Skip to content
The Future of CRM: How AI, Composable Architecture and Real-Time Data Are Reshaping SAP CX
Insights · First published ·Updated by Dario Pedol ·8 min read

The Future of CRM: How AI, Composable Architecture and Real-Time Data Are Reshaping SAP CX

Dario Pedol

Dario Pedol

CEO Spadoom AG & DSAG CX Switzerland Spokesperson

Share

CRM is going through its biggest architectural change since the move to the cloud. Three forces drive it: AI that takes over parts of sales and service work, composable architecture that lets you combine components instead of accepting a monolithic suite, and real-time data that makes personalisation work in practice rather than on a slide.

Here is what each of these means in SAP CX today, where the limits are, and what we recommend doing about it.

TL;DR: AI in SAP CX now sits inside the daily tools (Joule in Sales Cloud V2 and Service Cloud V2, recommendations in Commerce Cloud, predictions in Engagement Cloud), but it only helps where data is complete and current. Composable architecture and SAP’s clean core principle let you extend and combine products without locking yourself in. The practical order: switch on the AI you already license, move integrations from batch to events, then review which components still earn their place.

How is AI changing CRM?

AI in CRM is no longer a separate product. In SAP CX it appears inside the tools people already use:

Joule in Sales Cloud V2. Joule works as a copilot for sales reps: account summaries, suggested next steps, email drafts, and alerts on deals where activity has stalled. It works because it sits inside the rep’s existing workflow, with no second tool to open. New AI capabilities such as Joule agents are being built for the V2 products; our hands-on guide to Joule in Sales Cloud V2 shows what works today.

Case classification and routing in Service Cloud V2. Incoming cases are categorised, routed by skill and workload, and matched with knowledge articles. That shortens handling time and improves first-contact resolution.

Recommendations in Commerce Cloud. Machine learning drives product recommendations, adaptive search and personalised merchandising based on browsing and purchase behaviour. Not spectacular, but it moves conversion.

Predictions in Engagement Cloud. SAP Engagement Cloud (formerly SAP Emarsys) predicts purchase likelihood, churn risk and customer value, and triggers the matching journey automatically.

There is a useful counterweight. Gartner expects that by 2030, 75% of B2B buyers will prefer sales experiences that prioritise human interaction over AI (Gartner, 2025). In B2B, AI should give people more time for customers, not replace the conversation. Which AI features are in production and which are still on the roadmap is tracked in the SAP CX AI toolkit: what is real and what is roadmap; the roadmap itself is on SAP Roadmap Explorer.

What does composable CRM architecture mean?

Composable CRM means choosing the right component for each job and connecting the components through APIs, instead of buying one suite and living with whatever each module offers. Sales Cloud V2 for sales, Engagement Cloud for marketing and a third-party tool where it fits better: composable architecture makes that possible without the landscape falling apart.

SAP CX is built for this. Each product exposes REST and OData APIs, extensions run as independent services on SAP Business AI Platform (BAIP, formerly BTP), and SAP Integration Suite connects the pieces. You do not have to use all five products.

SAP’s clean core principle supports the same idea: keep each product standard, extend through published APIs, and do not let custom code turn the landscape back into a monolith. For commerce, composable commerce in B2B describes when a decoupled storefront makes sense and when it does not.

How is real-time data changing CX?

Traditional CRM systems exchange data in batches, nightly or at best hourly. Real-time means events flow between systems as they happen: a customer opens a service case and the account manager sees it immediately; a customer orders in the shop and the profile updates everywhere.

In SAP’s landscape, SAP Event Mesh on BAIP distributes such events to the CX products and extensions that subscribe to them.

Real-time data enables what batch processing cannot: web personalisation based on the last few minutes of browsing, prices based on current stock, proactive service when a machine reports a fault, alerts when a key account’s behaviour changes. We have built these patterns in customer projects; they work when the underlying master data is clean, and that includes the ERP. Prices, stock and orders live there, so real-time CX is only as current as the ERP connection behind it.

In the next 12 months. Switch on the AI features in the SAP CX products you already license. That needs no new project. Start with account summaries and next-step suggestions in Sales Cloud V2: little effort, visible benefit.

In 12 to 24 months. Invest in real-time data. If most of your integrations still run in batches, move the important ones to event-driven patterns. AI recommendations based on yesterday’s data are rarely better than a good rep’s instinct.

In 24 to 36 months. Review your architecture for composability. Which components are not pulling their weight? Could a weak module be replaced without rebuilding everything? A clean core makes that possible. If an ERP renewal is coming in the same period, plan both together; the transformation overview shows how we sequence ERP and CX.

For a broader picture of how SAP sees agents in sales, see what agentic AI means for sales. How our own five trend calls from March 2025 held up is reviewed in SAP CX trends 2025, revisited.

Trends only pay off with the right delivery partner. Our comparison of the best SAP CX consultancies in Switzerland sets out the criteria that matter, and our page on SAP CX consulting in Switzerland shows how we apply them.

FAQ

Will AI replace sales reps?

No. AI takes over repetitive work such as data entry, research, summaries and scheduling, so reps have more time for relationships, negotiation and complex deals. Gartner expects most B2B buyers to prefer human interaction by 2030, which matches what we see in Swiss B2B sales.

What is the biggest risk of adopting AI in CRM too fast?

Trusting recommendations without judgement. AI is only as good as the data it reads: if CRM data is incomplete or one-sided, the recommendations will be too. Start with AI as a suggestion engine, not a decision engine, and fix data quality in parallel.

Is composable CRM more expensive than a suite?

It depends. Composable can reduce licence costs because you pay only for what you use, but it increases integration and operating effort. For mid-sized companies, starting with SAP CX and replacing individual components where needed is usually cheaper than assembling everything from scratch.

How does SAP CX compare to Salesforce for AI?

Both invest heavily: Salesforce with Einstein and Agentforce, SAP with Joule. SAP’s advantage is that Joule works inside the CX processes and can use ERP data such as prices, stock and order history directly. Salesforce has the larger ecosystem of third-party AI apps. Your ERP usually decides which advantage weighs more.

When will AI agents run complete sales cycles?

Not soon, at least not in B2B. Agents can already handle individual tasks such as qualifying leads, drafting emails or proposing meetings. Negotiations and complex deals need judgement and trust, and that stays with people for the foreseeable future.

SAP CXAICRM TrendsComposable ArchitectureCustomer Data
Ask Spadoom · AI assistant

Ask Spadoom

Answers drawn from what Spadoom has published on this site, with links to the pages they come from.

Try one of these

Enter to send · Shift+Enter for a new line 0 / 600
Continue with an expert Opens the contact form with your question filled in.

AI-generated answers. Verify before acting. Questions are stored anonymously, without your IP address, so we can improve our content. Please do not enter personal data.

Next step

SAP Sales Cloud V2 implementation partner

Spadoom is the SAP Sales Cloud V2 implementation partner across Switzerland, Germany, Austria and Italy. 14-week median go-live. Live customers across DACH.

Related articles