For years, the case for moving from SAP ECC to S/4HANA was built around simplification, standardisation and real-time data. While those benefits still matter, they are no longer the full story.
Artificial intelligence is changing what customers expect from an ERP transformation. Robotics is extending digital processes into warehouses, factories and field operations. The destination is no longer simply a modern ERP platform, it is an increasingly intelligent enterprise in which systems can interpret information, recommend action and, within defined controls, execute work.
This changes the business case, architecture, programme scope and skills required to deliver S/4HANA. But AI doesn't remove the complexity of SAP transformation, it exposes it. AI is only as effective as the data, processes, integrations and governance beneath it.
That makes the quality of the S/4HANA foundation more important than ever.
This article contains insight from Conexus Managing Director Connor Wall.
AI is changing the S4HANA business case
The first generation of S/4HANA programmes was often justified by technical urgency: ageing infrastructure, the need to leave ECC, reduced customisation and a move to the cloud. Today, leadership teams increasingly expect the programme to create an operational platform for AI.
That ambition includes embedded AI for invoice matching, cash application, demand forecasting, anomaly detection and maintenance planning; natural-language interfaces such as Joule; AI agents coordinating work across functions; custom use cases built on SAP Business Technology Platform; and physical automation linking SAP decisions to machines, sensors and robots.
SAP's direction is explicit. Joule is evolving from a digital co-pilot into an interface through which agents, data and tools work together. It also positions Business Data Cloud, its knowledge graph, Joule Studio, Integration Suite and AI governance capabilities as connected parts of its business AI platform. The objective is to use trusted business context to turn intent into execution.
Crucially, this means an S/4HANA design decision made today may determine how quickly and safely a business can adopt AI tomorrow.
How customers are including AI in the S4HANA journey:
The more advanced organisations are building AI readiness into the transformation roadmap from the outset. Here's how:
1. Start with business outcomes
Implementing AI is not a usable requirement. Strong programmes define process-level outcomes first: reduce forecast error, shorten financial close, improve first-time-right production, lower unplanned downtime or reduce manual order intervention.
Valuable opportunities usually cross functions. A maintenance use case may require asset data, materials availability, workforce capacity and production plans. A working-capital use case may span sales, finance, credit and collections. Customers therefore need end-to-end process ownership rather than isolated technical pilots.
2. Make data readiness a transformation workstream
AI amplifies the quality of its environment. Clean, governed and context-rich data improves decisions; duplicated vendors, inconsistent material records and weak controls create unreliable outputs.
Data migration must therefore evolve beyond extraction, cleansing and load. Customers increasingly need semantic modelling, lineage, governance and clear ownership of the data used to ground AI. Data has become a permanent business capability.
3. Use clean core to enable innovation
Clean core is usually presented as a way to simplify upgrades, but it is also an AI-enablement strategy. When business logic is buried in heavily customised ECC code, exposing processes safely to cloud services and AI agents becomes difficult.
By adopting standard S/4HANA processes, using released APIs and placing appropriate extensions on BTP, customers create an architecture that is easier to integrate, govern and evolve. This does not mean removing every differentiating process. It means being disciplined about where that differentiation lives.
4. Apply AI to delivery as well as the end solution
AI is beginning to influence how programmes are designed and delivered. Teams are using AI-assisted tools to analyse documentation, support code assessment, generate test scenarios, accelerate knowledge retrieval and identify defects.
These capabilities improve productivity but do not remove the need for experienced judgement. Generated tests still require business validation. Proposed code requires architecture and security controls. AI-produced documentation is valuable only when it reflects the implemented process.
The likely outcome is not a project without consultants, but a project in which strong consultants cover more ground and weak controls create greater risk.
5. Establish governance before scaling
Enterprise AI introduces questions traditional ERP governance was not designed to answer. Which data can a model access? Which actions may an agent execute? Where must a human approve a decision? How are recommendations explained, monitored and audited?
These issues are especially important in regulated industries such as life sciences, energy, aerospace and financial services. Validation, privacy, cybersecurity, model risk and human oversight must be designed into the operating model rather than added shortly before go-live.
Robotics is where SAP intelligence enters the physical world
Robotics is sometimes discussed as though it sits outside ERP. In practice, SAP often provides the business context that makes physical automation useful.
A robot can move a pallet, but SAP EWM understands the warehouse task, inventory status, handling requirements and wider fulfilment priority. In manufacturing, SAP Digital Manufacturing and S/4HANA provide the production, material and quality context surrounding an automated action. In asset-intensive environments, inspection technology can feed observations into maintenance processes and work orders.
The emerging opportunity is embodied AI: AI-enabled machines that perceive and act in a physical environment while being informed by enterprise data and rules. In September 2026, SAP described customer experiments connecting Joule and its Business AI platform with robots for warehouse automation, asset inspection and material handling. Examples included inspection drones linked to SAP Asset Performance Management and humanoid robots supporting SAP Digital Manufacturing processes.
