What are the most and least promising categories to invest in, within vertical SaaS?

The vertical SaaS landscape is undergoing a significant re-evaluation driven by the advent of generative AI and a sharp downturn in investor sentiment. The market narrative of a "SaaSpocalypse" gained traction in early 2024, evidenced by a report indicating a **10-to-1 ratio** of selling versus buying volume in software stocks and a general feeling that the sector was "uninvestable" [14, 17, 23]. This sentiment shift is not just cyclical; large venture funds are deprioritizing vertical SaaS as a primary focus, citing insufficient potential exit sizes to generate returns for mega-funds [2, 30]. The leveraged finance market has also cooled dramatically, with software shifting from a leading category for lending to being almost unfinanceable . This broad-based investor exit is fueled by uncertainty over which companies will be negatively impacted by AI , leading to an approximate 20% year-over-year decrease in the valuations of publicly traded vertical SaaS companies .

The least promising categories within vertical SaaS are narrow, single-function "point solutions" that do not serve as foundational systems of record [3, 6, 15]. These tools, such as survey software, are highly vulnerable to being **replaced wholesale** by custom applications built on development platforms like Replit, which allow enterprises to create tailored solutions directly on their own data infrastructure [3, 12, 27]. This trend threatens to commoditize niche functionalities that can be replicated by AI or absorbed by larger, multi-product platforms . While some argue that incumbent platforms possess durable moats like customer trust, data governance, and security that are difficult for new entrants to replicate [6, 20, 21], others contend that the concept of a durable moat in software has always been weak, particularly in dynamic sectors . The critical distinction for investors is between these vulnerable point solutions and resilient, mission-critical platforms that are better positioned to adapt by integrating AI [6, 15].

Conversely, the most promising investment opportunities are in new vertical SaaS categories where AI can codify complex, unstructured workflows for the first time, particularly in industries with **no dominant software incumbents** [5, 9, 10]. Verticals such as legal, healthcare, education, and insurance are highlighted as prime targets because they rely heavily on unstructured data like emails and PDFs and lack standardized APIs [4, 5, 10, 16]. The opportunity extends beyond merely augmenting existing processes; modern AI enables the complete replacement of outsourced vertical services, such as those provided by BPOs . A viable go-to-market strategy involves targeting these outsourced workflows, as the budget is pre-allocated and the return on investment is clear . This approach creates entirely new software categories rather than competing in saturated markets [16, 22].

Successfully navigating this new environment requires a shift in investment thesis and business model evaluation. The focus is moving toward AI-native companies that can **own the outcome** in vertical markets, rather than simply selling AI tools . For new entrants targeting service industries, a relentless focus on improving gross margins is critical to ensure they are building a scalable software company, not a low-multiple, tech-enabled service . The financial metrics are also evolving, with the traditional SaaS growth model being replaced by the potential for explosive, non-linear adoption curves characteristic of successful AI products . While AI-native companies may have structurally lower gross margins due to inference costs, they can potentially achieve higher terminal operating margins through greater efficiency in sales and engineering . Foundational model providers like OpenAI and Anthropic are largely expected to become an infrastructure layer, partnering with rather than competing against these new application-layer companies [6, 13, 24].