SAP and Google Cloud deployed an agentic commerce architecture on June 19, 2026, connecting data, AI, engagement, and commerce operations so AI agents can help shoppers discover, buy, and resolve issues across retail experiences.
For marketers, the bigger story is not just retail automation. It is the shift from product discovery on a traditional search results page to product discovery inside the AI answer itself.
| The Numbers | What It Means |
|---|---|
| 78% | of businesses consider AI essential for retaining customers in 2026, according to SAP research. |
| 37% | share customer data across customer experience platforms, exposing the core data silo problem. |
| 39% | share customer data across CRM platforms, meaning most personalization runs on partial data. |
| 3 stages | The retail sequence an agent can now handle end to end: search, transaction, and post-sale resolution. |
What Did SAP and Google Cloud Announce?
Summary: SAP and Google Cloud expanded their partnership into an agentic customer experience architecture that lets AI agents discover, buy, and service products against a retailer backend without a human clicking through every step.
On June 19, 2026, SAP and Google Cloud deployed an agentic AI commerce architecture that connects data, AI, engagement, and commerce operations, first reported by AI News.
The headline problem they are solving is structural. SAP research shows 78% of businesses consider AI essential for retaining customers this year, yet fewer than two in five share customer data across their customer experience or CRM systems.
That gap matters because personalization breaks when the data behind it is fragmented. If customer profiles, inventory, campaign data, and transaction records sit in disconnected systems, an AI agent cannot reliably recommend, sell, or support the right product at the right time.
To address that, SAP Commerce Cloud adopts the Universal Commerce Protocol, a standard designed to let retailers, payment gateways, and autonomous agents exchange commerce data in a consistent way.
With a shared protocol, software can run the full retail sequence on its own, from initial search to transaction to post-sale resolution, without the brand rebuilding its existing infrastructure.
If the agent layer is new to your team, our explainer on what agentic AI means for marketers covers the fundamentals.
Why Is This an AEO Story?
Summary: SAP plans to surface merchant products organically inside the Gemini app and Google Search, including AI Mode, which means your catalog now competes to be the answer an agent returns.
This is the part marketers should not skim. A shopper can ask Gemini or Google Search AI Mode for a product, while the backend quietly handles the inventory check, cart, and payment.
The shopper may never touch a traditional product page. That is the same dynamic behind the zero-click marketing problem, but applied directly to commerce.
This is answer engine optimization applied to product discovery. If your product data is not structured for a machine to retrieve, verify, and trust, the agent can recommend a competitor instead.
The same playbook used to get a brand cited by ChatGPT, Gemini, Claude, and Perplexity now points straight at the storefront.
How Does the SAP and Google Cloud Architecture Work?
Summary: A Gemini-powered Shopping Assistant sits on the front end while a zero-copy BigQuery link feeds it live inventory and behavioral data, so it only recommends products it can actually fulfill.
SAP Commerce Cloud integrates Google Gemini to power a Shopping Assistant that brands can deploy directly to consumers across chat, voice, and text. The assistant keeps context across the full shopping cycle.
Behind it, SAP Engagement Cloud and Google Cloud form an autonomous multi-agent framework built on SAP Business Data Cloud Connect for Google BigQuery.
The link is bidirectional and zero-copy, meaning data stays in place rather than being duplicated. That can reduce storage cost, latency, and data movement complexity.
BigQuery pulls in live signals like weather, location, and ad interaction rates, while SAP Customer Experience supplies profiles, transaction history, and consented engagement records.
The Shopping Assistant can query live warehouse records before it shows a product, which targets a classic retail failure: a promotional message drives demand the warehouse cannot meet.
In this setup, the system verifies supply before it makes the suggestion.
How Do Generative Campaigns Adjust Themselves?
Summary: Google Gemini models generate localized copy, imagery, and campaign variations that can adjust based on live engagement data.
Instead of configuring rigid campaign parameters, marketing teams can set a business goal and grant data access to SAP Engagement Cloud. Autonomous agents then segment audiences using BigQuery analytics and generate content variations through Gemini.
The framework can upgrade plain text into interactive messages through Google Rich Communication Services, evaluate how each message performs, and adjust the next send without a person manually rewriting every variation.
For marketers, this changes campaign operations in a practical way. The work shifts from manually building every branch to defining the objective, data access, constraints, and measurement logic that the agent follows.
Retailers also keep ownership of the customer relationship even when the transaction happens inside a third-party agent, because the consented data flows back into SAP Customer Experience for the next cycle.
What Does Agentic Commerce Mean for Marketers?
Summary: Agentic commerce makes structured product data, live inventory accuracy, and trusted brand signals more important because AI agents choose what to surface before a shopper reaches a product page.
Agentic commerce changes the marketing job because it moves the decision point upstream. The consumer may never browse a category page, compare ten blue links, or click through several product pages.
Instead, they may ask an AI assistant for the right product and let the agent complete the next steps.
That means retailers need to make their products legible to machines, not just persuasive to humans.
Marketers should prioritize:
- Clean product titles and descriptions
- Accurate inventory and fulfillment data
- Structured product attributes
- Clear pricing and availability signals
- Reliable reviews and trust signals
- Consistent category taxonomy
- Content that explains product use cases in answer-friendly language
The brands that win agentic commerce will not only have better landing pages. They will have better data infrastructure behind those pages.
Related Reading
- What is Answer Engine Optimization?
- What agentic AI means for marketers
- The zero-click marketing problem
- How to get your brand cited by ChatGPT, Gemini, Claude, and Perplexity
- More AI news from Prompt Insider
Frequently Asked Questions
What is agentic commerce?
Agentic commerce is a model where an AI agent handles the full shopping sequence on a customer’s behalf, from finding a product to paying for it to resolving issues after the sale, by connecting directly to a retailer’s backend systems.
What is the Universal Commerce Protocol?
The Universal Commerce Protocol is a standard that lets retailers, payment gateways, and autonomous agents exchange commerce data in a consistent way, so agents can transact across platforms without each retailer building custom integrations.
How does agentic commerce affect SEO and AEO teams?
Product discovery is moving into AI surfaces like Gemini and Google Search AI Mode. To be recommended, product and inventory data must be structured, accurate, and machine-readable, which makes answer engine optimization a direct revenue lever for retail.
Do retailers lose the customer relationship to Google?
No. According to SAP, the architecture captures consented engagement data and feeds it back into the retailer’s SAP Customer Experience systems, so the brand retains ownership even when the transaction occurs inside a third-party interface.
Why does agentic commerce matter for marketers?
Agentic commerce matters because AI agents may decide which products to show before a shopper ever reaches a product page. That makes structured product data, inventory accuracy, and machine-readable brand trust signals more important for visibility and conversion.
