Your team is executing. Your product works. Customers are renewing. But somehow, growth feels heavier than it used to. Like you're running faster just to stay in place.

That's not a motivation problem. That's a platform problem.

SaaS is maturing. The hyper-growth era that defined the 2010s is consolidating and the rules that made SaaS so attractive to investors, founders, and talent are quietly being rewritten. Not all at once. Quietly, and faster than most teams realize.

This article explains what's actually happening, why capital has moved, and what it takes to build your next growth curve before the current one bends too far.

The SaaS Growth Rate Is Declining And That's Not Coming Back

Between 2021 and 2025, the average growth rate across public SaaS companies dropped from 38% to 15%. That's not a bad quarter. That's a structural shift.

Every major technology platform follows the same arc. Explosive growth for 10 to 15 years, a peak, and then a long consolidation period that lasts for generations. It happened with the railroad. With the automobile. With the internet. With mobile. SaaS is no different.

The businesses don't stop working. The premium disappears.

Customers still need software. Products still deliver value. But the valuation multipliers that turned ARR into unicorn status are gone. Permanently. The era of funding SaaS growth at any cost is over.

The Capital Didn't Dry Up. It Moved.

For over a decade, cheap capital flooded into recurring revenue businesses. Low interest rates made future revenue worth more today, and venture capital poured into SaaS at valuations that assumed hypergrowth would last forever.

When rates rose in 2022, that era ended. But here's the part most teams miss: the capital didn't disappear. It moved toward AI infrastructure.

The US has committed hundreds of billions in sovereign capital to AI development. China is matching it. The EU is mobilizing. This is no longer private capital chasing returns. This is governments treating AI as strategic national infrastructure, the same way previous generations funded the interstate highway system or the space race.

SaaS was built on cheap private capital. AI is being built on sovereign commitment. One of those can be reversed by a central bank decision. The other cannot.

This Is a Platform Switch, Not a Feature Update

The most dangerous mistake a SaaS company can make right now is treating AI as a product feature. It isn't. It's a fundamentally different infrastructure with fundamentally different economics.

Here's the core difference:

SaaS runs on I/O logic. Open a dashboard, retrieve a database row, display it. Fractions of a penny per interaction. Marginal cost near zero. Seat-based pricing works because adding a user adds almost no cost.

AI runs on compute. Every response requires billions of calculations. GPU cycles cost real money every time a user does something meaningful. The most active users become the most expensive users, the opposite of SaaS economics.

This changes everything: pricing models, infrastructure decisions, cost structures, go-to-market motion. You can't just layer AI on top of a SaaS business and expect the same economics to hold.

Two Companies, Two Outcomes

The clearest illustration of this difference is Intercom versus Zendesk.

Intercom saw its growth rate collapse from 37% to 4% over eight consecutive quarters starting in 2021. Instead of doubling down on the existing subscription model, founder Eoghan McCabe came back as CEO and made a deliberate bet: kill a $60M ARR subscription business and rebuild around Fin, an AI-native product on a consumption-based pricing model. Fin is now approaching $100M ARR, growing at 3.5x. Intercom's overall growth rate recovered from 4% to 24%, moving in the opposite direction of the SaaS market as a whole.

Zendesk took a different path. It kept optimizing its existing seat-based model, layering AI features on top of existing infrastructure rather than rethinking the foundation. The result: it was taken private in 2022 at a significant discount to its peak valuation, with growth stalling and no clear AI-native story to tell investors.

Same market. Same customer problem. Completely different decisions about what the platform shift required. One rewired the business. The other added features. The outcomes speak for themselves.

The Signal Most Founders Miss

The right moment to build your next growth curve is not when growth hits zero. It's when acceleration slows down.

If your growth rate is dropping from 30% to 20% to 15%, you're still growing, but the compounding has already reversed direction. That deceleration is the window. Wait for the decline to become obvious to your board, your investors, and your team, and you've missed the transition point.

The good news: SaaS and AI don't have to compete.

They run on different infrastructure. Different economics. Different physics. Your SaaS business keeps generating cash and building customer relationships. Your AI product builds compounding momentum. The companies that win aren't the ones who burned down their SaaS business to build AI. They're the ones who stacked the new curve on top of the old one, at the right moment, with the right architecture.

What the Transition Actually Requires

This is not a technical problem. It's an organizational one.

Every platform transition in history rewrote the rules at every layer of the business simultaneously, product, pricing, distribution, operations. Not sequentially. All at once. Mobile didn't just change the product. It changed everything else too.

Salesforce understood this when mobile arrived. They didn't just make their CRM mobile-friendly. They rebuilt their entire go-to-market motion around mobile-first workflows, acquired companies like ExactTarget to extend their platform, and repriced accordingly. The result was a decade of compounding growth while competitors who treated mobile as a UI update fell behind.

AI follows the same pattern. The companies that treat it as a one-dimensional problem lose. The ones that understand it as a complete business rewiring, rethinking pricing models, GTM motion, cost structures, and hiring priorities, are the ones building durable advantages right now.

At Millstone, we help startups make this transition without burning what already works. Using AI-native development, vibe-coding tools, and data-driven growth strategies, we help founders build the next curve while the current one still has fuel in it.

FAQ

Is SaaS actually dying?

No. SaaS businesses still deliver real value and generate real revenue. What's dying is the premium that investors placed on SaaS growth. The valuation multiples, the easy fundraising, the growth-at-any-cost era are over. The underlying businesses still work. The rules around them have changed.

How do I know if my business is at risk?

The clearest signal is deceleration, not decline. If your growth rate is trending downward quarter over quarter, even if the absolute numbers still look healthy, the compounding dynamics have already shifted. That's the moment to start building your next curve, not when growth actually stops.

Can I add AI features to my existing SaaS product without changing the business model?

You can ship AI features. But if your pricing assumes near-zero marginal costs and your new AI features carry real compute costs per active user, you have a structural problem in the making. Adding AI features is not the same as transitioning to an AI-native business model. One is a product update. The other is a platform shift.

Do I need to rebuild everything from scratch?

No, and companies that try usually fail. The goal is to architect your AI product on the foundation that already exists: your customer relationships, your domain expertise, your distribution. Great architects don't demolish what works. They design the new on top of the old.

What's the right pricing model for AI products?

Seat-based pricing breaks down when marginal costs are real. Most AI-native businesses are moving toward outcome-based or consumption-based pricing, where the customer pays for value delivered, not for access. The exact model depends on your product, but the underlying principle is the same: pricing needs to reflect the actual cost structure, not the SaaS default.

How fast does this need to happen?

Faster than it feels comfortable. The same compounding math that quietly collapsed SaaS valuations is now quietly determining who wins in AI. By the time the gap shows up in your numbers, it has usually been running long enough to be very difficult to close.

Can a small team realistically make this transition?

Yes, and small teams often have an advantage. Less legacy infrastructure, faster decision cycles, and access to modern vibe-coding and AI-native development tools that can take a product from idea to working MVP in weeks. The barrier to building is lower than it's ever been. The barrier to building the right thing is still high.

Build Your Next Growth Curve With Millstone

At Millstone, we help startups accelerate growth through AI-driven marketing, vibe-code development, and scalable product strategies. Whether you're validating an AI product idea, rebuilding your pricing model, or trying to understand where your next growth curve comes from, we help you move faster with less risk.

Get in touch