The Platform Pivot

TL;DR

• Your moat isn’t code.

• Let clients build their own features. Take a cut.

• You’re not a software company. You’re a platform in labor.

A client called me last week with a question that’s been keeping SaaS founders awake at 3 a.m. He’s the CEO of a profitable, growing B2B software company. Good numbers, solid trajectory. But he and his CTO had been staring at the same uncomfortable truth.

“If software is now basically free to build,” he asked, “what does that mean for us? We’re a SaaS company. Our moat is software.”

I told him his premise needed an update.

His moat was everything he’d built around the software: the relationships, the workflows, the trust.

The Numbers Are Unambiguous

The shift has gone vertical. Eighty percent of developers now use AI coding tools in their workflows, according to Stack Overflow’s 2025 Developer Survey. GitHub Copilot alone generates 46% of code written by its 20 million users, with 90% of Fortune 100 companies now using it. The AI coding tools market hit $7.37 billion in 2025, up from $4.91 billion just a year earlier.

When Andrej Karpathy coined “vibe coding” in February 2025, describing a world where you describe what you want in plain English and watch the code materialize, it sounded like a provocation. By year’s end, Collins Dictionary named it Word of the Year. A quarter of Y Combinator’s Spring 2025 batch shipped products built on codebases that were over 95% AI-generated.

Software development costs are collapsing at a rate that makes Moore’s Law look leisurely.

The Moat You Actually Have

Cheap inputs redistribute value. So where does it go?

When my client said his moat was software, I asked him a question: If someone cloned your codebase tomorrow, made it faster, cleaner, maybe even better, how many of your thousand clients would migrate next week?

He laughed. The answer was obvious: almost none.

Why? Because switching costs are about the workflows clients have built around your product. The institutional knowledge embedded in how teams use it. The integrations with other systems. The trust that when something breaks at 2 a.m., someone will pick up the phone.

Your clients bought the ability to stop thinking about the problem you solve so they could focus on their actual work. That contract, that relationship, doesn’t get cloned with a prompt.

This is the distinction that separates companies that will thrive from those that will scramble: Your software is the medium. Your service is the moat.

The Customer You’re Letting Walk

My client gave me an example that crystallized everything. He said clients regularly leave because they want hyper-specific features his team doesn’t have time to build or maintain.

“We’re willing to lose a customer rather than take on a feature we’d have to support forever,” he told me.

In the old world, this was rational. Custom features created technical debt, fragmented the codebase, and stretched engineering resources thin. Saying no was good product discipline.

In the new world, this has “MISSED OPPORTUNITY” written all over it.

The Platform Pivot

Here’s what I told him: You’re a platform waiting to happen.

The Salesforce AppExchange offers a template. Over 90% of Salesforce customers use third-party apps from its marketplace. The ecosystem generates an estimated five to six times Salesforce’s own revenue: we’re talking roughly $120 billion in value created by partners building on top of the platform. Spending on AppExchange tools is expected to grow from $2.5 billion to $9 billion by 2033.

Salesforce never tried to build every feature. They built the foundation and let an ecosystem flourish on top of it. They take 15% of partner revenue and provide distribution, trust, and infrastructure. Everyone wins.

Now imagine this model available to every SaaS company, including yours.

When software development costs were high, only Salesforce-scale companies could afford to build extensible platforms. When those costs approach zero, the math change entirely.

The Architecture of the Future

Here’s what the transition looks like:

Today: Your SaaS is a monolithic product. You build features. Clients use them or leave.

Tomorrow: Your SaaS is a trusted platform. Clients who need something specific can describe it in natural language, and AI generates a custom tool that runs on your infrastructure, plugs into your data model, and works seamlessly with your core product.

The client who wanted that hyper-specific feature? They build it themselves, in minutes, not months. Or their consultant builds it. Or an entrepreneur builds it and sells it to other clients with similar needs.

You don’t maintain it. You provide the environment, the APIs, the guardrails, the trust. Maybe you offer an AI agent that handles maintenance for a fee. You become the operating system for an ecosystem of solutions.

Think of it as an SDK for the age of natural language programming, an environment where anyone can summon functionality into existence.

The Objection, Steelmanned

The obvious pushback: “This sounds great in theory, but my clients are busy professionals who don’t want to build software. They want solutions that work.”

This is true today. When I’ve suggested to clients that they could build tools themselves, the response is usually a polite smile and a request to leave that to the experts.

But “today” has a short shelf life.

Twenty-year-olds are learning vibe coding in school: France now offers a state-recognized diploma in it. Business majors, marketers, operations people are growing up treating AI as a collaborative tool rather than a specialized technology.

In five years, expecting software that fits your exact workflow will be baseline. The companies that have already built the platforms and ecosystems for this customization will have an insurmountable head start.

Who This Breaks

The legacy mammoths still shipping monolithic products with eighteen-month feature roadmaps will watch their install base erode one frustrated client at a time, losing ground to more adaptable ecosystems. The real threat is a platform that lets customers build for themselves.

And if your only moat was code complexity, if you’ve been overcharging for years while delivering mediocre support, slow implementations, and a roadmap that ignores what customers actually need? Those clients will leave by the truckload. The small, agile ones go first. They’re already building alternatives this quarter. The enterprises running your ERP as their backbone will take longer, switching costs are real. But it’s a slow march to the same destination.

My bet is that the largest legacy software companies won’t make this transition. Think SAP, Oracle, the giants whose margins depend on heavy licenses, mandatory maintenance contracts, and implementation projects that run for years. You don’t go from a business model engineered around IT support and complexity to evolutive, user-driven software. There’s no path there, because the path would require dismantling their own profit engine.

What This Means for Strategy

When I work with SaaS companies now, the settled question is that software development costs are falling. The open question is what to do about it.

Here’s the reframe I offer: What would it mean to become the platform your clients build on?

For my client, this meant going back to the strategic drawing board. If software is free to build, could they now pursue global markets that were previously too expensive to customize for? Are there niche verticals that could self-build on their environment, segments that were never economically viable when every feature required internal engineering?

The company that used to say “no” to custom features might become the company that says “yes, and you can build it yourself this afternoon.”

The Real Opportunity

The SaaS market hit $316 billion in 2025 and is projected to exceed $1.4 trillion by 2034, according to Fortune Business Insights. But the shape of that value creation is shifting.

The winners will be companies that built the most trust, cultivated the deepest relationships, and created the platforms where their clients’ specific needs can be met, whether by the company, by partners, or by the clients themselves.

The question is whether you understand where that value actually lives.

Value lives in the community, the service, the ecosystem, the trust.

The companies that grasp this, really grasp it, will experience the next decade as the largest opportunity since SaaS itself was invented.

---

What I’m watching: The first mid-market SaaS company to successfully launch a client-facing “build your own feature” capability. When that happens, and it will, the race will be on.

---

Sources:

• Stack Overflow 2025 Developer Survey (stackoverflow.blog)

• GitHub Copilot Statistics, Quantumrun Foresight, January 2026

• Salesforce AppExchange ecosystem data via IDC and Foundation Inc.

• Fortune Business Insights, SaaS Market Report 2025

• Collins Dictionary Word of the Year 2025

• TechCrunch, “A quarter of startups in YC’s current cohort have codebases that are almost entirely AI-generated,” March 2025