Skip to main content

Onboarding Customer Data

Data onboarding brings the customer data you hold offline, such as CRM records, loyalty members and purchase history, into Surfside so you can use it for targeting, measurement and personalization across channels. You send us a file of identifiers and attributes. We match each record to a Surfside ID in our identity graph, discard the identifiers, and make the resulting audience available in the platform.

How it works​

  1. You prepare a file. One row per customer, with your own customer ID, the identifiers you hold and any attributes you want to segment on. The contract is in Batch Upload File Format.
  2. You deliver it. Upload to your encrypted S3 location on a schedule you choose, or upload a one-off file through the Cleanroom UI. See Secure File Transfer.
  3. We match it. Each record is linked to a Surfside ID, an anonymous people-based identifier, through our identity graph and our partnership with LiveRamp.
  4. We activate it. Surfside IDs resolve to mobile devices and other online identifiers, so the audience can be targeted, measured and enriched across the channels you choose.

What to send​

Matching runs on three identifiers. Send every one you hold for each record, because a record with both an email and a phone matches more often than either alone.

  • Email address, hashed with SHA-256.
  • Phone number, hashed with SHA-256. US and Canada only.
  • Mobile device ID, IDFA or GAID, sent raw.

Every record also carries your client customer ID, a stable identifier of your own that never changes between files. It is the audience key. It de-duplicates records, ties each file to the ones before it, and is how you act on what comes back.

You can send plaintext email and phone and we hash them on arrival inside the onboarding environment, or you can hash before sending. If you hash, preparation has to match ours character for character or the record will not match. Hashing Identifiers gives the rules, working code and test vectors.

Privacy by design​

Your data is protected at every step, and plaintext identifiers never reach the Surfside platform.

  1. In transit and at rest. Files land in a private, encrypted S3 location behind a firewall. If you want a second layer, encrypt each file with our public PGP key before upload. Your Surfside representative can supply the key.
  2. Into the processing environment. Files are mirrored into the onboarding environment through a second firewall and securely deleted from S3 once processed.
  3. Anonymized, matched and discarded. Personal identifiers are hashed and matched against the hashed identifiers in the Surfside identity graph. Each matched record is linked to a Surfside ID, and the identifiers themselves are discarded. Only the Surfside ID, your customer ID and the attributes you sent remain.

Never include anything beyond what matching and segmentation need. Government ID numbers, financial account details and health information must not appear in any file. The full list is in Batch Upload File Format.

Where to go next​