# Masked, seeded and synthetic test data strategy

*Platform Development Lifecycle and Deployment Architect Study Guide · Testing · Last updated October 26, 2026 · https://www.certifyforce.com/study-guide/platform-development-lifecycle-and-deployment-architect/testing/masked-seeded-and-synthetic-test-data-strategy*

How to give every environment representative test data without exposing regulated data: classify fields, mask production copies with Data Mask & Seed, seed related subsets with Seeding by Own from Salesforce, and use synthetic or curated datasets where production data is never needed.

- Certification: Salesforce Platform Development Lifecycle and Deployment Architect
- Exam section: Testing
- Study guide: https://www.certifyforce.com/study-guide/platform-development-lifecycle-and-deployment-architect

**Official exam objective:**

> Given a customer scenario, describe and recommend a unified test data strategy that utilizes representative data in a secure manner throughout the development lifecycle.

## What the exam expects

- Recommend one test data standard across scratch orgs, Developer sandboxes, Partial Copy and Full sandboxes that keeps regulated data inside production.
- Configure masking in a current org with Data Mask & Seed: policies, locale, value distribution, scheduling, and migration from the legacy managed package.
- Use data classification metadata to drive what gets masked.
- Choose a Seeding by Own from Salesforce template, source and method, and handle automation that fires during a seed.
- Decide which test layers need baseline data and whether a dataset is small (tree commands) or large (bulk commands).

Key terms: **masking** (replacing sensitive values in a sandbox with realistic fake values), **seeding** (copying a related subset of records from a source org into a destination), **synthetic data** (generated records never derived from production), **data classification** (field-level sensitivity and compliance metadata), **masking policy** (the Data Mask & Seed definition of which fields to mask and how).

## Key facts

| Fact | Detail | Source |
|---|---|---|
| Legacy Data Mask package | Support ended 2026-07-09; read-only from 2026-12-31 | [Retirement](https://help.salesforce.com/s/articleView?id=005299299&language=en_US&type=1) |
| Data Mask & Seed | Core app, no installation, available to all customers since Summer '26 (the Spring '26 release note says Beta) | [Retirement](https://help.salesforce.com/s/articleView?id=005299299&language=en_US&type=1) |
| Migration | New policies replace configurations; custom libraries shared | [Retirement](https://help.salesforce.com/s/articleView?id=005299299&language=en_US&type=1) |
| Distribution preserved | Masking jobs under 20M records | [Retirement](https://help.salesforce.com/s/articleView?id=005299299&language=en_US&type=1) |
| Masking jobs | Run manually or scheduled | [Data Mask and Seed release note](https://help.salesforce.com/s/articleView?id=release-notes.rn_data_mask_and_seed.htm&language=en_US&release=260&type=5) |
| Classification fields | Owner, usage, sensitivity, compliance categorization | [Classify Sensitive Data](https://help.salesforce.com/s/articleView?id=platform.data_classification_intro.htm&language=en_US&type=5) |
| Seeding source | Production org or a specific sandbox | [Seeding](https://help.salesforce.com/s/articleView?id=platform.seed_o_overview.htm&language=en_US&type=5) |
| Generate template | No source; 100 synthetic records per object | [Seeding Templates and Jobs](https://trailhead.salesforce.com/content/learn/modules/seeding-basics/build-seeding-templates-and-run-jobs) |
| Tree vs bulk | Tree under about 3,000 records; bulk for millions | [Work with Data](https://developer.salesforce.com/docs/atlas.en-us.sfdx_dev.meta/sfdx_dev/sfdx_dev_test_data.htm) |
| Sandbox templates | Partial Copy and Full only | [Sandbox Types and Templates](https://help.salesforce.com/s/articleView?language=en_US&id=platform.create_test_instance.htm&type=5) |

## Check yourself

- Why classify fields before building masking policies?
- When should a seeding job disable Salesforce automations?
- Which tests in a scratch org usually need no baseline data?
- Why can a Generate template feed an environment that must never hold production data?

## Sources

- [Data Mask Managed Package Retirement](https://help.salesforce.com/s/articleView?id=005299299&language=en_US&type=1)
- [The New Data Mask and Seed (Beta)](https://help.salesforce.com/s/articleView?id=release-notes.rn_data_mask_and_seed.htm&language=en_US&release=260&type=5)
- [Populate and Secure Data Sets with Seeding](https://help.salesforce.com/s/articleView?id=platform.seed_o_overview.htm&language=en_US&type=5)
- [Build Seeding Templates and Run Jobs](https://trailhead.salesforce.com/content/learn/modules/seeding-basics/build-seeding-templates-and-run-jobs)
- [Get Started with Seeding](https://trailhead.salesforce.com/content/learn/modules/seeding-basics/get-started-with-seeding)
- [Classify Sensitive Data to Support Data Management Policies](https://help.salesforce.com/s/articleView?id=platform.data_classification_intro.htm&language=en_US&type=5)
- [Trust](https://architect.salesforce.com/docs/architect/well-architected/guide/trust.html)
- [Work with Data](https://developer.salesforce.com/docs/atlas.en-us.sfdx_dev.meta/sfdx_dev/sfdx_dev_test_data.htm)
- [Sandbox Types and Templates](https://help.salesforce.com/s/articleView?language=en_US&id=platform.create_test_instance.htm&type=5)

## About the full unit

The "Core concepts", "Exam traps" and "Worked scenario" sections of this unit, its 13 linked practice questions, and unlimited timed mock exams are included with the Platform Development Lifecycle and Deployment Architect pack (USD $19): https://www.certifyforce.com/practice-exams