension Mapping section and click the + button to map each dimension to the corresponding column in your CSV.
Step 6 - Create the source
Click Create to save. The source is now ready to receive files.
Note: The hash algorithm is currently fixed to MD5. Other algorithms are available on request - contact your MX8 Labs account team if you need a different algorithm.
What the retention setting controls
Each snapshot expires at the next UTC midnight after ingestion plus the configured number of days. This guarantees the full configured days; it does not use the exposure timestamps inside the file. A newer successful upload replaces the previous snapshot. Failed, empty, or duplicate attempts do not extend it.
Expired snapshots stop matching before physical cleanup finishes. Operational ingest history remains available separately. See History and raw-data retention for cleanup timing and retention details.
What's next
- Server-Side File Delivery & Format Requirements - details on file format, naming, and upload process.
- Monitoring S3 Exposure Ingests - follow each uploaded file and troubleshoot rejected deliveries.
- IP Hashing Rules for Exposure Data - input formatting, salt handling, and output format guidance.