However, this remains an emerging field. Robotics programmes still face difficult questions around safety, latency, interoperability, exception handling and return on investment. Nevertheless, the boundary between enterprise applications and operational technology is becoming less distinct.
How resourcing demand is changing
The simplistic prediction is that AI will reduce the number of people required on SAP programmes. Some tasks will become faster and certain lower-value work will decline. Across the current transformation cycle, however, demand is more likely to be redistributed than removed. Here's how that looks:
1. Hybrid specialists will command a premium
Customers still need deep functional expertise across finance, manufacturing, supply chain, procurement, logistics and asset management. AI does not understand a company's target operating model by default. People must decide which processes to standardise, which exceptions are legitimate, what controls must remain and what an acceptable outcome looks like.
The strongest profiles will increasingly combine disciplines: S/4HANA functional knowledge with data literacy; BTP architecture with integration and clean-core design; EWM or Digital Manufacturing with robotics and operational technology; SAP data engineering with business semantics; security with AI governance; and change management with human-AI workflow design.
The differentiator here is context. Businesses need people who can translate between leadership, business processes, the SAP landscape and AI or robotics engineering teams.
2. Routine work will be compressed
Repetitive documentation, basic configuration support, first-pass code generation, routine test creation and low-complexity analysis are all candidates for AI-assisted productivity gains. This may lead to smaller teams in some work packages, shorter delivery windows and higher expectations of each consultant.
The long-term talent risk this creates cannot be overstated. If AI absorbs much of the entry-level work through which junior professionals learn, organisations must deliberately protect supervised experience and development pathways. Otherwise, the industry may improve short-term productivity while weakening its future expert pipeline.
3. Architecture integration data and governance demand will rise
AI increases the number of connections and decision points across the enterprise. SAP must interact with models, data platforms, external applications, factory systems, sensors and robotics fleets. This drives demand for enterprise and solution architects, BTP specialists, integration experts, API designers and professionals who understand both SAP and non-SAP ecosystems.
Clean core also changes development demand. Some businesses may need less traditional modification inside ERP, but more side-by-side extension, integration and governance capability around it.
Data roles become business-critical as well. Demand will grow for data architects, migration leads, master-data governance experts, data-product owners and engineers who understand how SAP business objects relate to operational decisions. Generic AI knowledge without SAP data context will often be insufficient.
4. Change expertise becomes harder to ignore
An AI-enabled process changes roles, approvals and accountability. Users must learn when to accept a recommendation, when to challenge it and when human judgement is mandatory. Managers need new performance measures, while risk teams need visibility of automated decisions.
This increases the need for credible change leaders, process owners, training specialists and transformation professionals who can turn technology adoption into behavioural change.
5. Customers need flexible access to specialists
Few organisations require every emerging skill throughout a multi-year programme. A robotics integration architect, AI governance lead or SAP validation specialist may be critical for one phase but unnecessary as a permanent hire.
The most resilient model combines stable internal leadership, systems-integrator capacity and targeted independent specialists deployed when needed. This gives customers access to scarce capability without becoming dependent on one partner's available bench.
The resourcing mistake to avoid
The greatest mistake is searching for AI people in isolation. AI capability without process knowledge produces demonstrations that fail to scale. SAP knowledge without AI awareness risks recreating yesterday's operating model on a newer platform. Robotics knowledge without ERP integration may automate one movement without improving the end-to-end process.
Customers should build multidisciplinary teams around measurable outcomes. For each use case, they should establish the process and financial objective, required data, level of process standardisation, permitted automation, human controls, specialist skills and measures of value, risk and adoption.
The future SAP team will be more valuable
AI will remove friction from SAP delivery and operations. It will automate elements of analysis, development, testing, support and user interaction. Robotics will automate more physical tasks across warehouses, plants and assets.
But automation doesn't eliminate expertise, it changes where expertise creates value. The premium will move from people who execute repeatable tasks towards those who design outcomes, connect technologies, govern risk and apply deep business knowledge.
For customers planning or delivering S/4HANA programmes, the message is this: do not bolt AI onto the end of the roadmap. Build the data, architecture, governance and workforce foundations now. The organisations that do this well will create a platform capable of learning, deciding and acting across digital and physical operations.
Build the team for an AI enabled SAP transformation
We support organisations across Europe and the United States with specialist SAP talent for complex S/4HANA and digital transformation programmes. Our network covers functional and technical SAP delivery, BTP and integration, data, supply chain, manufacturing, validation, cybersecurity, programme leadership and change.
Whether you need a targeted independent specialist, rapid contractor mobilisation or support building a broader transformation team, we can help identify capability gaps and provide access to proven expertise quickly. Let's talk.