SaaSpocalypse: Is it the end of SaaS business models?

As AI rapidly reshapes customer experience, the biggest competitive advantage comes from understanding how work actually flows across an organization. In this piece from Adrian Swinscoe in CX Today, our #client, Front CEO Dan O'Connell, shares how AI is changing where SaaS creates value. As frontier models become increasingly capable, differentiation will come less from the automation on top of the software interface itself and more from the coordination layer: workflow orchestration, governance, automation, and the shared customer context that helps teams and AI agents work together seamlessly. 💪 A great read on what will define tomorrow's market winners in enterprise software!

Knowing your customer better than ever means nothing if you can't act on it fast enough. That's the uncomfortable truth many CX teams are facing right now. We've spent years building beautiful dashboards, unifying customer profiles, and layering AI on top of everything. And yet, the gap between insight and action keeps widening. The SAP Customer Experience Q2 2026 release leans hard into closing that gap. A few things caught my attention. Agentic AI in Service Cloud The shift from "recommend the next step" to "execute the next step with a human in the loop" is real now. Joule agents can now resolve tier-one service cases end to end, escalating only when confidence drops. Early adopters are reporting meaningful reductions in average handling time, not because agents replace people, but because they clear the noise. Real-time context in Sales Cloud The updated integration between SAP Sales Cloud and S/4HANA finally makes credit exposure, open orders, and delivery status visible inside the opportunity view. No more toggling between five tabs to answer one customer question. Composable commerce gets simpler The new headless building blocks in SAP Commerce Cloud reduce the front-end lift considerably. Teams that were stuck maintaining monolithic storefronts now have a cleaner migration path. Here's my honest take though. None of this matters if your data foundation is weak. Agentic AI acting on bad master data just makes wrong decisions faster. Before you chase the shiny features, ask whether your Customer Data Platform is actually delivering a trustworthy single view. Insight was never the bottleneck. Trust in the data and the courage to automate decisions were. The organizations winning with CX in 2026 are not the ones with the most data. They're the ones who shortened the distance between knowing and doing. So I'm curious : Are your teams still stuck in the insight phase, or are you actually automating decisions in production? What's the biggest blocker stopping you from letting AI act, not just advise? Which Q2 feature are you most excited to test? #SAP #CustomerExperience #SAPCX #AgenticAI #S4HANA #SAPCommerceCloud #DigitalTransformation #CRM

🤖 SAP is positioning AI agents as the foundation of its Autonomous Enterprise strategy. Christian Klein, CEO and Member of SAP’s Executive Board, says, “In the age of agentic AI, SAP is leading the way.” Yet SAP’s Q2 evidence focused mainly on future launches, beta programs, and back-office automation. Rhys Fisher asks the question CX leaders care about: where are the proven customer service outcomes? 👉 Read the full article: https://lnkd.in/e45YaPAi #AgenticAI #CustomerService #EnterpriseAI #CXTechnology #SAP

🔌 Most AI integration projects fail before they start — because teams think they need to rip out the stack first. You don't need a full platform replacement to get real-time AI working across your CCaaS, CRM, and knowledge base. Event-driven middleware and API orchestration layers let you surface the right data at the right moment inside existing agent workflows — without a multi-year migration. The pattern that actually works: treat your CCaaS as the event source, your CRM as the data enrichment layer, and your knowledge platform as the inference target. A lightweight integration bus sits between them, routing context in under 200ms so agents get AI-assisted guidance that feels native, not bolted on. Practical tip for your next architecture review — audit your webhook and API rate limits before you design anything. Most teams discover the bottleneck isn't the AI model, it's the CRM throttling real-time calls during peak volume. Design around that constraint first. ⚙️ If you're working through integration architecture on your current CX transformation, drop your setup in the comments — happy to think through it with you. #theCXGuru #contactverse.ai #GenesysCloud #AmazonConnect #ContactCenter #CX 🌐 www.thecxguru.com

If every company has access to the same LLMs, where does competitive advantage actually come from? For the last two years, the question was: "𝗪𝗵𝗶𝗰𝗵 𝗟𝗟𝗠 𝘀𝗵𝗼𝘂𝗹𝗱 𝘄𝗲 𝘂𝘀𝗲?" Today, the question is becoming: "𝗛𝗼𝘄 𝗱𝗼 𝘄𝗲 𝗯𝘂𝗶𝗹𝗱 𝗔𝗜 𝘀𝘆𝘀𝘁𝗲𝗺𝘀 𝘁𝗵𝗮𝘁 𝗮𝗿𝗲 𝗿𝗲𝗹𝗶𝗮𝗯𝗹𝗲, 𝗴𝗼𝘃𝗲𝗿𝗻𝗲𝗱, 𝗰𝗼𝘀𝘁-𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁, 𝗮𝗻𝗱 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗱𝗲𝗹𝗶𝘃𝗲𝗿 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗼𝘂𝘁𝗰𝗼𝗺𝗲𝘀?" In enterprise software, long-term value is unlikely to come from the underlying model alone. It will come from how effectively AI is embedded into business processes to deliver measurable outcomes #EnterpriseAI #ArtificialIntelligence #EnterpriseSoftware #Reconciliation #DigitalTransformation #AIAutomation #FutureOfWork #Innovation #SaaS #B2B #GenerativeAI #AgenticAI OneCap https://lnkd.in/dweb4g_k

SAP Engagement Cloud CMO on why 43% of Gen Z using AI agents to buy is forcing brands to rethink how they act on customer data.

