Profit Is the New Moat for SaaS

Software used to be hard to build. That was the point.

You hired engineers, kept them for years, accumulated millions of lines of code, and ended up with something that took a competitor years to replicate. The code itself was the asset. The time required to build it was the barrier. That combination made software IP a genuine moat.

That dynamic is gone. Not weakening. Gone. The best AI models today generate production-quality code faster than most teams can review it. What used to take a team of engineers a year can now take a fraction of that time, and the gap keeps widening. When code becomes abundant, “we built this over years” stops being a defense and starts being a description of your cost structure.

So if the product is no longer the moat, what is?

Software IP no longer holds its value

The core of the old SaaS pitch was IP accumulation. You built something complex, the complexity created switching costs, and those costs kept customers in place even as competitors tried to undercut you. The pitch to investors followed the same logic: the software gets more defensible over time as the codebase grows and the integrations deepen.

That still applies in some narrow cases. Deep vertical software with years of domain-specific logic and embedded workflows has more durability than generic horizontal tools. But even there, the gap is closing faster than most founders want to admit.

The more fundamental issue is that your customers know this. They see the same AI tools you do. When a mid-market SaaS customer looks at their $50,000 annual contract and asks “could we just build this ourselves now?” the honest answer is increasingly yes, at least for the core functionality. That question is getting asked more often, and it changes the negotiating dynamic even when the customer never actually builds anything.

You cannot defend yourself with code complexity the way you once could. The companies that keep acting as if you can are the ones that will find themselves in a pricing war they did not see coming.

Twenty years of near-zero margins was always an anomaly

From roughly the mid-2000s through the early 2020s, the SaaS industry collectively decided that profit was optional. Growth was the metric. Revenue multiples rewarded top-line expansion regardless of what it cost to generate. Public SaaS companies with zero net income were valued at 20x or 30x revenue, and the implicit promise was that margins would materialize once the company reached scale.

Many never did. A 10% net margin was considered strong in the space. Plenty of public SaaS companies have never been profitable at all.

This was always unusual by the standards of any other industry. Most businesses expect meaningful profit because profit funds operations, services debt, and gets returned to owners. SaaS got a 20-year exemption from that expectation because capital was cheap, growth was fast, and the prevailing theory was that software businesses had unlimited upside if you kept investing in growth.

That exemption is expiring. Interest rates changed. Public markets stopped rewarding growth at any cost. And now AI is compressing the cost to build software so dramatically that the “we need to invest heavily in engineering to stay competitive” argument is harder to sustain.

The companies that built efficient, profitable businesses during the zero-margin era look prescient now. Hidden Levers, a wealth tech SaaS company, ran at a 52% pre-tax profit margin and sold at those margins. That was not an accident or a lucky niche. It was a deliberate choice to build efficiently and price for value. That kind of margin was achievable before AI. It is more achievable now, which means companies still running at 5% or 10% are making a choice, not facing a constraint.

You can now cut prices and improve features at the same time

Here is where AI changes the competitive math in a way that is genuinely new. For most of SaaS history, cutting prices meant accepting lower margins. Improving features meant spending more on engineering. You could do one or the other, but doing both simultaneously without hurting your unit economics was nearly impossible.

AI breaks that constraint. If you can build features faster and maintain your software with a leaner team, your cost per unit of output drops. You can pass some of that to customers through lower prices and keep the rest as margin improvement. Done right, you end up with a better product at a lower price point and healthier economics than you had before either change.

This matters strategically because it directly answers the build-vs-buy question your customers are starting to ask more often. When a customer considers building something internally, they run a rough calculation: what does it cost to build and maintain this ourselves versus what do we pay the vendor? The economics of that decision are shifting faster than most vendors realize. For a detailed look at where that line is moving, see The Future of Build vs. Buy: Throwaway Software, Dark Factories, and Liquid Code. If you have been raising prices while your product stagnates, the calculation starts favoring internal development. If you cut prices and ship meaningful improvements, the math flips back in your favor.

The goal is not to race to zero. It is to stay comfortably below the threshold where a customer rationally decides to build their own solution. That threshold is higher than most founders assume, including management overhead, context switching, and maintenance burden. That threshold is lower than it used to be software internally still has real costs even with AI assistance, including management overhead, context switching, and maintenance burden. But that threshold is lower than it used to be, and it will keep falling. Getting ahead of the shift by proactively lowering your price while improving your product is the move. Waiting until a customer actually cancels to start the conversation is the wrong order of operations.

AI compresses costs across the whole business, not just engineering

Most of the conversation about AI and SaaS margins focuses on engineering productivity. That is real, but it is a fraction of the available opportunity.

Every function in a software business is a candidate for cost compression. Marketing teams can produce more content, run more experiments, and analyze campaign performance without adding headcount proportional to output. Sales teams can qualify leads faster, personalize outreach at scale, and reduce the time from first contact to close. Product teams can synthesize customer feedback, draft specifications, and prioritize roadmaps without the overhead of large PM organizations. Back office functions, including finance, legal, HR, and customer support, all have meaningful surface area for AI assistance.

The companies that treat AI as an engineering tool and leave the rest of the business unchanged will capture partial benefit. The ones that push it into every cost center will see margin expansion that compounds across the full P&L.

This is not about eliminating teams. It is about not growing headcount at the same rate you grow revenue. A company that historically added one person for every $150,000 in new ARR might now sustain growth while adding one person for every $300,000 or $400,000. That ratio shift, applied consistently across functions, is what moves margins from 10% to 20% without requiring a single pricing change or a single layoff. It is a structural change to how the business scales, not a cost-cutting event.

The founders who get this right are the ones thinking about AI adoption as a business model question, not a technology question. The technology question is “what can AI do?” The business model question is “where exactly does AI let us grow revenue without growing cost?” Those are very different conversations, and the second one is the one that matters for margins.

20% is the floor, not a stretch goal

Twenty percent net margin should be the minimum target for a SaaS business operating with AI tools available to it. Not the aspirational ceiling. The floor. Here is why that number is the right one.

First, it provides a real buffer. The AI landscape is shifting fast enough that your cost structure and competitive position could look materially different in 18 months. Companies running at 5% margins have almost no room to absorb a misjudgment: a wrong bet on a product direction, a competitor that undercuts you, a key customer that churns. Companies at 20% can make mistakes, adjust their strategy, and still survive long enough to recover. Margin is optionality.

Second, it makes you an attractive acquisition target if and when that becomes relevant. Strategic buyers and private equity both run the same math: what is the normalized free cash flow of this business, and what multiple does it deserve? Understanding how strategic and financial buyers value a business differently is worth knowing before you are anywhere near an exit conversation. A company with 20% net margins and a clean cost structure commands a meaningfully higher multiple than an equivalent-revenue business running at breakeven. The acquirer is buying future cash flows, not revenue. If an exit is in your eventual plans, margins are one of the few variables you fully control in the years before a transaction.

Third, and most simply: profit is what you actually take home. Revenue is a vanity metric for founders. Growth is a means to an end. The end, for anyone with equity in the business, is cash. A 20% margin on $5 million in revenue is $1 million in net income. That is real money that does not require a liquidity event to materialize. Founders who spend years building companies at zero or near-zero margins and then sell are essentially giving up years of their working life for a single binary outcome. Profitable companies give you ongoing returns while you build.

The SaaS playbook of the last 20 years said profitability was something you got to eventually. The AI era says profitability is how you survive long enough for “eventually” to arrive.