A Salesforce Sandbox is a separate copy of your Salesforce organisation used for development, testing, and training — isolated from the production environment so that configuration changes, new automations, and integrations can be tested without any risk of affecting live data or disrupting your sales team’s daily operations. Understanding sandboxes is essential for every Salesforce administrator: making configuration changes directly in production without testing is the primary cause of Salesforce data corruption, broken automation, and disrupted workflows. This guide covers every sandbox type, when to use each, how to refresh them, and the Salesforce change management workflow that uses sandboxes correctly.
The best guide is the one that makes testing feel like a normal part of admin work.
A practical explanation should help readers see where sandbox usage fits into the change process.
That means the guide should focus on safety, experimentation, and controlled rollout.
For many teams, the real value is being able to validate changes before production is touched.
It should also show how different sandbox types support different kinds of work.
A good guide should explain what the sandbox is for and why testing in a separate environment matters.
That makes it an important part of the Salesforce admin workflow.
Salesforce Sandbox is useful because it gives teams a safe place to test changes before those changes affect real users or real data. It is one of the most practical tools for avoiding mistakes during configuration and development work.
The best guide is the one that makes testing feel like a normal part of admin work.
A practical explanation should help readers see where sandbox usage fits into the change process.
That means the guide should focus on safety, experimentation, and controlled rollout.
For many teams, the real value is being able to validate changes before production is touched.
It should also show how different sandbox types support different kinds of work.
A good guide should explain what the sandbox is for and why testing in a separate environment matters.
That makes it an important part of the Salesforce admin workflow.
Salesforce Sandbox is useful because it gives teams a safe place to test changes before those changes affect real users or real data. It is one of the most practical tools for avoiding mistakes during configuration and development work.
The best guide is the one that makes testing feel like a normal part of admin work.
A practical explanation should help readers see where sandbox usage fits into the change process.
That means the guide should focus on safety, experimentation, and controlled rollout.
For many teams, the real value is being able to validate changes before production is touched.
It should also show how different sandbox types support different kinds of work.
A good guide should explain what the sandbox is for and why testing in a separate environment matters.
That makes it an important part of the Salesforce admin workflow.
Salesforce Sandbox is useful because it gives teams a safe place to test changes before those changes affect real users or real data. It is one of the most practical tools for avoiding mistakes during configuration and development work.
The best guide is the one that makes testing feel like a normal part of admin work.
A practical explanation should help readers see where sandbox usage fits into the change process.
That means the guide should focus on safety, experimentation, and controlled rollout.
For many teams, the real value is being able to validate changes before production is touched.
It should also show how different sandbox types support different kinds of work.
A good guide should explain what the sandbox is for and why testing in a separate environment matters.
That makes it an important part of the Salesforce admin workflow.
Salesforce Sandbox is useful because it gives teams a safe place to test changes before those changes affect real users or real data. It is one of the most practical tools for avoiding mistakes during configuration and development work.
The best guide is the one that makes testing feel like a normal part of admin work.
A practical explanation should help readers see where sandbox usage fits into the change process.
That means the guide should focus on safety, experimentation, and controlled rollout.
For many teams, the real value is being able to validate changes before production is touched.
It should also show how different sandbox types support different kinds of work.
A good guide should explain what the sandbox is for and why testing in a separate environment matters.
That makes it an important part of the Salesforce admin workflow.
Salesforce Sandbox is useful because it gives teams a safe place to test changes before those changes affect real users or real data. It is one of the most practical tools for avoiding mistakes during configuration and development work.
Why Salesforce Sandboxes Exist
Salesforce is a live, multi-user production system where your sales team depends on data integrity and automation reliability every day. Without sandboxes, every new Flow, validation rule, page layout change, or integration configuration would need to be built and tested in production — where a mistake can corrupt live opportunity records, fire automation emails to real customers, or break the pipeline your management is reporting from. Sandboxes eliminate this risk by providing an isolated environment where changes can be built, tested, and validated before being promoted to production via the Change Sets or Salesforce DX deployment process.
