“One more thing.” This phrase is more than just a marketing cliché for product launches; it implies that innovation never stops. There is always a new product, feature, or shift on the horizon, forcing companies to constantly adapt. While many of these changes are merely incremental and temporary, some are truly disruptive. The company that drives these disruptions gains a significant competitive advantage and the opportunity to lead the market. Apple did this with the iPhone, Netflix with film consumption, Spotify with music, Google Maps with navigation, and OpenAI with how we find and process information.

Regarding the one more thing disruptions, SAP was for the last decades on the reaction side. Trends in IT such as web 2.0 (AJAX, thanks to Microsoft), API first, mobile first (Apple), UX focus (humans), everything regarding Single Sign-on, cloud-first (k8s, Google), and so on. In all of these trends SAP was not initiating them. They were not the ones saying: one more thing. SAP reacted. Mostly late to the game, catching up slowly, and in the end, delivering good enough results. This is also fine as SAP acts in a risk-averse business. Burning billions for creating something new that later on will be buried as SAP simply does not sell products directly to end user is not something SAP’s customers want. SAP is used to enable companies to deliver disruption. No wonder companies like Apple, Google, Microsoft or NVIDIA are SAP customers.

The Playbook

For decades, this reactive strategy paid off. SAP successfully satisfied shareholders and maintained a strong market presence without needing to pioneer cutting-edge innovation. Being seen as relevant in a trend without having contributed something substantial is an art. When business intelligence gained traction, SAP acquired Business Objects. When SaaS grew popular, it bought SuccessFactors. When mobile access became essential, it acquired a mobile platform. The same happened with IoT — which was a wise move, considering SAP’s own in-house IoT solution suffered from severe architectural security flaws. However, SAP does more than just buy companies. When mobile apps took off in the early 2010s, SAP introduced Fiori in 2013. As cloud-native computing gained momentum, SAP evolved its offering from Neo and SCP to its current niche cloud platform, BTP. Of course, not every bet succeeded. Initiatives like Leonardo (IoT), blockchain, and NFTs ultimately failed to gain mainstream adoption.

Still, this reactive strategy delivered results. Most of the time, SAP’s solutions survived long enough and gained sufficient customer adoption to justify their maintenance costs. SAP’s teams stayed busy, a reasonable number of customers were satisfied, and the revenue kept flowing. The playbook was simple: launch a barely functional product while the hype is still relevant, secure a few early buyers, and then spend the next several years refining it into something workable.

The API hype is another perfect example. SAP reacted, but what do customers actually have today? A severe lack of APIs. At best, the standard response is a request to “open a Customer Influence request.” It makes you wonder if SAP’s product teams have any real-world experience working on the consumer side of an integration. Even when SAP gracefully decides to provide an API, chances are it is only available in the latest release, locked behind a specific cloud product you have to buy, or only partially functional; as anyone who has tried working with the Business Partner API knows all too well. Looking the UX topic of the last decade (Fiori), developers can now use Fiori Guidelines and Fiori Elements to build visually appealing apps relatively easily, these solutions quickly break down under real-world demands. Try running these apps offline, bringing the Fiori UX to non-SAP business solutions, or embedding them into a standard website. Fiori was built for (internal) business apps. Designed to replace Excel spreadsheets. Instead of staying the course long enough to fix these systemic UX issues, SAP simply lost interest, shifting its focus to react to the next market trend.

Reacting is, however, not a competitive vantage. Failing to stay the course long enough to deliver lasting results is a major problem. Not just for SAP. The company’s original DNA was rooted in core business processes, enabling customers to excel. While this core still exists and pays the bills, SAP is no longer seen as a leader in other critical areas. By shifting focus too quickly, SAP is abandoning its core identity without ever establishing a new one.

The Top-Down Trap

This approach succeeded because these changes were IT-driven backend technologies. Sold directly to C-level executives and mandated top-down. This top-down enforcement allowed SAP to deliver cumbersome, feature-incomplete software, knowing that enterprises had enough engineers to build necessary workarounds.

