The enduring debate of "build vs. buy" has been a foundational question since the dawn of the technology industry, influencing strategic decisions across every sector, from traditional IT departments to agile product teams. At its core, this dilemma addresses how organizations choose to acquire or develop solutions to their operational challenges. While multiple commercial "buy" alternatives often exist for common problems, each comes with inherent costs and functional limitations. Conversely, building a custom solution presents its own well-documented hurdles, encompassing significant initial creation time and expense, coupled with the ongoing burden of maintenance and evolution.
Historically, the decision has largely hinged on whether the problem being addressed falls within a company’s core competency. If a solution is deemed strategic and central to the business’s competitive advantage, the inclination has typically been to build it internally. However, for functionalities outside this core, procurement of an existing solution has been the preferred path. This general guideline, while useful, is replete with exceptions; numerous specialized problems lack viable commercial alternatives, forcing organizations to develop custom solutions regardless of core competency. Furthermore, the "buy vs. build" dichotomy has often proven an oversimplification. In the complex reality of large enterprises, it is exceedingly common to acquire a commercial solution only to embark on extensive customization to tailor it precisely to specific business needs, blurring the lines between the two approaches.
For decades, a distinct power dynamic existed within organizations: product development teams often had the luxury and resources to build bespoke solutions, while the broader "business" segments were largely reliant on a centralized IT department to fulfill an often-endless list of internal requests. This created bottlenecks and friction, with business units feeling constrained by IT’s capacity and priorities.
The Genesis of User Programming: A Historical Overview
The landscape began to shift dramatically with the advent of "user programming," a concept referring to the ability of non-technical individuals to create functional software solutions. This transformative trend traces its roots back to 1979 with the introduction of VisiCalc, the first spreadsheet program designed for personal computers, specifically the Apple II. VisiCalc was nothing short of revolutionary. It empowered accountants, financial analysts, and business managers to manipulate data, perform complex calculations, and model scenarios without needing to write traditional code or rely on IT. This represented an enormous democratization of computing power, moving beyond the realm of specialized programmers.
The success of VisiCalc paved the way for subsequent generations of spreadsheet software, most notably Microsoft Excel, which became ubiquitous across businesses worldwide. Today, countless millions of user-created programs—primarily in the form of formulas, macros, and embedded scripts—run silently within Excel and similar platforms, forming the backbone of countless operational processes. This widespread adoption underscores the profound, albeit often unacknowledged, impact of user programming.
Building on this foundation, the early 1990s saw another significant leap with the release of Visual Basic in 1991. Visual Basic provided a graphical development environment that allowed non-professional developers and "power users" to create more sophisticated applications with greater ease than ever before. It simplified the creation of user interfaces and connected to databases, effectively becoming one of the earliest widespread "low-code" options. This period marked a growing recognition that empowering more people to build applications, even if simpler ones, could unlock significant productivity gains. Even before the recent explosion of generative AI, the industry was witnessing a growing wave of low-code and no-code platforms designed explicitly to enable non-technical users to build applications, automate workflows, and create custom solutions without traditional programming knowledge. These platforms offered visual interfaces, drag-and-drop functionality, and pre-built components, making application development more accessible than ever.
Generative AI: Ushering in a New Era of User Programming
The current technological epoch is witnessing an even more profound transformation in user programming, driven by the rapid advancements in Generative AI (GenAI). This new generation of tools, exemplified by innovative platforms such as Lovable and Bolt, is fundamentally redefining the concept of a programming language. In essence, the programming language is now natural language – English, or any other human language. This paradigm shift means that the ability to articulate a problem or a desired outcome in plain language is increasingly sufficient to generate functional software. This opens up software creation capabilities to virtually anyone with a problem to solve, drastically lowering the barrier to entry for application development.
It is crucial to clarify that the choice between "build vs. buy" has always existed, and non-technical individuals have, for decades, possessed options to create solutions to varying degrees. The primary variables have historically been the specific technical skills required to operate the tools and the scope and type of applications those tools were designed to produce. What distinguishes this new generation of user programming tools, often dubbed "vibe coding" or "prompt engineering," is the dramatically reduced skill requirement—primarily natural language proficiency—and the significantly expanded range of application types that can be created. This marks a qualitative leap in accessibility and capability, moving user programming from the margins to a potential mainstream activity.
The Future of Business Software Solutions: Beyond the Hype
While the historical context of user programming is undeniably useful, the immediate impetus for this discussion is a prevailing sentiment in some circles: that the rise of these powerful new user programming tools, particularly those powered by GenAI, will lead to a future where "everyone will build, and nobody will buy." Consequently, this perspective suggests, the vast ecosystem of Software-as-a-Service (SaaS) providers is doomed to obsolescence. This article aims to explain why this sweeping conclusion is almost certainly inaccurate.
At the risk of oversimplification, the primary reason why core enterprise business software – encompassing solutions for procurement, invoicing, payments, budgeting, forecasting, payroll, staffing, sales force automation, customer relationship management (CRM), customer service, and countless others – is unlikely to be entirely supplanted by individually user-programmed solutions lies in the intricate web of business rules and the associated business logic.
What many proponents of the "everyone will build" hypothesis often fail to fully grasp is the sheer scale and complexity embedded within most enterprise business solutions. These systems are undergirded by literally thousands of often highly complex business rules and millions of lines of corresponding business logic. These rules and the code that implements them are meticulously crafted to capture and enforce critical constraints and processes related to corporate policy, regulatory compliance, data security, legal mandates, financial reporting standards, pricing structures, and much more. These are not trivial functionalities; they represent the codified intelligence of an organization’s operations and its adherence to external requirements.
