The notion of plug-and-play enterprise software is a fallacy. For decades, large software implementations have secretly relied on extensive services from firms like Accenture for configuration. GenAI simply makes this reality transparent, requiring customization upfront rather than dressing it up as a simple software sale.
Related Insights
The rise of AI services companies like Invisible and Palantir, which build custom on-prem solutions, marks a reversal of the standardized cloud SaaS trend. Enterprises now prioritize proprietary, custom AI applications to gain a competitive edge.
Despite the hype, LinkedIn found that third-party AI tools for coding and design don't work out-of-the-box on their complex, legacy stack. Success requires deep customization, re-architecting internal platforms for AI reasoning, and working in "alpha mode" with vendors to adapt their tools.
Traditional SaaS switching costs were based on painful data migrations, which LLMs may now automate. The new moat for AI companies is creating deep, customized integrations into a customer's unique operational workflows. This is achieved through long, hands-on pilot periods that make the AI solution indispensable and hard to replace.
The key for enterprises isn't integrating general AI like ChatGPT but creating "proprietary intelligence." This involves fine-tuning smaller, custom models on their unique internal data and workflows, creating a competitive moat that off-the-shelf solutions cannot replicate.
The one-size-fits-all SaaS model is becoming obsolete in the enterprise. The future lies in creating "hyper-personalized systems of agility" that are custom-configured for each client. This involves unifying a company's fragmented data and building bespoke intelligence and workflows on top of their legacy systems.
The ease of building applications on top of powerful LLMs will lead companies to create their own custom software instead of buying third-party SaaS products. This shift, combined with the risk of foundation models moving up the stack, signals the end of the traditional SaaS era.
Off-the-shelf AI models can only go so far. The true bottleneck for enterprise adoption is "digitizing judgment"—capturing the unique, context-specific expertise of employees within that company. A document's meaning can change entirely from one company to another, requiring internal labeling.
Unlike deterministic SaaS software that works consistently, AI is probabilistic and doesn't work perfectly out of the box. Achieving 'human-grade' performance (e.g., 99.9% reliability) requires continuous tuning and expert guidance, countering the hype that AI is an immediate, hands-off solution.
Dylan Field is skeptical that disposable, AI-generated apps will replace complex SaaS products. Real business software must handle countless edge cases, scale with teams, and support shared workflows—a level of complexity and institutional knowledge that today's agents are far from mastering.
The "last mile" difficulty of implementing AI agents makes them economically viable for huge enterprise deals (justifying custom engineering) or mass-market apps. The traditional SaaS sweet spot—the $30k-$50k mid-market contract—is currently a "missing middle" because the cost to deliver the service is too high for the price point.
