Blog Details

Every ERP programme has a date circled on a calendar. It is the least interesting part of the project. What decides whether the first month after that date is quiet or ugly is the work that happened weeks earlier — the design decisions, the data, the testing and the plan for the day the old system stops.

Design against the process, not the demo

A fit-gap workshop that only maps happy paths produces a system that works in the demo and fails on the third exception. The processes worth the most design time are the ones the business runs under pressure: month-end, returns, credit holds, the order that has to ship today.

Data migration is a project, not a task

Master data is where go-lives slip. Duplicate customers, part numbers with three formats and opening balances that never quite tie are all discoverable months ahead — and all painful to discover in the cutover window. Migrate early, migrate repeatedly, and reconcile every run against the source.

Test with the people who will use it

User acceptance testing run by the project team proves the configuration. UAT run by the people who will live in the system proves the process. Only the second one tells you anything about the first week of trading.

Plan the fallback before the cutover

A cutover plan needs a stated point of no return, a rollback that has been rehearsed, and a hypercare period with named people on it. Stabilisation is part of delivery — not an afterthought once the invoice is raised.