← All DORA Capabilities

Database Change Management

Engineering Excellence
DIRECT DRIVER

Databases underpin critical business data and system behaviour, making changes inherently high risk. Unlike application code, database modifications can affect large volumes of data, introduce irreversible errors, or cause downtime if executed improperly. Poor database change management often becomes a major bottleneck to delivery, forcing manual coordination, lengthy testing cycles, and reluctance to evolve schemas.

Effective database change management integrates schema evolution into normal delivery practices, enabling safe, repeatable, and reversible changes. Mature organisations treat database changes as versioned, testable artefacts deployed through automated pipelines alongside application code. At the highest level, databases evolve continuously without disrupting service, supporting rapid innovation while protecting data integrity and availability.

Manual and Risk-Prone Changes
(Ad hoc updates to live systems)

Description

Database modifications are performed manually, often directly against production environments, with limited safeguards.


Observable Characteristics

  • Changes executed through manual scripts or interactive tools
  • Little or no version control for schema changes
  • Documentation incomplete or absent
  • High dependence on specialist knowledge
  • Testing limited or performed informally
  • Rollback procedures unclear or unavailable

Outcomes & Risks

  • High likelihood of data corruption or downtime
  • Significant operational risk
  • Bottlenecks around database specialists
  • Limited ability to evolve systems
Scripted but Coordinated Changes
(Controlled execution, limited automation)

Description

Changes are documented and executed through scripts, often coordinated as part of scheduled releases.


Observable Characteristics

  • Change scripts maintained for deployments
  • Formal approval required before execution
  • Testing performed in staging environments
  • Deployment windows scheduled to minimise impact
  • Rollback scripts sometimes prepared
  • Processes vary across systems

Outcomes & Risks

  • Improved stability but reduced agility
  • Continued dependency on specialist roles
  • Large batches of database changes
  • Difficult alignment with continuous delivery
Versioned and Repeatable Database Changes
(Schema evolution integrated with development)

Description

Database changes are managed as version-controlled artefacts deployed through consistent processes alongside application updates.


Observable Characteristics

  • Schema changes tracked in version control
  • Migration scripts applied automatically
  • Changes tested in multiple environments
  • Deployment processes standardised
  • Rollback strategies defined
  • Developers involved in database evolution

Outcomes & Risks

  • Better alignment with modern delivery practices
  • Reduced operational risk
  • Improved collaboration between teams
  • Requires disciplined testing and design
Automated and Safe Schema Evolution
(Changes deployed with safeguards and monitoring)

Description

Database changes are executed through automated pipelines with validation, compatibility strategies, and observability.


Observable Characteristics

  • Automated deployment of migrations
  • Backward-compatible change patterns used
  • Data integrity checks and validation steps
  • Monitoring during and after deployment
  • Minimal downtime techniques employed
  • Coordination with application changes

Outcomes & Risks

  • Predictable database evolution
  • Improved system reliability
  • Reduced need for maintenance windows
  • Increased complexity of automation
Continuous Database Delivery
(Database evolves seamlessly with the system)

Description

Database changes are routine, low-risk operations integrated into continuous delivery, enabling rapid innovation without compromising data integrity.


Observable Characteristics

  • Incremental, backward-compatible schema evolution
  • Near-zero-downtime deployment strategies
  • Automated rollback or recovery mechanisms
  • Data migrations handled transparently
  • Strong alignment between application and database evolution
  • Minimal need for specialised coordination

Outcomes & Risks

  • Ability to evolve data models quickly
  • Reduced technical constraints on innovation
  • Lower total cost of ownership
  • Competitive advantage through adaptability
Manage database changes safely, version-controlled, and aligned with application delivery practices.