AI Is Surpassing SaaS in Revenue Milestones: What Does this Mean for the Future

There’s been a lot of discourse recently around the concept of SaaS “dying” and ultimately being replaced by AI and agents. While many experts have chimed in to challenge this claim, it’s no secret that AI startups are enjoying significantly more financial growth than SaaS companies at the moment, with Stripe releasing some damning numbers in their most recent annual report.

With Stripe CEO and co-founders Patrick and John Collison saying that AI startups are “building businesses at a record pace” and Microsoft CEO Satya Nadella claiming that SaaS should be worried about displacement, there are clear signs that this conversation isn’t going away anytime soon – if anything, it’s accelerating. 

But although AI is enjoying its moment in the sun (and with that, a lot of revenue), it doesn’t ultimately mean we’re seeing SaaS being replaced. Let’s take a closer look at the numbers and what this really means for SaaS.

AI Outpaces SaaS in Revenue Targets

As suggested by Stripe’s CEOs, AI startups are experiencing much faster revenue growth when compared to SaaS offerings.

Data from Stripe – pitting Stripe’s 100 fastest-growing AI customers against the 100 most promising SaaS companies – indicates that AI startups are reaching $1M in annualized revenue approximately four months faster than their SaaS counterparts. On top of this, AI startups that have scaled up to $30M in revenue have done so five times faster.

From their findings, Stripe pointed to AI-powered coding assistant startup Cursor reaching $100M in revenue, as well as Lovable hitting $17M in ARR (annual recurring revenue) in three months, and Bolt achieving $20M in ARR in just two months, as key players in this current ‘AI boom’.

Are LLM “Wrappers” Skewing the Numbers?

AI’s continued progression over the last year has encouraged many startups – like the successful ones above – to break into the field. According to Stanford’s 2024 AI index report, there are just over 10,000 AI startups across the top ten leading countries for AI innovation, and half of those startups are founded in the United States.

With so many new companies emerging quickly, critics are beginning to question the depth of innovation behind them. Many of these startups are described as “LLM wrappers”, which are applications that offer a thin layer of functionality or interface on top of a foundational LLM, like OpenAI’s GPT 4 or Anthropics’s Claude.

These wrappers don’t have their own unique moat and are often built on top of established third-party APIs. If these LLM providers were to introduce features that serve the same niche, the startups’ business model could be quickly undermined.

For example, a tool focused on summarizing documents may find itself in trouble if OpenAI releases a native document summarization plugin with ChatGPT.

Cursor, for instance – Stripe’s most notable AI success story – is often categorized as a wrapper, leveraging external LLMs to enhance their development workflow. While it offers deeper functionality than a basic prompt entrance (like autocomplete and debugging tools), it still depends on external model providers in order to function.

This overreliance introduces some risks, especially as more capabilities are introduced into the foundational models themselves.

Stripe mentions in their research how many of these startups they’ve showcased may be categorized as wrappers, and those doing so are “missing the point.”

They argue that domain-specific AI products like Cursor are demonstrating the true capabilities of LLMs by showcasing real-world use cases at scale. These foundational models are not equipped to do this, and the revenue generated by these domain-specific products reflects this.

Whether today’s AI startups are truly building long-lasting products or riding on the hype remains to be seen, but it certainly has put SaaS companies on high notice.

What’s Next for SaaS?

Even with AI’s growing success in revenue, it’s hard to see a reality where SaaS is ever displaced. It’s more likely that rather than replacing it, we will continue to see more integration of AI into SaaS.

Describing it as a “Microsoft narrative”, Salesforce CEO Marc Benioff is not convinced by the premise of AI replacing SaaS and believes there is no way agents can perform on their own without the necessary level of data.

Speaking in Salesforce’s Q4 call in February, Benioff said: “There is a holy trinity of AI CRM: the apps, the data, and the agents. These three things have to work together.

The agentic layer is very important. But it doesn’t operate by itself – it operates with data, with a data cloud that has to be federated through your company to all your data sources.”

Recently, we saw Klarna consolidate its SaaS relationships in order to rely more solely on agentic AI offerings. But even Klarna’s CEO, Sebastian Siemiatkowski, expressed his own doubts about this becoming the norm for companies in the future.

And a lot of this comes down to SaaS having something that AI doesn’t – hard-coded business logic. While AI agents specialize in fetching information quickly and automating your workflows, SaaS has the advantage of deep domain knowledge, as well as a structured process. Overall, it’s a more reliable option when really lined up against AI.

While SaaS might not be making as much median revenue as AI right now, it’s still quietly crushing it. Stripe’s data shows SaaS companies on its platform are growing revenue seven times faster than the S&P 500. Vertical SaaS, in particular, is booming, with 6.3% of small businesses using it hitting $1M revenue in their first year

And it’s not just tech startups benefiting – SaaS is modernizing everything from healthcare and legal to hospitality and even churches.

Although the numbers speak of the short-term success of AI, the reality that this has a detrimental impact on SaaS is still slim – or at least not apparent as of yet.

Final Thoughts

Whether it was mobile apps threatening web software or no-code platforms supposedly replacing developers, history shows that every “SaaS killer” ends up strengthening the ecosystem. The rise of AI appears to be no different.

AI will continue to reshape the software landscape, but the future is likely to be collaborative, not competitive. Rather than a zero-sum game, the growth of AI will make SaaS more powerful – not obsolete.