When AI Can Finally Act: How SAP And Google Are Turning Real Customer Signals Into Business Outcomes

What is the SAP and Google partnership for customer experience?

SAP and Google are combining enterprise data with real-time signals, advanced AI (Gemini), and interactive messaging channels to help marketers deliver highly personalized, real-time customer engagement grounded in business context.

What is agentic AI in marketing?

Agentic AI refers to AI systems that go beyond insights to plan, execute, and optimize marketing campaigns autonomously at the direction of marketers, enabling faster and more relevant customer interactions at scale.

How does this improve AI-powered customer engagement?

By connecting live customer signals with operational data, marketers can act instantly on intent – delivering the right message, through the right channel, at the exact moment it matters – while ensuring that what the customer was promised is available.

Marketers sit on rich customer and operational data, but it’s scattered across CRM, commerce, service, analytics, and media platforms. AI is generating more insights than ever, but too often those insights stay stuck in dashboards, disconnected from live execution.

The disconnect between what customers feel and what businesses believe is what we call the Engagement Divide: the growing gap between what customers expect and what brands can actually deliver. This widening gap happens because customer signals live across disconnected systems. Data arrives too late or without the right context. Then, execution happens separately from those critical insights. And while customers are the first to feel this friction, many companies don’t realize how disconnected their experiences are until it’s too late.

Now, AI is accelerating this divide. Agents can generate content, launch campaigns, and optimize engagement at unprecedented speeds. But when those agents act on incomplete, outdated, or fragmented data, they only exacerbate inconsistency and poor customer experiences.

In fact, our Global Engagement Index reveals that 82% of consumers say brands have disappointed them. Yet only 22% of brands recognize they have a problem with creating seamless experiences.

The first and most foundational capability is a bi-directional, zero-copy data integration between SAP’s Business Data Cloud (BDC) and Google BigQuery. This connection is more than a one-way data export or a batch sync; it’s a live, two-way bridge.

SAP’s operational and customer data can flow into BigQuery, and Google’s rich contextual signal data can flow back into SAP. Neither dataset needs to be moved permanently or duplicated; it’s shared fluidly across both environments.

What this unlocks for marketers: Think about what lives in Google’s ecosystem that most marketing teams can’t access natively inside their CRM or customer engagement platform today:

• Ad engagement signals: which ads a customer clicked, how long they engaged, what they searched for after
• Weather data: real-time and forecasted conditions by geography
• Geo and location signals: where customers are, where they’re traveling, regional demand patterns
• Trend data: what’s spiking in search interest, what’s emerging in a product category or destination market
• Seasonal demand indicators: low-fare windows, peak travel periods, shifting consumer intent tied to the time of year

Now, pair those signals with what SAP already holds:

• Inventory levels and fulfillment status
• Active promotional offers and finance-approved discounts
• Customer purchase history, loyalty tier, and preferences
• Open service cases and support history
• Audience insights, product details and campaign context

How this looks in practice: Imagine you’re a travel brand. Google signals tell you that spring break searches are spiking, short-haul airfare prices have just dropped to seasonal lows, and a specific customer cohort has been browsing family entertainment destinations. Your SAP data indicates that your Legoland San Diego package is in stock, the promotional offer is finance-approved and active, and this particular customer has a young child and has historically taken only short-distance trips.

Without this integration, you might serve that customer a generic “Explore San Diego” banner ad. With it, you serve them: “Hey, spring break is coming – Legoland San Diego packages are available and on sale this week.” That level of personalization converts quicker and drives more loyalty because of its relevance at exactly the moment of highest intent. And best of all, it’s grounded in the operational business context, so that when they click through, everything they were promised is actually available and waiting for them.

The second capability embeds Google’s Gemini large language model directly into SAP Engagement Cloud, specifically through the AI-assisted Content Composer feature. This means marketers can generate, iterate, and personalize both copy and images within their campaign workflows without leaving the platform or starting from a blank slate.

Critically, this isn’t a generic AI writing or image tool. The model is already aware of the full context it’s working with:

• The email’s existing content and structure
• The target audience segment and their characteristics
• Brand and campaign guidelines
• Related products being promoted
• Adjacent campaigns that link to the one you’re building

Imagine the time you’ll save because you don’t have to re-explain your brand or your audience to the AI every time—it already knows.

What this unlocks for marketers: Consider how you would target and personalize a campaign today to various audiences. Serving meaningfully different creatives and offers to each segment means briefing a designer, waiting for iterations, going through review and approval cycles – a process that can take days or weeks.

With generative content in the composer, you can now:

• Generate contextually-aware image variants and copy blocks for each distinct segment that appeals to their interests and motivations.
• Fix image alignment or orientation for promoted products to look consistent throughout the message, without waiting on source updates in the catalog
• Assemble a complete look by merchandising complementary products and generating a scene-style image that shows the full outfit or product collection together.
• Apply the creative across channels by building it once in email, then extending it to other channels with minimal additional effort.

How this looks in practice: Imagine you’re running a campaign for a new shoe line with two distinct audiences: eco-conscious buyers and urban fashion-forward consumers. Instead of briefing a designer and waiting through rounds of revisions, you prompt the AI to generate a contextually aware image aligned to brand guidelines for the eco-conscious segment, such as a shoe styled in a natural, outdoor setting.

