Skip to main content
Your audience data is yours. TrackPlay exports it in two shapes, and the difference between them matters: one is safe to upload straight to an ad platform, the other contains contact information and needs to stay inside your team.

The two formats

A two-column CSV: email, phone, both SHA-256 hashed. Nothing else.Every value is normalized the same way ad platforms expect: lowercase, trimmed, then hashed. That is deliberate. Meta, Google, and TikTok all hash a customer list the same way before matching it, so a hash TrackPlay produces from Jane@Acme.com matches the hash the ad platform produces from the same address on its side. Upload it directly as a Custom Audience or Customer Match list.
No raw email address and no raw phone number leave your app in this format. If a value did not already have a hash on file, TrackPlay hashes the raw contact at export time rather than including it plain, so the file that downloads is always hash-only.
Rows with neither an email hash nor a phone hash, and exact duplicate hash pairs, are dropped before the file downloads. There is no reason to upload noise to an ad platform.

Two independent choices

Every export is one request with two parameters, chosen independently. So scope=hot&format=full is your call list with real phone numbers and emails attached. scope=all&format=customer_match is every viewer you have ever identified, hashed and ready to upload as a retargeting audience.

How the export runs

This is a direct download, not a background job. The file streams to you the moment you request it. There is no queue, no email with a link, and no waiting.
That directness comes with hard limits, so one export request can never turn into an unbounded database scan: If your audience is larger than the cap, export in the scope that fits your need (hot is almost always what you want for follow-up) rather than expecting a second file to pick up where the first left off. There is no pagination on this endpoint today.
This is different from the video-report export in the API, which does run as an asynchronous job and hands back a download URL for larger jobs. Audience exports and video-report exports are separate mechanisms with separate limits. Do not assume the async behavior of one applies to the other.