Seven warning signs that your S/4HANA go-live is at risk

Go-live risk · 6 min read

None of them appears on a status slide. All of them are visible eight weeks ahead, if you know where to look.

Flat reporting line while a dashed delivery line drifts away before go-live.

Programs do not go from green to red. They go from green to green to green to cancelled, because the reporting line and the delivery line are the same people. These are the points we look at when we are asked for an assessment eight weeks before a date.

1. Nobody can name the three critical activities. Ask the program lead — without an open tool — which three activities determine the date. If the answer takes longer than fifteen seconds, the plan is being maintained, not managed.

2. Test progress is reported as a percentage of executed test cases. Executed is not passed, and passed is not accepted by the business. Ask for open priority 1 and 2 defects per cycle, with a closure curve. A flat curve four weeks before the date is the strongest indicator we know.

3. The cutover has never been rehearsed in the real window. A rehearsal that ends with “we will manage it over a weekend” is an assumption, not a measurement. The relevant number is elapsed hours against the downtime window you actually have — with the real data volume.

4. Data cleansing sits with IT. The business owns the data quality problem, because the business created it. Cleansing delegated to IT or the partner is the most common late surprise in an S/4HANA program — and it shows in the last migration cycle, when there is no time left.

5. Key users are still doing their day jobs. Ask for the agreed percentage and have their line manager confirm it. The gap between the two numbers is your real testing and training capacity.

6. Change requests are still being accepted. Every request accepted in the last quarter consumes testing capacity — and that is already the binding constraint. A scope freeze is unpopular and the cheapest decision available.

7. The forecast to completion contains no hypercare. Hypercare is not a risk item. It is a known cost block with a known duration — leaving it out means the budget discussion is still ahead of you.

If four or more of these apply, the date is a decision, not a plan. That is not automatically wrong — but it should be made explicitly, by the people accountable, with quantified trade-offs.