How does the migration from SAP to Business Central succeed?

An SAP-to-BC migration is a re-implementation: Business Central is built on its standard, the data comes across from SAP cleaned and selectively, and the old Z-logic is deliberately not rebuilt. That is exactly where the value of the move lies, because otherwise you also carry over everything that made the old system heavy.

When is the right time for the move?

The calendar now answers that question: mainstream maintenance for SAP ECC runs until the end of 2027, the optional extended maintenance at extra cost until the end of 2030. Anyone still working on ECC today is no longer deciding whether, but where to. For the mid-market the destination is by no means automatically S/4HANA: whether your organisation really needs the group-level depth of S/4 or finds the more sustainable home in Business Central is what the comparison Business Central vs. SAP in the mid-market.

In practice this means: anyone who wants to be cleanly switched over by the end of 2027 needs twelve to eighteen months of lead time depending on the company structure, because the clarification work on processes and data comes before the system change. The latest sensible starting point is therefore considerably earlier than the year suggests.

Source: SAP, maintenance strategy (support.sap.com)

Why is the move a rebuild and not a conversion?

From ECC to S/4 there are conversion paths; from SAP to Business Central there deliberately are none. That is not a drawback but the opportunity of this route: instead of lifting grown Z-programs, workarounds and historical configuration into a new shell, the business is mapped anew on the Business Central standard. Every deviation from the standard needs a justification, not every habit a replica.

For this, the custom developments are inventoried and sorted into three piles: the function is replaced by the standard or an AppSource app, it is rebuilt as an extension, or it is dropped without replacement. In practice the third pile is bigger than any IT department believes beforehand.

In what order does the migration run?

A sequence of five steps has proven itself, and the second is the one most often skipped:

1. Clarify the starting position: which modules and interfaces are actually used, where custom developments sit and what shape the data quality is in.
2. Data foundation first: clean up master data before migrating. Every error that travels along becomes more expensive in the new system than in the old.
3. Map processes to the standard: first the target in the Business Central standard, then the justified deviations, never the other way round.
4. Pilot and waves: one entity goes first, and the sequence of the rest follows the similarity of the processes, not the size.
5. Operations and enablement: key users take functional ownership of the system before the cutover forces them to.

What happens to the data from SAP?

Selective takeover instead of a full copy: master data, open items and the transaction data the business needs move to Business Central cleaned up. The old history remains readable via archive access. How the takeover runs cleanly, technically and functionally, is described in detail in the answer on data migration to Business Central .

This selection is not a cost-saving measure but architecture: the new data foundation starts clean and stays clean, because nothing enters it that nobody has checked. At the same time it is the foundation on which automation and AI will later carry.

How long does the move take and what does it cost?

Considerably less than an S/4 programme, but more than a software swap: the duration depends on the number of entities, the quality of the data and the discipline on the standard. A single entity with orderly data can be switched in months; a group of entities runs in waves over a longer period. The cost logic of an implementation, from licences to clarification work, is broken down in What does a Business Central implementation cost? auf.

What role does AI play in the migration?

A double one. At the destination waits a system in which Copilot is included in every online licence and agents take over routine processes. And along the way every hour of data cleansing pays in twice, because AI amplifies good data as much as bad. Whoever runs the migration as a data project makes the company AI-ready along the way.

The bigger picture behind this question: ERP migration: the overview.