Embedding AI with SAP: What Dutch organizations are doing differently

By now, most organizations have already experimented with copilots, assistants, or isolated automation use cases. However, according to SAP Nederland and VNSG (Vereniging van Nederlandstalige SAP Gebruikers), that is not where the real challenge lies. The hard part is not starting with AI, but embedding AI structurally into business processes, data models, and daily decision‑making. Success at scale, they argue, depends far less on enthusiasm for the technology itself and far more on integration, data quality, and organizational change.

A standalone AI tool might generate attention, yet it rarely changes how a business actually operates. Therefore, sustainable value emerges when AI, underpinned by strong governance and clear objectives, is connected to the systems that drive planning, logistics, customer engagement, service, and finance. In other words, AI delivers value when it is not layered on top of the business, but built into its operational flow. In recent publications, SAP and VNSG make exactly this point, urging organizations to move beyond experimentation and focus on making their processes and data genuinely AI‑ready.

A clear example is dairy producer FrieslandCampina, which has described how it is moving toward a single global SAP platform to make its supply chain smarter and to create the foundation for broader AI adoption. SAP is explicit about the rationale where migration to SAP S/4HANA and the harmonization of processes are intended to enable wider use of AI across the value chain. FrieslandCampina’s own positioning aligns with SAP and VNSG by getting the basics right, then scaling AI. That is a far more realistic roadmap than attempting to automate on top of fragmented data landscapes and local process variations.

The FrieslandCampina case is particularly helpful as it shows what “embedded AI” actually looks like in practice. In the new environment, production, transport, energy, and sales data are captured in a more uniform way, making them usable for process optimization and for advanced analytics and AI-driven scenarios.

Importantly, FrieslandCampina also shows that embedded AI is not limited to customer‑facing use cases. SAP reports that AI is being used behind the scenes as part of the transformation itself, i.e., for software testing, validation of data conversions, and the generation of documentation during the migration to S/4HANA. This is a useful reminder that AI often delivers its first substantial value in the less visible layers of the organization, where it speeds up execution, reduces manual effort, and improves consistency long before customers ever notice it.

Another strong example comes from Heineken, which has embedded GenAI into operational decision-making and process support using SAP Business Technology Platform (SAP BTP), SAP Datasphere, and SAP’s GenAI hub. Heineken’s internal AI chatbot, “Hoppy,” is built on SAP BTP and leverages SAP Datasphere to create a unified data model, enabling employees to access consistent, reliable KPIs and business data. Through natural-language queries, employees can not only retrieve information but also generate visualizations directly inside Microsoft Teams, showcasing seamless integration between SAP and Microsoft platforms. This cross-platform approach reduces data lookup time from 15 minutes to just 1 minute, helps identify missing cost-center data in purchase orders, and demonstrates how embedding AI in daily workflows can deliver tangible efficiency and data quality improvements across the organization.

Another Dutch example comes from fashion brand Studio Anneloes. In 2026, SAP Nederland reported how the company implemented SAP Engagement Cloud to create a scalable foundation for more relevant and personalized customer communication. Previously, customer data was spread across multiple sources, requiring manual linking and analysis, while segmentation and campaign execution involved too many intermediate steps. The company underlines that the core challenge was not creativity, but scalability.

This provides a steer in the AI discussion as personalization only becomes repeatable when the underlying customer data and interaction processes are centralized. By bringing together customer profiles, purchase history, and interaction data in a single environment, Studio Anneloes enabled communication that can automatically respond to customer behavior. This is exactly the type of operating model in which embedded AI thrives, i.e., a capability integrated into lifecycle marketing, loyalty, and engagement workflows.

The example above underscores that the more advanced use of AI (e.g., Agentic AI) and derived value ultimately depend on pillars such as clear governance, unified data, manageable and standardized processes, and a platform architecture that can support automation at scale. SAP’s recent partnership with the Open Data Institute reinforces this point. In March 2026, SAP and ODI announced a collaboration focused on helping organizations build AI‑ready data infrastructures, arguing that the biggest gap is often not the AI model itself, but trust in the data that feeds it.

This focus on data trust and foundational readiness is also reflected in the way advisory firms see enterprise AI evolving in practice. KPMG Netherlands, for example, reports growing demand for applying AI directly within core business processes where the underlying data, governance, and platforms are already in place.

That perspective is seen in initiatives such as SAP Joule for Consultants, which KPMG references as a way to explore emerging AI‑driven experiences and applications. The emphasis is not on experimentation for its own sake, but on identifying where AI can be responsibly embedded into existing processes and decision flows. In that sense, organizations such as FrieslandCampina are well-positioned to benefit from new AI capabilities, e.g., multi-agents, as they have invested first in harmonization, governance, standardization, and data quality.

Trust and governance remain essential to achieving value‑driven AI outcomes. IT services providers such as Fujitsu Benelux position AI governance as a strategic differentiator through offerings such as the Enterprise AI Execution & Governance Suite. These approaches introduce structured oversight from design through go‑live, with mandatory checkpoints for safety, monitoring, and accountability. (For more information visit Fujitsu advances its AI strategy and industry focus with a governance led approach). 

The focus should shift from “Where can we try AI?” to “Which core processes are ready for AI to improve decisions, speed, and relevance?” The examples discussed here suggest that Dutch organizations making real progress are not the ones running the most pilots. Instead, they are investing early in governance, building common data foundations, standardizing workflows, and embedding intelligence into the systems that already run the company.