Will this traditional approach to handling innovation continue to work for SAP? Yes and no. As long as a shift targets the backend infrastructure — where executives can simply mandate its adoption — the answer is yes. However, any change driven by end-user demand is incredibly difficult for SAP to manage. The Fiori example highlights this perfectly. More than a decade after its introduction, it is still not widely adopted because SAP attempted to dictate a user experience revolution from the top down. In reality, users are heavily accustomed to working with classic transactions. For anything involving data tables, they still pivot to Excel, while business context remains scattered across documents, emails, websites, and chats. For developers, the shift to Fiori felt like a mandate forced upon them rather than a tool they actually wanted. As a result, countless customers continue to rely on legacy transactions or restrict Fiori usage to just a few specific Lines of Business.

Decisions regarding SAP are almost always made within closed circles, where SAP representatives and customer executives align on what they think is best for employees. Occasionally, the actual end users are brought in, though usually just to maintain the illusion of influence. The resulting IT infrastructure clearly reflects what happens when a software vendor dictates what is good for a business: if a solution generates revenue for SAP, it gets sold; if it solely benefits the customer, they are left to fend for themselves. Yet, it is not that enterprises are incapable of delivering modern software. Employees today routinely enjoy seamless, remote access to their time-tracking apps, intranets, non-SAP business platforms, and ticketing systems. This is possible because these modern applications are driven directly by end-user demand and managed by business (non-SAP) departments, completely bypassing the rigid, SAP-centric IT procurement loop.

End-User Demand

This user-driven demand is increasingly becoming a major threat to SAP, particularly in the AI era. Today, end users are natively adopting AI. They routinely use ChatGPT, carry Google Gemini on their smartphones, and navigate Windows with a dedicated Microsoft Copilot key built directly into their keyboards. They are leveraging tools like Claude so heavily that current Wall Street IPO forecasts project Anthropic’s valuation at up to $2 trillion, backed by staggering future revenue targets. To put that in perspective, $200 billion - Anthropic’s long-term revenue milestone - is roughly the entire current market capitalization of SAP itself. And what does SAP offer its users in response to this monumental AI shift? A heavily restrictive API policy. One can only hope that the executives responsible for that API strategy have already received their 360-degree feedback and are actively seeking new opportunities outside the IT industry. And that someone sane at SAP is allowing the every kind of API usage a customer wants at least for any non-PRD system.

Why is AI the “one more thing” that finally breaks SAP’s playbook? Because AI is unapologetically direct. AI is mean. From day one, ChatGPT was built exclusively for the end user. There was no enterprise hardware “box 🔗” to pitch to management—just a clean, Google-like interface with a single input field. You type text, and you get an answer. There is no complex setup, no login maze, and no need to wait for a developer to fix a broken configuration. All it takes is a web browser and a user-friendly site. By making the entry barrier as low as human language itself, AI achieved immediate mass adoption 🔗. It is no longer just tech-savvy early adopters who leverage these tools; everyday users across the globe rely on them. Once Google embedded Gemini directly into smartphones and AI-enabled its core search engine, AI became ubiquitous. People have naturally integrated AI into their daily habits—and, by extension, into their workflows. This organic, grassroots adoption is exactly where the terrain becomes incredibly dangerous for SAP.

The AI Reality Gap

The traditional enterprise playbook: closed-door sales cycles, long-running contractual negotiations, multi-year product roadmaps, and the forced bundling of foundational prerequisites just to unlock basic AI capabilities … is fundamentally broken. A company that requires more than two years just to deploy SAP Joule to its employees is operating in a past era. Anyone who has ever had the “pleasure” of managing an S/4HANA migration knows the drill: transitioning from ERP Central Component (ECC) to S/4HANA takes months, followed by an exhausting stabilization phase. Then comes the cleanup: implementing everything that was aggressively cut from the initial project scope just to meet the go-live deadline, followed by tackling the massive remaining backlog. By the time a company completes this cycle and braces for the next inevitable upgrade, years have passed, and they still might not have migrated to an SAP RISE contract.

