AI might finally make SMB SaaS work

Written by Luis Garay - Partner at Samaipata

For decades, serving SMBs has been a graveyard for startups - too expensive to acquire, too costly to serve, too prone to churn. Despite the sheer number of small and medium-sized businesses (over 23 million in Europe alone), SaaS companies have historically struggled to make the economics work.

With fewer employees, there are fewer seats to sell, and lower average contract values (ACVs) make scaling harder. The problem compounds with churn. Small businesses operate in a volatile environment, where nearly 50% fail within five years, making retention a constant battle for SaaS companies. Unlike enterprise SaaS, where expansion revenue offsets customer loss, SMB-focused companies struggle to sustain strong net revenue retention. Even best-in-class SMB SaaS players barely hover around 100% NRR, while enterprise SaaS companies routinely exceed 120%.

Margins are squeezed further by the cost to serve. SMBs still need onboarding, support, and troubleshooting, but their smaller contract values do not always justify high-touch service models. Unlike B2C users, who are accustomed to self-serve tools, small businesses do require guidance, and unlike enterprise, they lack internal IT teams to manage software rollouts. The result is a high support burden with little room to absorb it.

As a result, the SaaS playbook has always been clear: serve mid-market or enterprise. High ACVs, multi-year contracts, and expansion revenue make these customers the logical target. Meanwhile, SMBs, despite accounting for 99% of all businesses, remain largely underserved - a fragmented and complex market dismissed as economically unviable.

But AI might be rewriting the economics.

AI is collapsing the cost to serve. AI-driven customer service automation is shifting the equation, turning service into a scalable, margin-friendly function. Instead of requiring expensive human support teams, AI agents can now own end-to-end customer workflows, providing a new model: service as a software. Companies like Sierra are proving that AI-powered agents can deliver real-time, personalized support and troubleshoot issues - all at a fraction of the historical cost. An analysis from a16z shows that human agents incur a cost of c.$37 per support ticket, whereas AI agents reduce this cost to approximately $0.69 per ticket. This is not just making support more efficient, it is enabling a level of customization that rigid SaaS models could not previously offer SMBs.

And the impact of this shift is massive. Even if ACVs remain low, AI-driven efficiencies mean that cost to serve is also low, allowing for healthy margins. More importantly, it creates a flywheel: as AI enhances customer support with personalized services and SMBs operate more efficiently with an AI tech stack, their survival rates increase, and so do SaaS retention metrics. Churn goes down, net dollar retention goes up, and the entire business model stabilizes. What was once an unscalable segment now becomes a blue ocean opportunity.

So, what does this mean for SaaS companies today? Should enterprise-focused companies pivot to unlock the SMB opportunity? The answer is not straightforward. Many SaaS incumbents have built their entire operating models - pricing, sales, retention strategies - around high ACVs and multi-year contracts. Even if the economics of SMB SaaS become viable, inertia will probably keep many enterprise-first SaaS vendors from shifting focus.

This is where startups would have a headstart against incumbents. AI-native challengers can build for SMBs from day one, leveraging automation to serve a blue ocean with healthy unit economics. Without the burden of sales motion inertia and investor expectations, innovators can capitalize on this opportunity before incumbents react.

While AI is already reshaping the economics of serving SMBs, its impact on GTM strategies might be just as profound. Outbound-led motions will see a sharp decline in CAC as AI agents like Genesy AI automate sales funnel workflows, enabling leaner structures. Inbound motions may take longer to evolve since the bulk of CAC is still funneled into digital ad monopolies (+50%). However, if AI-driven assistants begin replacing traditional search for software discovery, inbound CAC could shift just as dramatically.

For investors, this moment is a call to rethink the playbooks as we know them. The frameworks that have long dictated what is scalable, profitable, and investable are being rewritten in real-time. And beyond the investor lens, this is a defining moment for SaaS companies themselves. AI is unlocking underserved segments that were once deemed unviable and now constitute an untapped frontier.