The SaaS Death March
For decades, Software as a Service (SaaS) reigned supreme.
Companies like Microsoft, Salesforce, and ServiceNow have earned massive margins built around a powerful moat: your data.
But a seismic shift is underway.
AI makes it easy for businesses to create their own custom solutions.
And gives them a powerful new reason to unify their data.
The starting gun for the SaaS Death March has been fired.
A story as old as time
High castle walls surrounded by a moat were once the ultimate defense.
Then the advent of gunpowder and cannons turned those walls into rubble.
Legacy media recently met a similar fate.
Their moats were built by expensive printing presses, studios, and local monopolies on distribution.
The internet democratized content creation and made distribution free.
And legacy media companies have been on a death march ever since.
Now it’s SaaS’s turn
SaaS company moats are built on two key factors:
- Cost and Complexity - Building software used to be complicated, time-intensive, and expensive.
- Data lock-in – The more you use them, the more of your data they hold, making it increasingly difficult to switch providers.
This is why even huge SaaS businesses continue to generate such insane margins and growth rates.
AI makes these once impenetrable moats trivial.
1. AI makes software cheap and easy to build.
The primary breakthrough that brought us ChatGPT, is being able to talk to a computer using natural language.
And to AI, code is just another language.
This is why software development is the first area to be disrupted.
Tools like Cursor and Replit are democratizing coding like smartphones and Youtube democratized video creation and distribution.
Why use the same CRM system as everyone else, when it’s cheap and easy to build one tailored to your business?
Why pay $100 per seat (or more) for Salesforce when you can develop your own custom system for a fraction of the cost?
The answer is simple, you won’t.
2. AI makes fragmented data easy to unify - and gives a powerful reason to unify it.
SaaS tools attracted businesses with useful software but locked them in by controlling their data.
For example, once you start using Slack or Teams, your company’s communication history resides with them, not you.
Even if switching providers were straightforward (it rarely is), migrating data in a usable format is nearly impossible.
This lock-in forces companies to pay exorbitant fees for commodity software - not because it’s exceptional, but because leaving is too costly.
Tools like Open AI’s recently released Operator will take data migration from being complicated and expensive to cheap and easy.
AI also gives a powerful new reason to unify your data.
Data silos have always been a challenge - CRM data in Salesforce, communication data in Slack, operations data in ServiceNow, and so on.
These silos slow decision-making, hinder innovation, and obscure the full picture of your business.
AI changes this.
It thrives on centralized, unified datasets.
The larger and more comprehensive your data, the better your AI performs.
Custom-built AI tools trained on all your data - not fragments scattered across SaaS platforms - offer a massive competitive edge.
Why settle for fragmented SaaS insights when AI can provide far superior unified insights?
Why pay SaaS premiums when unlocking your data’s full potential means moving beyond them?
The answer is simple: you won’t.
What’s Next for SaaS
SaaS companies are now faced with 2 choices:
Choice 1: Become a commodity producer
AI will soon make copying software as easy as cutting and pasting an article is today.
That means charging for software will quickly become a race to the bottom.
This is a great opportunity for AI-first startups operating with lean teams and minimal overhead to eat the incumbent’s lunches.
But this isn't an option for large SaaS companies with huge cost bases and high stock prices.
They need choice number 2.
Choice 2: Find a new form of lock-in
This is what the first movers are doing, and AI agents are the obvious choice.
AI agents are the next phase - where we don’t just talk to AI, it acts on our behalf.
This new digital workforce will integrate into every aspect of business.
It will replace certain roles entirely and heavily augment the rest.
Once these agents are deeply embedded in your business, switching platforms will be like firing your entire team and starting over.
If SaaS providers control this new workforce, their lock-in effect, growth rates, and margins will be even more extreme.
This is the new workforce.
And they want to own it.
The question is: do you want them to?
Thanks for reading.
Dave