Then, you swap to the urban segment and generate a variant that places the same shoe in a city environment, without altering the shoe itself. From there, you can regenerate copy blocks that speak to each segment’s specific motivations, pull in complementary products to complete the look, and share the finished creative across different channels. In a single session that would have previously taken days, you can have an output that is more personalized and relevant than a one-size-fits-all campaign could ever produce.

The third capability unlocks an entirely new engagement channel: Rich Communication Services (RCS). As an industry messaging standard, RCS is supported through Google infrastructure and innovation, enabling both consumers and businesses to engage in more interactive text messaging conversations

What this unlocks for marketers: Think of RCS as a significant evolution beyond traditional SMS. This technology brings the rich, interactive experience of channels like WhatsApp or LINE into the native messaging environment on Android or iOS devices (no additional app required), with capabilities that go meaningfully further by having:

• Rich cards combining text, images, and video with interactive buttons
• Suggested replies and actions that guide customers without requiring them to type
• Carousels for browsing products or options within the message thread
• Web view, which can pull up a full product page or content experience inside the channel, without the customer ever leaving the conversation

RCS transforms the customer journey from a series of handoffs between channels into a single, uninterrupted experience spanning from discovery to decision to purchase to support.

How this looks in practice: A customer receives a promotion from a guitar retailer via RCS. They’re interested, but they have questions, such as, “Will this amp work with their specific guitar? What’s the return policy?”

Instead of clicking out to a website, hunting for a chat window, and losing momentum, they simply reply within the message thread. The brand responds with interactive cards that help them navigate the right information. When they’re ready to buy, they don’t need to open a browser or re-enter their details on a new page. Instead, a web view pops up inside the RCS experience, and they complete the transaction right there.

The entire journey from awareness to consideration to purchase happens in one fluid conversation where the brand blurs marketing with customer support, all while building confidence to buy. No friction, no drop-off, no lost intent.

For marketers who have watched customers abandon their carts or lose interest between touchpoints, RCS creates an entirely new architecture for the customer journey.

The fourth capability is the most forward-looking, unlocking the full potential of the partnership’s data foundation and AI innovation. This is where a company’s live signals from Google converge with enterprise data in SAP, and a generative AI content engine to enable truly hyper-personalized advertising and customer engagement at scale for marketing organizations.

What this unlocks for marketers: As an example, rather than a marketing team manually planning and creating variations of ad creative for different segments, regions, and contexts, they can now turn to a team of AI agents to analyze all available signals, identify the highest-intent audience segments, generate the appropriate creative variants, pair them with the right offers, and assemble campaigns that can be served dynamically and refined continuously with feedback.

This is agentic marketing: where the marketer sets the goal, and the AI handles the ideation, iteration, and optimization at a scale. At SAP, this is orchestrated through Joule Assistants coordinating teams of specialized agents, which can also inter-operate with AI agents across other systems, in this case from Google.

How this looks in practice: Your brand is launching a new shoe. You tell SAP Joule your marketing goal to drive awareness and conversion for this launch and target the segment most likely to convert. SAP Joule analyzes your customer data, cross-references live Google signals (trending style searches, seasonal demand across regions, recent ad engagement patterns), and returns a recommended tactic you can run immediately or customize.

But it goes further. You flag that demand is strong, but certain regions are out of stock. So, SAP Joule adapts and, in those regions, Joule shifts the messaging from a conversion push to a nurture journey. This way, you don’t promote what you can’t deliver and waste valuable ad spend. Instead, you’re building interest for when inventory is restored. In regions where stock is healthy, serve the full conversion-focused campaign with the localized, segment-specific creative generated through the content composer.

Meanwhile, a customer in your eco-conscious segment who just searched for sustainable fashion on Google and has a loyalty discount expiring this week gets a version of the ad that emphasizes the shoe’s materials and their exclusive member offer. A customer in the urban explorer segment who’s been browsing streetwear content gets a version featuring the fashion-forward city imagery. Both are generated from the same campaign brief and automatically adhere to brand standards, so creative and marketing teams can deliver hyper-relevant content with confidence and consistency.

And when either customer clicks through, they land on a page where the product is in stock, the discount code is active, and the offer is exactly what was promised — because the SAP ERP signals ensured the campaign was only served when the operational reality could support it.

Strip away the technology for a moment, and what this partnership is really solving for is the marketer who has always known, intuitively, that they could be doing better – better targeting, timing, creative, follow-through—but who has been constrained by the manual work required to translate that instinct into action at scale.

For the SMB marketer: execute more personalized campaigns efficiently without the support of a large creative team behind them. For the enterprise marketer: run campaigns across dozens of segments, regions, and channels simultaneously without everything becoming a bottleneck. For the brand manager: ensure that the right message is sent at the right moment to avoid budget waste, disjointed customer experiences, and eroded trust.

This partnership shows what’s possible when data, AI, and channels truly work together: reaching the right customer with the right offer, in the right context, at the moment intent is highest – backed by the confidence that every promise can be delivered.