Salesforce Sandbox Types
Salesforce offers four sandbox types with different characteristics — the right choice depends on what you need to test and what your Salesforce edition includes:
1. Developer Sandbox
- Data included: Configuration and metadata only — no production data records. Contains a copy of your production org’s structure (objects, fields, page layouts, automation) but empty of actual Account, Contact, Opportunity, and other records
- Storage: 200 MB data storage, 200 MB file storage
- Refresh interval: Can be refreshed once per day
- Number included: Enterprise edition includes 25 Developer sandboxes; Unlimited edition includes more
- Best for: Individual developer or admin work — building and unit testing a new Flow, creating custom fields, developing Apex code. The most commonly used sandbox type for day-to-day admin configuration work
- Limitation: No production data means you cannot test automations against realistic data volumes or test data-dependent logic against real Account hierarchies
2. Developer Pro Sandbox
- Data included: Configuration and metadata only — no production data
- Storage: 1 GB data storage, 1 GB file storage (5× more than Developer)
- Refresh interval: Once per day
- Number included: Enterprise edition includes 5 Developer Pro sandboxes
- Best for: Testing that requires importing larger data sets for realistic testing — testing bulk data processing jobs, testing integrations that depend on large data volumes, or multi-user test scenarios where the standard Developer sandbox storage is insufficient
3. Partial Copy Sandbox
- Data included: Configuration metadata plus a sample of production data — defined by a Sandbox Template that specifies which objects and what percentage of records to copy (maximum 10,000 records per object)
- Storage: 5 GB data storage
- Refresh interval: Once every 5 days
- Number included: Enterprise edition includes 1 Partial Copy sandbox; Unlimited includes more
- Best for: UAT (User Acceptance Testing) where end users need to work with realistic CRM data — testing new reports against real account data, validating automation behaviour on representative records, training sessions with familiar-looking data
- Important consideration: Production data is copied into the sandbox — implement data masking if personally identifiable information (PII) or sensitive customer data must not exist in the sandbox environment for compliance reasons
4. Full Sandbox
- Data included: Complete copy of all production data — every record, file, and attachment in production is replicated to the Full Sandbox
- Storage: Matches production storage allocation
- Refresh interval: Once every 29 days
- Number included: Unlimited edition includes 1 Full Sandbox; Enterprise does not include a Full Sandbox by default (must be purchased as an add-on)
- Best for: Performance testing (validating that new automations perform at production data volumes), load testing integrations, full regression testing before major releases, and disaster recovery testing
- Limitation: Creation and refresh take many hours for large orgs — a Full Sandbox of a 500K record org may take 24–48 hours to complete. The 29-day minimum refresh interval means the Full Sandbox data becomes stale quickly for active orgs
- Cost: Full Sandboxes are expensive add-ons for Enterprise edition — a primary reason most Enterprise customers use Partial Copy for UAT rather than Full Sandbox
How to Create and Refresh a Sandbox
Creating a New Sandbox
- Navigate to Setup → Environments → Sandboxes → New Sandbox
- Enter a name (e.g., “Dev1”, “UAT”, “Staging”) — the sandbox URL will be
https://[orgname]--[sandboxname].sandbox.my.salesforce.com - Select the sandbox type
- For Partial Copy, select or create a Sandbox Template defining which objects to copy
- Click Create — sandbox creation takes 10 minutes to several hours depending on type and org size
- You receive an email when the sandbox is ready — log in using your production username appended with
.[sandboxname](e.g.,admin@yourcompany.com.uat)
Refreshing a Sandbox
Refreshing a sandbox resets it to a fresh copy of production — overwriting all changes made in the sandbox since its last creation or refresh. This is how you get an updated copy of production configuration and data into your sandbox environment. Navigate to Setup → Environments → Sandboxes → click Refresh next to the sandbox name. Refresh is subject to the minimum interval for each sandbox type.
Important: Refreshing a sandbox destroys all work done in that sandbox since the last refresh — any automation, fields, or code built in the sandbox that has not been promoted to production will be lost. Always deploy sandbox changes to production before refreshing.
