Building AI Readiness Into an SAP S/4HANA Migration

Many SAP customers treat AI as a decision to make after their SAP S/4HANA migration. But the migration is already shaping that decision. Choices about process improvement scope, deployment model, and data harmonization set the processes, data, and platform services that embedded AI will later depend on.

That does not mean every migration should become an AI project. Reopening every decision for AI would expand scope, cost, and risk well beyond what most programs can absorb. Program teams can instead pinpoint the migration decisions most likely to limit AI later and address those selectively.

cbs Corporate Business Solutions addresses this through AI Infusion, one of the options in its Brownfield+ approach. The idea is that targeted work during the migration, such as standardizing core processes and harmonizing master data, creates the foundation that analytics, planning, and later AI capabilities build on.

Why AI Plans Depend on Migration Decisions

Many SAP landscapes have grown through acquisitions, local customizations, and years of accumulated data. cbs sees many organizations pursuing AI while those landscapes remain fragmented, with inconsistent processes and uneven data quality.

The SAP S/4HANA migration is one of the points at which companies decide whether to preserve those differences or reduce them.

Deployment model is the first decision. SAP delivers several newer AI and data services through its cloud offerings, so the choice between on-premise, private cloud, and RISE with SAP affects which AI services and architectures a company can use after go-live.

Data harmonization determines what AI has to work with. If a migration carries different structures for the same customers, materials, or accounts into S/4HANA, those inconsistencies carry into every AI tool that draws on them.

Process scope affects how much value embedded AI can deliver. cbs holds that standardized business processes let tools such as SAP Joule deliver more. Local process variants give embedded AI less consistent patterns to follow.

Most of these decisions are made on cost, risk, and timeline. Each can still shape what AI can do later. That makes it useful to know whether a migration approach leaves room to revisit them once their effect on AI becomes clearer.

How AI Infusion Fits Into a Brownfield+ Migration

AI Infusion is one of the Brownfield+ options, and cbs describes it as applying embedded and custom AI use cases to core business processes.

cbs groups that work into three areas. Operational efficiency covers tasks such as summarizing RFPs, capturing meeting knowledge, and automating routine administrative work. Project acceleration applies AI to the migration itself, including code analysis, test automation, and change management. Business AI covers SAP Joule, agent-based processes, and cbs’s own tools, such as its cbs AID document automation solution.

The three areas depend on the migration to different degrees. Efficiency tools can run on existing systems; cbs AID, for example, works on SAP ECC and S/4HANA. Business AI relies more heavily on the processes and data the migration produces, and cbs’s view is that process harmonization and data readiness should come before large-scale AI initiatives.

Brownfield+ is structured so process and data harmonization can happen during the migration. cbs converts an empty copy of the existing system to S/4HANA once, then moves selected data through repeated cycles, which cbs says reduces the need to lock choices early.

Kemira’s experience shows how harmonization work during a migration can carry into business AI. During its move to SAP S/4HANA Private Cloud with cbs, the chemicals company standardized processes and harmonized data across 400 plants in 37 countries. It also set up SAP Datasphere and SAP Analytics Cloud as a shared data layer for reporting and planning.

At SAP Sapphire 2026, Kemira presented its S/4HANA move as a step toward an AI-ready enterprise, and referenced AI operators that augment work with real-time intelligence.

Keeping AI Adoption Economically Sustainable

Total cost of ownership is already a primary concern for SAP customers. cbs points to data footprint, technical debt, and system complexity as factors that shape the cost of the environment a company operates after go-live, and deployment choices add to that picture.

AI adds a separate cost layer with its own model. SAP’s premium AI capabilities, including agents, consume AI Units as they perform work, so spending depends on how widely each use case runs. Companies adopting AI Infusion need to know which use cases will generate ongoing consumption and who owns that spend.

The two layers should stay distinct in the business case. A cleaner SAP landscape may lower infrastructure or maintenance costs, but it does not reduce AI consumption. Process and data harmonization can also make AI easier to scale, but that alone does not justify adding them to the migration scope. They still need a clear business case of their own.

cbs’s view is that AI adoption should be sustainable and economically viable, which means treating cost as part of readiness from the start. Planning AI Infusion alongside the migration gives program teams a chance to estimate ongoing consumption for each likely use case and weigh it against the expected operational benefit.

That planning follows the same selective approach behind Brownfield+. AI can start where the benefit is clear and the foundation already supports it, then expand as harmonized processes and data come into use. AI readiness is strongest when the technical foundation and the cost model are planned together.

What This Means for SAPinsiders

• Use migration scoping to find AI constraints early. A selective approach lets teams review deployment, process, and data choices against the AI uses the business expects to pursue, then address the ones that matter during the migration. That gives teams a way to prepare for AI without turning the migration into a full redesign.
• Sequence AI Infusion by dependency. Some parts of cbs’s AI Infusion, such as efficiency and project acceleration tools, can start before the migration is complete, while business AI needs the harmonized processes and data the migration produces. Starting with the first group lets teams show results early while that foundation is put in place.
• Plan AI Infusion costs alongside the migration. Premium AI costs follow usage and sit apart from migration and hosting budgets. Building consumption estimates and ownership into the same planning as the migration scope gives finance and IT one view of what the migration and AI will cost to run.