From feature races to outcome-driven software
Outcome-driven software is a model where SaaS vendors compete not on lists of features, but on whether their technology, services, and integrations produce repeatable, measurable business results for customers across real-world operational contexts shaped by data, workflows, and organizational constraints.
AI is making generic software functionality abundant, exposing how thin many traditional SaaS advantages always were. When any credible tool can tick the same checkbox requirements, the feature checklist loses its power. In mature categories, buyers already assume a baseline level of capability: a CRM that manages customer data, a workflow tool that routes tasks, a content platform that supports approvals. The differentiation now lies in whether those tools change how work is done and sustain those changes over time. Two customers can buy the same platform and end up in wildly different places: one streamlines production quickly, while another spends 18 months on unused workflows and stalled migrations before falling back to manual work. Software value has shifted from product purchase to operational consequence, and vendors that still sell “seats” and feature parity are in denial.
AI feature commoditization and the new SaaS business model
AI is now the great equalizer of software features. Vendors can no longer rely on polished interfaces, roadmaps, and sales narratives when generative models can recreate common functions with minimal effort. As one source puts it, anyone can generate software, so domain knowledge—not code—becomes the defensible moat. Markets are repricing accordingly: value is moving from tools that help humans work to software that does the work on their behalf. This is not a collapse; it is a migration of value.
The SaaS business model is also being rewritten. Per-seat pricing assumed a human in every seat; when a single agent does the work of ten people, ten seats do not follow. According to one venture firm, “the thing we called Software-as-a-Service is becoming something else … Service-as-Software.” For years, vendors protected high-margin subscription revenue by limiting their reliance on lower-margin services, even though those services were often needed for adoption. That tradeoff is breaking down. Founders are no longer selling tool access; they are selling units of work and full capability systems that combine technology, data, governance, workflows, skills, and partner models.
Disintermediation, agents, and the myth of the SaaS apocalypse
AI agents are attacking the most visible layer of SaaS: the user interface. In this new pattern, professionals ask an AI system for an outcome, and that agent calls multiple applications behind the scenes to deliver it. You can say, “I need these things,” and a model can write code, connect to capabilities, and present an ephemeral application as your workspace, disintermediating you from the traditional app interface. By definition, that reduces direct interaction with SaaS UIs while keeping their underlying functions in play.
The scale of this shift is not theoretical. Gartner says up to USD 234 billion (approx. RM1,076 billion) in application spending is exposed to agentic arbitrage between now and 2030, and AI-driven interactions could account for roughly 20% of enterprise SaaS spending by the end of the decade. Another projection: 40% of enterprise applications will include task-specific AI agents by the end of 2026, up from under 5% in 2025. Yet the same analyst warns this is a metamorphosis, not an apocalypse; SaaS will emerge in a different form, with winners defined by how they respond to disintermediation, not whether they avoid it."The SaaS apocalypse is overrated, but disintermediation is real," one executive notes.
From tools to capability systems: what enterprises now demand
Enterprise buyers have learned a hard lesson: buying software is not the same as buying capability. Many organisations bought martech into environments riddled with fragmented data, unclear process ownership, inconsistent governance, and competing agendas. AI has made this disconnect harder to ignore by exposing operational weaknesses and showing how quickly features can be copied. Software was never valuable because it had features; it only creates value when it changes how work is done in durable ways.
As a result, the software market is becoming a market for capability systems: combinations of technology, data, integrations, permissions, workflows, governance, skills, operating roles, partner models, and value measures. The more durable moat now is operational consequence—the extent to which a platform embeds into operational, financial, governance, and decision-making fabric. Enterprise customers are demanding measurable business outcomes: faster production, fewer errors, better decisions, lower churn. Founders are responding by selling units of work rather than licenses and by going vertical, owning proprietary workflow data that generic tools and incumbents cannot easily copy. In plain terms, the winner is no longer the product that demos best; it is the one that offers the clearest path from product to working, scalable capability.
Workday, systems of record, and the road ahead
Leading vendors are not buying the “SaaS is dead” story; they are reorienting around outcomes, integration, and data. One HR and finance giant describes its ambition as building the “front door to work,” where employees log in once, ask natural-language questions about issues like payroll variations, and receive personalised answers grounded in enterprise data. This is outcome-driven software in action: value defined by whether people solve problems faster, not whether they click through nicer menus.
These vendors are betting that durable value sits in trusted systems of record and cross-business workflows, even if agents handle more of the interaction layer. Another vendor argues that, despite AI disruption, operational data will still live in core platforms that structure how organisations run. At the same time, history suggests that new intermediaries reshape, but do not erase, incumbent roles. The more serious risk is poorly conceived AI projects: the same analyst firm that predicts an agent boom also expects more than 40% of agentic AI projects to be cancelled by the end of 2027 due to cost, weak governance, or unclear value. The message for SaaS leaders is stark: integration, data quality, and strategic implementation are no longer optional; they are the product.