The Change Management Workflow Using Sandboxes
The recommended Salesforce change management process for admin-level changes (Flow, fields, page layouts, validation rules):
- Build in Developer Sandbox: Make all configuration changes in a Developer sandbox — never in production. Test thoroughly in the sandbox
- Deploy to staging (Partial Copy or second Developer Sandbox): Promote the changes from the Dev sandbox to a staging sandbox for UAT if the change is complex or user-facing
- User Acceptance Testing: End users (sales reps, managers) test the changes in the staging environment against realistic data
- Deploy to production via Change Set: Create an Outbound Change Set in the sandbox containing all modified components — deploy it to production via Setup → Environments → Change Sets → Inbound Change Sets in the production org
- Post-deployment verification: Verify the components deployed correctly in production and test the key user workflows
Sandbox Best Practices
- Never make configuration changes directly in production: Even “small” changes like adding a picklist value or modifying a validation rule should be built and tested in a sandbox first
- Label sandbox email alerts: In sandbox Setup → Email Deliverability, set “Access Level” to “System email only” — preventing sandbox Flows from sending real emails to real customers. Or configure email relay to redirect all sandbox emails to an internal test inbox
- Deactivate production integrations in sandbox: After a refresh, production API credentials and integration endpoints are copied to the sandbox — immediately deactivate or reconfigure these to prevent sandbox testing from writing data to production external systems
- Name sandboxes by purpose: “dev1”, “dev2”, “uat”, “staging” — clear naming conventions prevent accidental production deployments from the wrong environment
- Use Salesforce DX for developer workflows: For teams with Salesforce developers, Salesforce DX (Developer Experience) combined with scratch orgs and a CI/CD pipeline provides more granular version control than Change Sets — each metadata change tracked in Git and deployed via automated pipeline
Sandbox vs Production: Key Differences to Know
- Sandbox login URL ends in
.sandbox.my.salesforce.com; production ends in.my.salesforce.com - Sandbox username suffix: production username +
.[sandboxname] - Sandbox records have different IDs than production records — IDs are not portable between environments
- Connected app credentials and OAuth tokens are not automatically valid between sandbox and production — integration authorisation must be re-done after sandbox refresh
- Email deliverability is restricted in sandboxes by default — configure carefully before testing email automation
Conclusion
Salesforce Sandboxes are the foundation of responsible Salesforce administration — they are the mechanism by which administrators can build, test, and validate changes without risking the production system that sales teams depend on daily. Developer sandboxes for configuration work, Partial Copy sandboxes for UAT with realistic data, and a Clear Set deployment workflow for promoting changes to production: this is the standard change management pattern that separates well-governed Salesforce orgs from chaotic ones where production is the testing environment. Every Salesforce administrator should be comfortable creating, refreshing, and managing sandboxes before making non-trivial configuration changes — it is a foundational competency covered in the Salesforce Administrator certification and essential for any organisation treating Salesforce as a critical business system.
Sources
Salesforce, Sandbox Types and Limits Documentation (2026)
Salesforce, Change Set Development Documentation (2026)
Salesforce, Trailhead — Salesforce DX Trail (2026)
Salesforce, Sandbox Best Practices Guide (2026)
Salesforce Sandbox Strategy: Getting the Most From Your Testing Environment
Problem: Sandbox Refresh Wipes Important Test Data and Configuration
Salesforce Sandbox environments are refreshed from the Production org on a scheduled or manual basis — but when you refresh a Sandbox, all data, custom configuration, and changes made in that Sandbox since the last refresh are permanently deleted and replaced with a new copy of the Production org. Many teams lose hours of test configuration because they forgot a Sandbox refresh was scheduled. To protect against accidental refresh loss: (1) Never use a Sandbox as your only copy of in-progress development work — commit all Apex code and metadata to a version control system (Git) so it can be re-deployed to any Sandbox after a refresh. (2) Document all in-progress Sandbox configuration (custom fields, page layout changes, automation rules) in a Change Log before refreshing. (3) Use Salesforce DX (Scratch Orgs) for development work that needs to survive across Sandbox refreshes — Scratch Orgs are created from source code rather than Production refreshes and are not affected by Production Sandbox refresh cycles.
Problem: Full Sandbox Takes Too Long to Refresh and Delays Development Cycles
Salesforce Full Sandboxes (which copy all Production data including records) can take 24-72 hours or more to complete a refresh for large orgs. This delay slows development cycles, especially when multiple teams are waiting for an updated Sandbox to test against realistic Production data. To manage Full Sandbox refresh delays: (1) Schedule Full Sandbox refreshes quarterly rather than on-demand — stagger them with your major release cycles so teams know when a fresh copy will be available. (2) Supplement Full Sandbox with Partial Sandbox (which copies a configurable subset of data in 1-5 hours) for teams that need fresh data but can work with a representative sample rather than the full dataset. (3) Use Developer Sandboxes (which copy configuration but no data, and refresh in under 30 minutes) for unit testing and automation development — reserve Full Sandbox for UAT (User Acceptance Testing) and performance testing that genuinely requires Production-scale data volumes.
