Every SAP customer running ECC now has a deadline in the diary. Mainstream maintenance ends in 2027, with extended support available beyond that at a premium. The date has been public for years, which is precisely why it has been easy to postpone.
What we see in practice is that the deadline is rarely the hard part. The hard part is that an S/4HANA move is presented internally as a technical upgrade when it is, unavoidably, a business process decision.
Brownfield, greenfield, or somewhere in between
The three routes are well documented and badly explained. In plain terms:
- Brownfield converts your existing system in place. History comes with you, including the customisations and the workarounds. Fastest to a running system, slowest to a clean one.
- Greenfield rebuilds on standard processes and migrates only the data you choose. The cleanest outcome, and the one that asks most of the business, because every deviation from standard has to be argued for rather than inherited.
- Selective transition takes some entities or company codes across at a time. Useful for groups with several plants or subsidiaries on different maturity levels.
There is no universally correct answer. There is a correct answer for a given organisation’s tolerance for disruption, and it is usually visible within two weeks of looking honestly at the custom code inventory.
Start with the custom code, not the roadmap
Run the readiness check early and read the output properly. In most ECC systems of any age, a large share of custom objects have not been executed in years. Each one still has to be assessed, and assessment is where budgets go.
Retiring dead code before the project starts is the cheapest work in the whole programme. It shrinks testing scope, it shrinks the conversion window, and it is entirely reversible if you get it wrong.
Where the estimates break
Three items are consistently under-scoped:
- Data quality. Duplicate vendors and materials that were tolerable in ECC become blocking issues under the business partner model.
- Integration. Interfaces to warehouse systems, banks, dealer portals and shop-floor equipment usually outnumber the ones anyone has documented.
- Testing and cutover rehearsal. Teams plan one dry run. Two or three is normal, and the later ones are what make the real cutover uneventful.
A sequence that holds up
Assess and clean custom code. Fix master data while the old system is still the system of record. Agree the target operating model with finance and supply chain before configuration begins. Rehearse the cutover until the runbook is boring. Then move.
None of that is novel. It is simply the order in which the work is cheapest. Programmes that go wrong usually did the same tasks — in a different sequence, under time pressure, at three times the cost.
What to decide this quarter
You do not need to pick a go-live date this quarter. You do need the custom code inventory, an honest data quality assessment and a decision on whether the business is prepared to adopt standard processes. Those three artefacts turn a deadline into a plan.
