What S/4HANA Conversion Articles Miss When You Work in EWM

There's a gap between how S/4HANA conversion projects are described and what they actually mean if you live in the warehouse layer.

A recent piece in Mittelstand Heute outlines five success factors for S/4HANA conversions in the German mid-market. It's a decent contextual read — the kind of thing you send to a business stakeholder who's asking why a conversion takes two years. But it doesn't go near the warehouse. And that's the problem I keep running into on projects.

EWM decisions get treated as downstream of the "real" S/4 conversion work. They're not. They can easily become the longest-running open item on a migration.

Here's what I'd add to that list of success factors, specifically for anyone running EWM in a mid-market S/4HANA context.

1. Decide Early: Embedded EWM or Decentralised?

This is the fork in the road that shapes everything else, and it rarely gets made early enough.

With S/4HANA, Embedded EWM is the default direction — it runs directly within the S/4 system without a separate EWM server. That changes your landscape design, your transport management, and how your basis team thinks about system sizing.

If you're migrating from WM (classic Warehouse Management in ECC) or from decentralised EWM, the architecture decision isn't just technical — it has licensing, upgrade, and operations implications. I've seen this get punted past go-live planning more than once. It shouldn't be.

At minimum, the EWM deployment model should be confirmed before the blueprint phase closes.

2. WM-to-EWM Migration Is Not a Technical Uplift — It's a Redesign

The Mittelstand Heute piece (and most S/4 conversion content) treats WM-to-EWM as a checkbox. It isn't.

If a site is running classic WM with storage types, transfer orders, and lean configuration built up over ten years, moving to EWM means rethinking:

Storage Process Types and Warehouse Process Types — EWM's process control is significantly more granular
Handling Unit management — if the site doesn't use HUs today, the go-live will surface that gap fast
RF Framework — EWM's RF transactions are not the same as LM transactions; screen flows differ, and custom LM logic doesn't migrate automatically

None of this is impossible. But calling it a technical conversion understates the work by a wide margin.

3. Labour and Yard Management Are Often Missed in Scope

Mid-market sites often run lean operations — a few dock doors, a yard that's managed in spreadsheets, labour tracked loosely in a TMS or not at all. When they move to S/4 with Embedded EWM, the temptation is to scope only what they already have.

That's fine as a starting position, but Labour Management and Yard Management in EWM are genuinely useful for mid-size distribution operations, and they're easier to activate during an initial go-live than to bolt on later. At minimum, raise the question during scoping. Let the business decide with full information.

Skipping this conversation and then reopening scope twelve months post-go-live is avoidable.

4. Interface Mapping Takes Longer Than You Think

For mid-market clients — the DACH Mittelstand that article is writing for — EWM often interfaces with:

A transport management system (sometimes SAP TM, sometimes not)
A WMS-to-shop-floor connection for production supply
Carrier or customs systems for outbound

In S/4 with Embedded EWM, some of these interfaces change. eWM-PP integration for production orders behaves differently from classic WM-PP. IDocs that worked fine in ECC may need to be reviewed against the new message types in EWM.

This isn't a reason to panic — but it is a reason to get the integration architect and the EWM consultant in the same room before the interface spec is locked.

5. Post-Go-Live EWM Stability Needs Dedicated Hypercare

EWM failures at go-live are often invisible in a normal helpdesk queue. A Transfer Order that didn't post, a Handling Unit that got stuck, a Warehouse Task that's sitting open — these don't always throw hard errors. They show up as inventory discrepancies a week later.

Build a hypercare plan that includes someone with EWM access watching LX02, queue monitors, and exception queues in the first two weeks. Not just checking tickets — actively monitoring.

Takeaway

S/4HANA conversion advice written for finance or IT stakeholders rarely reaches the warehouse layer with enough specificity. For EWM practitioners, the five things worth pushing for are: an early deployment model decision, honest WM-to-EWM scope framing, a labour/yard conversation during scoping, interface review before spec lock, and dedicated EWM monitoring in hypercare.

None of this is exotic. It's just the stuff that tends to slip until it doesn't.

If you're working through any of these on a current project, I'm happy to compare notes — find me at markus-mildner.com or drop a comment below.