Problem: Sandbox Metadata Deployment to Production Fails Due to Apex Test Failures
When deploying changes from Sandbox to Production using Change Sets or Salesforce DX, Salesforce runs all Apex tests in the Production org as part of the deployment process. If any test fails, the entire deployment is blocked. This is particularly disruptive when existing Production tests break due to new configuration changes (like adding a required field validation rule that causes existing test data to fail). To manage deployment test failures: (1) Before deploying from Sandbox, run all Production Apex tests in your Sandbox environment using the Apex Test Execution page (Setup > Apex Test Execution > Run All) to identify test failures proactively. (2) Review the failing test classes to understand whether they need updated test data (to accommodate new required fields or validation rules) or genuine bug fixes. (3) For urgent deployments blocked by unrelated test failures, Salesforce allows specifying individual test classes to run during deployment rather than all tests — use this option to deploy critical changes while fixing failing tests separately.
The best sandbox setup is the one that keeps testing separate from production. If the test environment is not respected, the safety benefit disappears.
The best sandbox setup is the one that keeps testing separate from production. If the test environment is not respected, the safety benefit disappears.
The best sandbox setup is the one that keeps testing separate from production. If the test environment is not respected, the safety benefit disappears.
The best sandbox setup is the one that keeps testing separate from production. If the test environment is not respected, the safety benefit disappears.
The best sandbox setup is the one that keeps testing separate from production. If the test environment is not respected, the safety benefit disappears.
Frequently Asked Questions
What are the different types of Salesforce Sandboxes?
Salesforce offers four sandbox types: (1) Developer Sandbox – copies org configuration and metadata only (no records), refreshes in under 30 minutes, 200MB data storage. Best for Apex development and unit testing. (2) Developer Pro Sandbox – copies configuration only, 1GB storage. Best for teams needing more storage for development work. (3) Partial Copy Sandbox – copies configuration plus a configurable subset of Production data (selected object types, up to 5GB). Refreshes in 1-4 hours. Best for integration testing with realistic data. (4) Full Copy Sandbox – complete copy of all Production data and configuration. Storage matches Production. Refresh time is 24-72+ hours for large orgs. Best for performance testing and final UAT. Different Salesforce editions include different numbers of each sandbox type — Enterprise includes multiple Developer Sandboxes and one Partial Copy sandbox, while Unlimited adds Full Copy Sandboxes. Additional Sandboxes can be purchased as add-ons.
How often should you refresh a Salesforce Sandbox?
Recommended refresh frequency varies by sandbox type and use case. Developer Sandboxes should be refreshed on-demand when developers need to work against current Production configuration — typically at the start of each major development sprint (every 2-4 weeks). Partial Copy Sandboxes used for integration testing should be refreshed quarterly or after significant Production data model changes that affect the test data. Full Copy Sandboxes used for UAT should be refreshed at the start of each major release cycle (typically 1-4 times per year). Avoid over-refreshing Sandboxes that teams are actively using for testing — each refresh wipes all in-progress work. Establish a Sandbox refresh calendar with release managers and team leads to coordinate refresh timing across the development team.
Can you do a Sandbox-to-Sandbox deployment in Salesforce?
Salesforce Change Sets, the primary deployment tool, only support Sandbox-to-Production or Sandbox-to-Sandbox deployments using specific routes — typically from a child Sandbox to a parent Sandbox or to Production within the same Sandbox lineage. Deploying from one independently-refreshed Sandbox to another requires Salesforce DX (the Salesforce developer experience toolchain) or a CI/CD pipeline. Salesforce DX allows source-driven deployments using the Salesforce CLI that can deploy metadata from any source to any target org. For teams with multiple Sandboxes at different development stages (Development Sandbox > UAT Sandbox > Production), Salesforce DX with a Git repository is the recommended approach for controlled multi-stage deployments. Change Sets work adequately for simpler single-stage deployments but become unmanageable for complex multi-environment release pipelines.
How do you set up and manage Salesforce Sandbox permissions?
Salesforce Sandbox environments inherit user access from the Production org at the time of the last refresh — all Production users are copied to the Sandbox with the same profiles and permission sets. After refresh, Sandbox administrators typically: (1) Reset Sandbox user passwords to prevent accidental confusion with Production credentials. (2) Disable sandbox users who should not have access to the Sandbox (contractors, departed employees whose Production records were not yet deactivated). (3) Enable ‘Is Sandbox’ checkbox verification in your organization’s security guidelines — Salesforce adds a visual banner to Sandbox environments to distinguish them from Production, but additional org-level customization (like a custom Sandbox warning banner using Custom Metadata or a Theme) helps prevent accidental data changes in the wrong environment. Sandbox data is isolated from Production — changes in a Sandbox do not affect Production data, which is the core safety benefit.