# SAP and Google Cloud Deploy Agentic Commerce Architecture > Source: > Published: 2026-06-19 15:09:37.411006+00:00 # SAP and Google Cloud Deploy Agentic Commerce Architecture SAP and Google Cloud announced a collaboration to deploy an agentic commerce architecture for enterprise marketing and retail operations. According to SAP, **78%** of businesses view AI as essential for retaining customers in 2026, yet SAP reports fewer than two in five companies share customer data across CX (**37%**) or CRM (** 39%**) platforms. SAP's Architecture Center notes participation in the Agentic AI Foundation (AAIF) and lists open projects such as the **Model Context Protocol (MCP)**, **goose**, and **AGENTS.md**. PwC, reporting from Google Cloud Next, frames the effort as part of a broader agentic-commerce roadmap that emphasizes data readiness, governance, agent identities, and observability. Sources for these details include SAP News, SAP Architecture Center, PwC, and industry reporting summarized by AI News. ### What happened SAP and Google Cloud announced a collaboration to deploy an **agentic commerce architecture** intended to automate multi-agent marketing and retail operations at enterprise scale, reporting coverage appears in SAP News and industry outlets. According to SAP research cited in SAP News, **78%** of businesses say AI will be essential for retaining customers in 2026; SAP also reports that fewer than two in five companies share customer data across CX (**37%**) or CRM (** 39%**) platforms. Per SAP's Architecture Center, SAP is a Gold Member of the **Agentic AI Foundation (AAIF)**, and the AAIF manages open projects including the **Model Context Protocol (MCP)**, **goose**, and **AGENTS.md**. PwC's post from Google Cloud Next outlines a practical roadmap for agentic commerce and highlights Google Cloud's new products and infrastructure signals for supporting agentic workloads. ### Technical details Editorial analysis - technical context: Agentic commerce implementations combine multi-agent orchestration, governed data access, identity for agents, and observability. PwC's coverage of Google Cloud Next emphasizes that moving beyond agent prototypes requires a platform approach where "every agent has an identity, every action leaves a trail, and every data call hits governed, contextualized sources" (PwC). The AAIF artifacts named by SAP, notably the **MCP** and **AGENTS.md**, are intended to standardize how LLM-driven agents integrate with external data and tools, which addresses interoperability and tool-integration friction highlighted in enterprise deployments. ### Context and significance Editorial analysis: The announcement sits at the intersection of enterprise CX modernization and the rising capabilities of agentic AI. SAP's cited research points to a gap between business ambition and operational readiness: many organizations expect AI to be central to customer retention, yet reported data-sharing rates across CX systems remain low. Reporting by PwC frames the hyperscaler role as supplying the cloud-native infrastructure, governance features, and developer tooling that enterprises need to shift from experiments to production-scale agentic workflows. ### What to watch Editorial analysis: Observers should track three operational signals: - •product releases or documentation from SAP and Google Cloud that define runtime, identity, and observability primitives for agentic workflows - •adoption of AAIF standards such as the **MCP** in vendor SDKs and partner integrations - •case studies showing how enterprises resolve the cited data-sharing shortfalls (the **37%** CX and**39%** CRM figures from SAP) Additionally, practitioners should watch for third-party audits or compliance guidance addressing traceability and governance for agentic actions, which PwC flags as critical for retail and commerce use cases. ### Bottom line Editorial analysis: The coverage collectively frames agentic commerce as an infrastructure problem as much as a model problem. For enterprise practitioners, the immediate implication is that successful agentic deployments depend on investing in governed data layers, agent identity and audit trails, and adopting interoperability standards such as those emerging from the AAIF. ## Scoring Rationale This is a notable enterprise-infrastructure story: it highlights practical steps and standards work enabling agentic commerce at scale. It matters to practitioners building production agentic systems but is not a frontier-model or market-shifting release. Practice with real Retail & eCommerce data 90 SQL & Python problems · 15 industry datasets 250 free problems · No credit card [See all Retail & eCommerce problems](/problems/datasets/retail)