Skip to content

Engineering glossary

Multi-Tenancy

Multi-tenancy is an application architecture in which one deployed system serves multiple customer organizations, called tenants, while enforcing defined boundaries around each tenant's data and behavior.

In plain language

Multiple organizations use the same product, but each should experience an appropriate private context. Adding a tenant identifier to a table is not sufficient isolation by itself; authorization, queries, storage, jobs, caches, logs, and administration must all respect tenant boundaries.

Why it matters to a business

  • Supports operating one product for many customer organizations
  • Enables shared deployment and product updates
  • Creates a consistent place for tenant-specific access, configuration, and billing

How it works

  1. 1A trusted tenant context is established from identity and membership.
  2. 2Data-access and service layers enforce tenant scope.
  3. 3Background work, files, caches, and integrations preserve the same boundary.
  4. 4Administrative access and cross-tenant operations receive explicit controls and auditability.

Common use cases

  • B2B SaaS applications
  • Customer and partner portals
  • Franchise or multi-organization operations platforms

Important implementation decisions

  • Shared, schema-separated, or database-separated storage
  • Tenant-aware identity and role scope
  • Per-tenant configuration, limits, and encryption needs
  • Migration, backup, export, and deletion boundaries

Common mistakes and misconceptions

  • Trusting a tenant ID supplied by the browser
  • Filtering only some queries by tenant
  • Ignoring caches, queues, files, analytics, or support tools
  • Assuming one storage model is universally safest

Limitations and trade-offs

More isolation can reduce blast radius but increase operational complexity and cost.

Shared infrastructure improves efficiency but demands disciplined enforcement and testing.

Need to apply this concept to a real system?

A focused technical discussion can identify the decisions and constraints that matter before implementation.