Moreover, the process of discerning, defining, and codifying these rules into software is an arduous and time-consuming endeavor. The vast majority of non-technical individuals who might aspire to create business applications using new GenAI tools have little to no awareness or understanding of these underlying business rules. Even experienced technical professionals often struggle with these rules, partly because they are frequently deeply embedded within existing codebases as business logic. The individuals who originally defined and implemented these rules may have long since departed the organization, leaving behind code that functions correctly but lacks clear, up-to-date documentation explaining the rationale and nuanced implications of each rule.
This deep institutional knowledge, the ability to discern and translate complex operational requirements into viable technical solutions, is precisely what product managers—and their predecessors, business analysts—specialize in. It is this expertise that allows them to define solutions that are not only technically feasible but also strategically sound and compliant. This inherent complexity is also a major reason why addressing technical debt in legacy systems is so challenging; it requires a painstaking effort to extract and understand the embedded business rules, and then to determine which ones remain relevant and necessary for future operations.
Therefore, a robust system that effectively captures, manages, and enforces these thousands of essential business rules, ensuring that all transactions and processes are handled precisely as required, represents a significant investment of effort to create. However, it delivers immense, sustained value to an organization by ensuring operational integrity, compliance, and efficiency.
The Model Context Protocol: Enabling a Hybrid Future
Because of this profound value, while strong SaaS vendors are not likely to disappear, it is equally true that significant changes are on the horizon. The paradigm of how these business solutions are designed and consumed is evolving. Historically, these solutions were primarily created with human users in mind, with user interfaces optimized for human interaction. However, moving forward, these solutions will increasingly be utilized not just by humans, but also by intelligent AI agents and by new custom solutions crafted (whether "vibe-coded" or hand-coded) on top of these component services.
The major enabler for this sophisticated new architecture, especially critical for complex enterprises, is a technology that has been conceptually needed since the early days of the internet but is only now becoming a practical reality. The industry has long required a widely accepted protocol capable of describing business services in a manner that can be readily interpreted and acted upon by computers, not just by human users. Approximately a year ago, Anthropic, a leading AI research company, proposed the Model Context Protocol (MCP). This protocol has rapidly gained traction because it offers an elegant solution to a critical and long-standing architectural challenge: how to allow large language models (LLMs) and other AI agents to understand, interact with, and orchestrate complex business services programmatically and reliably.
The MCP Protocol provides a standardized way to represent the capabilities, inputs, outputs, and constraints of various business services in a machine-readable format. This allows AI agents to dynamically discover, understand, and invoke these services, enabling them to automate multi-step processes, synthesize information from disparate systems, and execute complex workflows without explicit, pre-programmed instructions for every single step.
If this analysis holds true, then the future of the "build vs. buy" decision will overwhelmingly be "yes to both." Companies will continue to strategically procure complex, high-value component services for critical parts of their operations. These services will remain essential for encapsulating complex business logic, ensuring compliance, and providing robust, scalable infrastructure. However, a key differentiator will be that these purchased components will be explicitly designed to be seamlessly accessed and controlled by both human users and sophisticated software entities, including AI agents. Some of this custom software will consist of AI agents acting autonomously or semi-autonomously on behalf of human users, performing tasks and orchestrating workflows. Other custom solutions will be bespoke workflows generated from GenAI tools, developed by customers or system integrators, dynamically integrating and orchestrating these underlying purchased services. These AI agents, in turn, are expected to be developed by the SaaS vendors themselves (as integrated features), by specialized system integrators, and increasingly, by end customers using advanced user programming tools.
The Age of User Programming: Implications and the Enduring Challenge
While user programming has historically operated somewhat on the margins, often confined to individual productivity hacks or departmental solutions, the new generation of AI-powered tools is swiftly bringing these capabilities into the mainstream. This broad democratization of software creation is, overall, a profoundly positive development. It promises to liberate countless individuals and teams who have long contended with severely constrained IT resources, enabling them to address their unique operational challenges with unprecedented agility and autonomy.
However, as more non-technical individuals are empowered to create solutions that extend beyond simple personal time-savers, they will inevitably need to internalize many of the hard-won lessons learned by the professional product development world. Perhaps the most important lesson, one that transcends the technical act of coding, is that the truly difficult part of software development is rarely the actual building and delivering of a solution. The far greater challenge, and the one that separates impactful solutions from costly failures, is discovering the right solution to build in the first place.
This discovery process involves deep user research, understanding underlying problems, identifying unmet needs, validating assumptions, and navigating the complex landscape of business rules and organizational constraints. While AI can make the act of building easier, it does not inherently simplify the intellectual rigor required to define a truly viable, valuable, and usable solution. This aspect of problem-solving—the strategic thinking, the empathy for the user, the understanding of the business context—will remain paramount.
The implications for various stakeholders are significant. IT departments will see their role evolve from primarily building and maintaining applications to governing, integrating, and enabling user-driven innovation. SaaS vendors will face pressure to open up their platforms through robust, machine-readable APIs and protocols, moving towards a more composable future. System integrators will find new opportunities in helping enterprises navigate this hybrid landscape, connecting purchased services with custom AI-driven workflows. Ultimately, the "Age of User Programming," powered by AI, promises to unlock immense innovation, but it also underscores the enduring importance of strategic thinking and disciplined problem discovery in the pursuit of effective technological solutions.