Stop Buying Software. Start Building Leverage.

Quick exercise: go pull up your credit card statement. Count the software subscriptions. I'll wait.

If you're a typical mid-market company, you're paying for somewhere between 15 and 40 SaaS tools. CRM. Email marketing. Accounting. Inventory management. Project management. Analytics. Scheduling. Help desk. Social media management. And a dozen others you forgot you were paying for.

Productiv's 2025 State of SaaS report found that the average company with 200-500 employees spends $4.6 million annually on SaaS subscriptions. For smaller companies in the $5M-$20M revenue range, it's typically $80K-$200K per year. That's not a rounding error. That's a salary.

And here's the part that should make you angry: you don't own any of it.

The Rent Problem

Every SaaS subscription is rent. You pay monthly for access to someone else's system, running on someone else's servers, governed by someone else's roadmap. When they raise prices, you pay more. When they change features, you adapt. When they get acquired and the product dies, you start over.

This was an acceptable trade-off when the alternative was building custom software from scratch — a process that historically cost hundreds of thousands of dollars, took 12-18 months, and failed 70% of the time.

That trade-off no longer exists. AI coding has collapsed the cost and timeline of custom software by an order of magnitude. The SaaS apocalypse isn't coming. It's here. Wall Street has already repriced the entire sector.

“Software spending as a percentage of revenue is declining for the first time in two decades. Companies are replacing generic tools with purpose-built AI systems at a fraction of the cost.” — Bain Technology Report, 2025

What Leverage Actually Means

Naval Ravikant has this concept that stuck with me: the modern forms of leverage are code, media, and capital — things that work while you sleep, with zero marginal cost. Software is leverage. But only if you own it.

When you pay $200/month for a CRM, you're renting leverage. You get the features they built for everyone, configured the way they allow, integrated with the tools they support. It works. But it's generic. It doesn't know your business. It doesn't learn from your operations. And if they double the price, your options are “pay it” or “start over.”

When you build a custom system — even a simple one — you own the leverage. It knows your workflows because it was built for your workflows. It integrates with your specific stack because it was designed to. It learns from your data because that's its only job.

And nobody can take it away from you.

The Cost Inversion

Here's where the math has fundamentally changed. In 2020, building a custom dashboard that connects your ERP, eCommerce platform, and CRM might have cost $50K-$100K and taken 3-6 months. In 2026, an AI-assisted developer can build the same thing in 2-3 weeks for $5K-$15K.

GitHub reports that developers using AI tools write code 55% faster. In my experience with operator-led development, the multiplier is higher — because the bottleneck was never typing speed. It was translating business requirements into technical specifications. AI eliminates that translation layer.

Let me give you a concrete comparison:

• Off-the-shelf CRM (HubSpot Pro): $1,200/month × 12 = $14,400/year. Generic features. Their roadmap. Their data policies. Price increases at renewal. Customization limited to what they allow.
• Custom CRM built on Supabase + Next.js: $5K-$10K to build. $50-$100/month for hosting and database. Exactly the features you need. Your data, on your terms. Modifiable anytime. Ownable forever.

The custom build pays for itself in 4-8 months. After that, it's pure savings — $13K+ annually that goes to your bottom line instead of someone else's ARR.

What You Should Own vs. Rent

I'm not saying cancel every subscription tomorrow. Some tools are genuinely better rented. The question is: which tools give you competitive advantage, and which are commodities?

Keep renting (commodities):

• Email (Gmail, Outlook) — nobody needs custom email
• Payment processing (Stripe, Square) — compliance and security make this a no-brainer
• Basic accounting (QuickBooks) — well-regulated, rarely needs customization
• Domain hosting, DNS, CDN — infrastructure utilities

Start owning (competitive advantage):

• Customer intelligence. How you understand, segment, and respond to customers should be proprietary. A generic CRM treats every business the same. Your customer relationships are unique.
• Operational dashboards. The way you monitor and react to your business should fit your business. Generic analytics dashboards show you what the tool-maker thinks matters. Your dashboard should show you what actually matters.
• Workflow automation. The processes that run your business — order processing, fulfillment triggers, follow-up sequences — these are your operational DNA. Renting them means running your business the way the software vendor designed, not the way you operate.
• AI agents. By definition, agents that learn from your data and optimize for your goals become more valuable over time. That's not something you want owned by a third party who could raise prices, change terms, or sell to a competitor.

The Vendor Trap, Inverted

I wrote about the vendor trap — how accumulating SaaS tools creates lock-in that's expensive to escape. The leverage model inverts this trap.

Instead of being locked into 20 vendors, you're building on open-source foundations (PostgreSQL, Next.js, Python) that nobody controls. Your data lives in databases you own. Your logic lives in code you control. Your agents run on models you choose.

OSS Capital's research shows that companies building on open-source infrastructure spend 40% less on technology while maintaining 30% more flexibility in how they operate. The open-source ecosystem provides the building blocks. AI provides the assembly speed. What used to require a team of developers now requires an operator who understands the business and a builder who understands the tools.

The companies that win the next decade won't be the ones with the best software subscriptions. They'll be the ones who own their operational intelligence.

How to Start the Shift

You don't rip out your SaaS stack overnight. That would be insane. You migrate strategically:

1. Audit your subscriptions. List every tool, its annual cost, and how deeply it's integrated into your operations. The Stack Audit on this site helps with this.
2. Identify the most expensive commodity tools. These are the ones charging premium prices for features you could replicate in a weekend with AI assistance. Typically: CRM, reporting/analytics, project management, customer support ticketing.
3. Build replacements incrementally. Don't rebuild everything at once. Replace one tool. Get stable. Replace the next. Each replacement is a piece of leverage you now own.
4. Connect, don't replace, the keepers. For the tools you're keeping (accounting, email, payments), build integration layers so your owned systems can talk to them. This is the plumbing work that makes everything flow.

A realistic timeline: replace 2-3 SaaS tools in the first year, saving $30K-$80K annually, while building a custom operational layer that's worth more than all the tools it replaced.

The Endgame

The endgame isn't zero subscriptions. It's conscious ownership. You rent the commodities. You own the intelligence. You build leverage that compounds over time instead of paying rent that compounds against you.

Every month you pay for a tool you could own is a month your competitor might be building the same thing — and keeping the savings. The compounding cost of delay applies here too.

This is the shift from consumer to operator. From renter to owner. From someone who uses tools to someone who builds leverage.

If you're curious what your specific stack looks like through this lens, the Stack Audit will map it out. Or book a call and we'll walk through it together.