Every major implementation has issues. The executive question is whether those issues are visible, owned and being resolved while they can still be changed economically.
1. Progress is measured by activity
Completed workshops and configured functions do not prove that the organisation is ready to plan, schedule and execute maintenance differently.
2. Future processes exist mainly in slides
Teams need clarity on decisions, roles, information, exceptions and hand-offs—not only a high-level diagram.
3. Master data is a late cleansing task
Asset hierarchy, job plans, coding and maintenance programmes embody operational choices. Loading technically valid data is not the same as creating usable information.
4. The implementer is the only source of assurance
Delivery partners should assure their work. Sponsors may still need an independent view representing operational interests.
5. Testing proves transactions
Realistic testing should demonstrate end-to-end working practices, data fitness, exceptions and decision outcomes.
6. Training begins after design
By then, role and capability gaps may already be embedded. Users need the maintenance purpose behind each transaction.
7. Integrations have no business owner
An interface decides which information can be trusted, when it is available and who owns it. This is not merely technical.
8. Go-live criteria focus on the system
Operational readiness also requires process clarity, fit data, capable roles, support and credible stabilisation plans.
What a useful review produces
A short review should distinguish material delivery threats from longer-term improvements, connect each finding to its operational consequence and provide a prioritised action sequence.