We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Articles
May 08, 2025
Reading Time: 5min
Is SaaS Dead
Is SaaS dead? Discover why traditional SaaS is shrinking while AI SaaS is booming. Learn how founders can adapt with smarter pricing models, LLM integrations, and bootstrapped growth strategies to thrive in 2025 and beyond
Software as as Service (SaaS) market is undeniably shrinking, from lower company evaluations to less venture capital in the ecosystem to shrinking year-over-year growth. It may seem like now is a bad time to have a SaaS company. However, these shifts are opening up new opportunities for founders and businesses.
SaaS Headwinds
It's no secret that the SaaS ecosystem is going through contractions following the pandemic boom. Some major shifts are:
- Evaluation of SaaS companies is shrinking from the 2021 peak of 17x ARR (Annual Recurring Revenue) to 6-7x ARR. This means that a SaaS company that was making $1 million in recurring revenue would go from being evaluated at ~$17M to now $6-7M.
- These lowering evaluations stem from less venture capital in the ecosystem. In 2024 SaaS companies in total raised ~$4.7B, which is down from 2023 of ~$17.4B.
- Lower capital translates into slower growth on average as fundraising (on average) increases your growth rate at the cost of equity. SaaS companies on average went from 35% YoY (Year-over-Year) growth in 2022 down to 30% YoY growth in 2024.
- Enterprises and large companies are using less SaaS in general as the major players are adding more functionality and eating up the market share of smaller niche-specific solutions.
This paints a pretty bleak picture of the future of SaaS, not to mention how it seems every niche, pain point, and problem has already been solved. You name it, it's been done.
At least so it seems...
Better, Faster, Cheaper
With recent developments in AI and better LLM models, developers are able to exponentially increase their productivity and output. Some people think that all developers do is write code in a language like JavaScript or Rust. In reality, there is so much more that goes into crafting good software, such as handling configurations, deploying cloud services, and designing good architecture.
LLM's are excellent at taking a menial task and empowering the dev to get it done in minutes instead of hours. A very common task is having to look through documentation and forum posts of an obscure tool or dependency just to use it one time. Now with LLM's those sorts of menial tasks can be completed in a couple of prompts.
Of course LLM's are no replacement for good developers as an LLM is only as useful as the person using it, and the more complex the codebase becomes, the harder it is for LLM's to keep up.
This increase in productivity means smaller teams have a bigger punch for less cost.
New Opportunities
While the SaaS industry as a whole is shrinking, AI SaaS is expected to grow by 34.7% in 2025 and projected to become a trillion-dollar market by the mid 2030s.
The choice for founders is pretty clear to integrate AI into their products.
Superficially, integrating basic LLM functionality into dashboards is an easy no-brainer. Like most radical technological advancements, the true potential isn't in sustaining innovation but in disruptive innovation.
Disruptive innovation is not just doing the same thing but better. It's about creating entirely new products, services, and platforms that simply could not exist without the new technology in question.
Ask yourself what problems can be solved with LLM's that could not be solved before? Specifically, what categories are overlooked or too small for the big players.
Before the internet, the only way to buy something was with a catalog or walking into the store. The internet enabled not just normal shopping (sustaining innovation), but radically new business models and ways of shopping such as Amazon and Ebay (disruptive innovation).
Paying-for-usage Replacing Subscriptions
Instead of the traditional freemium model, some SaaS companies are transitioning to a pay-as-you-go pricing model.
This means paying only for what you use replacing monthly subscription payments. Platforms such as AWS, Azure, and other Cloud Service Providers have pioneered this new pricing model to great success (AWS is the most profitable part of Amazon accounting for nearly half of operating profits). This model is more complex to implement, and you'll need a robust tracking and alert system for your users.
One of the biggest advantages of this pricing model is that your costs, profit, and revenue can be in near perfect alignment, reducing some of the liquidity bottlenecks that a traditional freemium pricing model experiences.
Bootstrapping
With the increasing capital constraints, profitability is becoming more and more valued over raw growth. Whereas you used to be able to raise a funding round with a pitch deck, prototype, and a plucky attitude, now VCs are looking for profitability and traction before investing.
When money was easy to obtain, it caused a glut of startups with shaky business models and a general culture of throwing money at problems until they were fixed.
This renewed focus on solid fundamentals means bootstrapping is more effective and valuable than ever. While bootstrapped companies on average grow at a slower rate, they last longer and tend to have a better product-market fit.
One of the advantages for bootstrappers is that less capital means less overall pressure for growth. Now there is a much smaller possibility of your competitors raising funding overnight and beating you on growth by out-spending you.
Takeaways To Thrive
- Focus less on headcount and more on increasing productivity per team .
- Are their new opportunities with LLM's you can use to solve problems that could not be solved before?
- Would a pay-for-usage pricing model make sense with your app?
- It's okay to grow small. Not everyone needs venture funding and you might be better without it.
I have just scratched the surface of the trend data. Below are my sources for my fellow data nerds out there.