AI for Intelligent Product Discovery Most e-commerce platforms are still designed around one assumption: Customers know exactly what they are looking for. Modern commerce is different. Customers expect businesses to understand their intent, preferences, and context—not just the keywords they enter into a search box. Traditional product discovery relies on search, categories, and filters. Enterprise AI transforms product discovery by connecting customer behavior, product intelligence, inventory, business rules, and real-time enterprise data to recommend the most relevant products. This shifts product discovery from a search capability to an enterprise intelligence capability. For organizations using SAP, this is enabled by combining SAP Commerce Cloud, SAP Business Technology Platform (SAP BTP), and SAP Business AI to create a connected intelligence layer across the commerce ecosystem. With Connected Enterprise AI, organizations can deliver: Intent-aware product discovery Personalized recommendations Context-aware shopping experiences Smarter merchandising decisions Real-time inventory awareness The competitive advantage is no longer helping customers search faster. It is helping customers discover the right products through connected enterprise intelligence. AI for Intelligent Product Discovery is the first capability in building Intelligent Commerce. How is your organization using AI to improve product discovery? #EnterpriseAI #SAPBTP #SAPCommerce #BusinessAI #ProductDiscovery #IntelligentCommerce #EnterpriseArchitecture #DigitalCommerce #CustomerExperience #ArtificialIntelligence

"SAP now has 130,000+ AI agents running across customer landscapes. Somewhere, a governance team just felt their coffee go cold." Here's the distinction that actually matters, buried under the "agents" hype cycle everyone's currently living through: agents don't answer questions. They act. Trigger one, and it plans the steps, executes across systems, monitors the result, and adapts if something changes — no human babysitting each step. That's a genuinely different animal than a chatbot with a nice UI. And they show up in three flavors: you ask directly (conversational), they surface inside the app you're already in (embedded), or — the one that should raise an eyebrow in a good way — they trigger themselves off a business event with nobody asking (system-triggered). An agent quietly watching for trade policy changes and proposing reclassifications before a human even noticed the regulation shifted. That's either the most useful thing on this list or the one your compliance team wants a very long conversation about. Here's the line that should actually reassure the skeptics: every agent action carries the requesting user's identity and authorization boundaries. An agent can't do anything you personally couldn't do. No shadow-superuser problem, at least by design. Which brings us to the part that makes 130,000 agents survivable instead of terrifying: the AI Agent Hub — one command center covering discovery, build, observability, governance, and eventual retirement, for SAP agents, custom Joule Studio agents, and third-party agents alike. Full audit trail, even at decommission. That's the unglamorous plumbing that decides whether "autonomous enterprise" is a real operating model or a compliance incident waiting to happen. The specifics are legitimately eye-catching where they land: 50-70% cash cycle time reduction from 3 coordinated agents in Treasury, 85% faster supplier creation, 34% higher first-contact resolution in service. And — important for a lot of landscapes still mid-transformation — this isn't cloud-only. Hybrid support means S/4HANA on-premise and even ECC environments get agent-led automation now, not "after your migration finishes in 2028." Caveat worth saying out loud: these are projected/targeted ranges across a 2026 rollout, not universal guaranteed outcomes — the usual "results depend on your landscape" applies here too. If your org turned on just one Autonomous domain first, which one would earn back its investment fastest — Finance, Supply Chain, or CX?

Logistics SaaS platforms have mastered the system of record. The next competitive leap is building a system of action. Today, your TMS, WMS, or forwarding platform tells users what to do. It surfaces data and provides a screen to click through. The next generation won’t just show the work; it will help do the work. Enter browser automation as a core capability layer. This isn't about bypassing external platforms or replacing your APIs. It’s about using AI agents to execute workflows directly within your own authorized environment and the systems your customers explicitly connect. The architectural principle is simple: Use APIs where data exchange is the best interface. Use browser automation where the user interface is the practical reality. Combine both to eliminate operational friction. Imagine exposing your existing features to specialized agents: • Quote Agent: Gathers shipment details, searches rates, and drafts the proposal. • Document Agent: Identifies missing customs forms, extracts data, and attaches them to the file. • Booking Agent: Validates constraints and executes authorized steps once approved. This isn't unrestricted AI clicking buttons. Production-grade browser agents require strict guardrails: deterministic workflows, explicit customer permissions, human-in-the-loop approvals for sensitive actions, and immutable audit trails. Constrained, specialized agents will always outperform general-purpose web autonomy. By bridging the gap between decision and action, you stop selling "booking functionality" and start selling software that executes logistics workflows alongside your users. You dramatically reduce context switching, manual data entry, and time-to-book. Your software already knows how the work should be done. The question for product leaders is: will your platform just display the workflow, or will it actively execute it? #Logistics #SaaS #SupplyChain #FreightTech #AgenticAI

68% of enterprise tech buyers are consolidating their AI stack. This is a big deal... New market data from North America Entry shows 68% of enterprise tech buyers are actively reducing their AI vendor count to lock in a single platform. The study found that enterprise software applications shipping with embedded AI agents will jump from 5% in 2024 to 40% in 2026. Enterprise software buyers are tired of managing twelve different point-solution logins. They want single-platform architecture where intelligence runs quietly inside existing operations. If your GTM motion relies on selling another standalone dashboard, enterprise accounts will pass. Buying criteria shifted toward background orchestration. The front-end UI is becoming secondary while the execution layer moves directly into the system of record. When enterprise buyers consolidate vendor stacks, they keep the platform that sits closest to their core data model and executes end-to-end tasks. Here is how to adjust your positioning and GTM playbook this week: → map your top enterprise customer churn risks → audit how many manual steps your tool requires to finish one job → THE MAIN ONE: rebuild your core feature into an embedded agent connector that executes inside your customer's primary workflow Build complete workflow execution directly into their existing platform stack.