Operation and switching

Keep or switch? Ten checks before moving automation

Migration is not export, import and a tasteful announcement. Triggers, identities, data mappings, histories and failure behaviour often need rebuilding.

Compare the cost of staying with the complete cost and risk of moving.

EDITOR’S NOTEOperation and switching
KeepPilotSwitch
Implementation includes the part after it stops.

THE SHORT ANSWER

Switch only when a material constraint, cost or operating failure survives a pilot. Dissatisfaction is evidence; it is not yet a migration plan.

OPERATING CONDITIONS

Test the awkward path, not the demo

01

Inventory

List workflows, owners, credentials, schedules and dependencies.

02

Portability

Separate reusable logic from product-specific behaviour.

03

Pilot

Run critical flows in parallel and define rollback.

04

History

Decide what logs and audit evidence must remain accessible.

Price is one operating condition

Include rebuild, validation, training, dual-running, lost history and recovery risk.

What to test before committing

Pilot the most critical and least portable workflow first.

GOOD FIT

Teams with a named reason to move, an inventory and time for parallel testing.

SKIP FOR NOW

A migration driven by fashion, a new interface or an untested assumption about lower monthly price.

Before changing the operating model

Repair governance, remove unused workflows or renegotiate the current plan before moving.

A new platform can remove old problems. It usually brings its own stationery.