Did the world stop while the SAP department was busy upgrading? During those exact two years, those same employees have already gained access to Microsoft Copilot and internal corporate ChatGPT instances trained on company documentation. Outside the rigid SAP ecosystem, AI is already deeply embedded in software development and actively powering customer-facing solutions. By the time executive management finally aligns with SAP on licensing terms, real-world users have already moved on—reality has already decided how AI will be used.

Furthermore, AI is rapidly turning everyday employees into “citizen developers.” Generating a fully functional application, complete with backend services directly from a text specification, is now a reality. Faced with this capability, why would users wait years for an infrastructure upgrade to S/4HANA? Why wait months for a Line of Business (LoB) manager to procure an expensive SAP (cloud) solution that will still require endless customization to be useful? From an enterprise architecture and IT governance perspective, this shadow IT is a nightmare. But for the people evaluated on actual performance, they just want to deliver results—and the consumer AI tools available to them today help them do exactly that. They don’t need SAP Joule, which might finally become available in two, three, or more years.

While enterprises will inevitably purchase and deploy Joule, it will merely serve as a niche gateway to query SAP information. The moment alternative AI agents find a way to reliably access that same SAP data, the future looks incredibly bleak for Joule—and for SAP. This is a grassroots shift that executive management cannot suppress top-down. In fact, C-level executives are being heavily judged by stakeholders on their immediate AI strategies. Announcing that the company might have a functional AI roadmap in over two years is no longer a guarantee of job security.

Businesses Can’t Wait for SAP

Access to AI is no longer a luxury; it is an aggressive, bottom-up demand from employees. Workers actively seek these tools because they drastically improve efficiency and output quality. This widespread adoption has put immense pressure on management, as employees are already leveraging AI daily — even when explicitly restricted by corporate policies. They don’t just want permission; they want their employers to officially integrate AI, and they expect the same friction-free experience they enjoy in their personal lives. Crucially, users aren’t asking for just any AI platform; they want highly capable, deeply integrated assistants that seamlessly augment their existing daily workflows. Enterprises find themselves completely aligned with their staff. Organizations desperately want their workforces to utilize AI. Both to optimize current processes and to rapidly build new solutions. In an era plagued by severe labor shortages and retiring talent, companies view AI as a vital defense mechanism to maintain productivity. The mandate for capable AI isn’t a milestone on a distant multi-year roadmap; that demand is happening right now.

The Question Everyone Must Answer

Simply put, none of this aligns with SAP’s core strengths. Their business model is built on selling to C-level executives and Line of Business (LoB) managers. Their solutions are rigidly tailored to satisfy enterprise architecture frameworks, multi-year product roadmaps, specific release cycles, and complex contractual requirements. [1] This playbook worked perfectly before the AI revolution. It will likely continue to work after the AI dust settles for non-AI solutions — but it cannot work with AI. The enterprise procurement process is simply too slow for a technology that evolves by the week.

Ask yourself: When everything is finally aligned to pave the way for Joule and the Autonomous Enterprise, how long will it actually take your company to execute? How long will it take to put all the technical, functional, and licensing prerequisites in place? Will you truly wait for your sluggish SAP landscape to be ready? Or will you force the adoption of AI today because your business simply cannot afford to wait? Will every single employee in your organization be willing to play exactly by the rigid rules SAP demands? Or will they start actively finding ways to circumvent the monolith and turn to agile, non-SAP solutions?

In the fast-moving era of AI, a software giant that relies on top-down bureaucracy and reactive strategies will inevitably be bypassed by the very users it aims to serve.

[1] Don’t get me wrong: SAP was once good in this. Currently, they are also losing on these capabilities.

Before people ask: this article was first started in January. At that time, Fiori was still the (only) go-to UI technology. The Autonomous Enterprise did not exist.