The key question

A list of convertible objects is not a roadmap. Delivery has to consider applications, business processes and operational requirements together.

What to inventory

  • Group findings by business process and accountable team.
  • Identify blocking dependencies and unresolved architecture decisions.
  • Capture test data, cutover windows and acceptance criteria.

Design the target deliberately

Start with a pilot that tests an important assumption. Then group related objects and callers into migration waves. Every wave needs data reconciliation, regression testing and a defined operational hand-over. Effort ranges should remain linked to their assumptions.

A concrete validation step

A useful migration unit

Keep the description, affected objects and callers, assumptions, target decision, test evidence, owner and acceptance criterion together. A single person-day estimate hides those dependencies.

Evidence required before sign-off

  • Define a testable completion criterion for each wave.
  • Confirm dependencies between teams and external systems.
  • Clarify rollback point and consequences once the target receives new writes.
  • Update remaining effort and sequence after the pilot.

Primary sources & further reading

These recommendations are engineering guidance. Specific options depend on source and target versions, privileges and operating model.