In 2018 Revlon could not ship orders it had already taken, and the cost came to $64M. A business that cannot deliver what it has sold loses twice: the revenue now and the customers later.
What happened
Orders piled up that the company could not ship, at a cost of $64M. As widely reported, the trouble followed a major change to the systems that run operations.
When the systems that plan, make and ship products change all at once, small mismatches show up as stopped production lines and waiting customers.
Where an independent check was missing
Big system changes are usually tested in pieces, and each piece passes. What often goes untested is the whole flow, from order to warehouse to delivery, under real volume.
An independent readiness review that asks what happens if the new system is slower, wrong or down on day one, and checks that there is a plan for each, is the safeguard that is easy to skip.
What to do differently
Move over in stages where you can, and run old and new side by side for the most important flows until the new one has proved itself.
Decide before the switch what would trigger a rollback, and keep the old path available until that point has passed.
Protect the customer first
Whatever else happens in a migration, the customer should still get the order. Identify the handful of flows that must never stop and give them the most testing and the closest watching.
Plan extra support for the first weeks. The problems that matter appear only when real orders arrive.
- Orders that could not ship cost Revlon $64M in 2018.
- Test the whole order-to-delivery flow, not only the parts.
- Run old and new side by side where possible.
- Agree rollback triggers before go-live.
