How we redact company data before it is licensed

The first question every company asks us is some version of: “What happens to our people’s names, our customers and our secrets?” The short answer is that they never leave processing. Here is how that works.

Read-only, scoped access

Everything starts with read-only connections to the sources you choose. We cannot change, send or delete anything in your systems, and you can revoke access at any time. You also set the scope: specific channels, repositories, folders, date ranges or customers can be excluded before processing begins.

Removing identifiers

Each record passes through a redaction pipeline that detects and removes identifying and sensitive details, including:

• →Names of employees, customers and contacts
• →Email addresses, phone numbers and physical addresses
• →Account, invoice and payment numbers
• →Credentials, API keys, tokens and secrets in code or config
• →Company-specific confidential terms you flag, such as product codenames

Keeping the structure intact

Deleting identifiers outright would break the data. A thread where every name is blank no longer shows who handed off to whom. So instead of blanking, we replace each identifier with a consistent placeholder. The same person becomes the same token everywhere they appear, which preserves the shape of the work without revealing who did it.

The terminal animation on our home page shows a simplified version of this: a raw record, the same record with identifiers replaced, and the priced asset that results.

Quality checks and review

After redaction, assets go through quality checks for leftover identifiers, broken links between records and low-value noise. You review what each listing covers before it is offered to buyers. Raw identifiers are never exported.

Questions about security

If your security or legal team wants more detail on our handling of data, contact security@datalogix.ai. We are happy to walk through it.