Overview

De-identification masks sensitive data and replaces it with secure PII alternatives. Configuration runs as a fixed sequence — name, method and data source, rule definition, filter, output, summary — that is shared by all three configuration methods. This page documents the shared sequence and the reference tables common to every method: generators, modes, and output options.

The three methods differ only in how rules are created. Each has its own reference page: Auto, Excel, and Manual De-identification.

When to use it

• Understanding the end-to-end configuration sequence before choosing a method.
• Looking up the generators, modes, or output options shared across methods.

Configuration sequence

Step	Screen	Purpose
1	Configuration (start)	Select the project and enter a title.
2	Method + data source	Choose Auto, Excel-based, or Manual, and select the data source.
3	Rule definition	Create de-identification rules (method-specific).
4	Filter	Optionally restrict processing to a subset.
5	Output configuration	Choose output store and table-name prefix/suffix.
6	Summary	Review and submit.

Configuration methods

Method	Description
Auto	Uses AI to detect PII and de-identify it.
Excel-based	Applies rules from a downloaded, filled-in template uploaded back to Maskbits.
Manual	Defines each field rule directly in the UI.

Rule settings

Choosing the De-identification option on a data type enables three settings, used by all methods:

Setting	Description
Generator	The kind of value produced in place of the original (for example First Name).
Mode	How replacement values are generated across the data (for example Consistent).
Field Path	The path to the field being de-identified.

Generators

Named in the source; partial list.

Generator	De-identifies
First Name	First name values.
Last Name	Last name values.
Date / Date of Birth	Date and date-of-birth values.
Email	Email addresses.
Address	Address values.
Phone	Phone numbers.
SSN	Social security numbers.
State	State values.
City	City values.
Zip	Zip / postal codes.

De-identification modes

One mode named in the source.

Mode	Behavior
Consistent	The same input value maps to the same de-identified output across the data.
Random	Randomize the data

Output configuration

Option	Description
Output store	Where de-identified data is written (for example Mongo).
Prefix / Suffix	Text added to the output table name.
Output preview	Shows the resulting table name.

Key behaviors

All methods converge on the same flow. Auto, Excel, and Manual differ only at the rule-definition step; filter, output, and summary are identical.

Output is a separate table. De-identified data is written under a name derived from the source table plus a prefix or suffix, leaving the original in place.