When you're a startup, SaaS is a gift. You get enterprise-grade CRM, accounting, and project management for a few thousand rupees a month — tools that would have cost crores to build five years ago.

But somewhere between ₹1Cr and ₹5Cr in annual revenue, SaaS starts working against you. The pricing scales with your success. The features you need are always one tier up. And the data your business runs on lives on someone else's server.

The SaaS tax you don't notice

A mid-size company with 30 employees typically runs on: CRM (₹3,000/user/month) + Project management (₹1,200/user/month) + Accounting (₹8,000/month) + HR (₹500/user/month). That's ₹15–20L per year in software subscriptions — for tools that own your data and can raise prices at will.

"We were paying ₹12L/year for software that refused to add the one feature our business model depended on. The vendor said it was 'on the roadmap.'"

The subscription fee is visible on your P&L. What's invisible: the engineering cost of working around limitations, the hours lost to manual exports, and the strategic cost of building your business process around someone else's constraints.

When the math flips

A custom CRM for a 30-person company typically costs ₹3–5L to build and ₹15–30K/year to maintain. Compare that to ₹10–18L/year in SaaS subscriptions. Payback period: 6–18 months. After that, every year is pure savings — with a system built exactly for how your business works.

What you can't do with SaaS

• Integrate directly with your ERP or legacy systems without expensive middleware
• Customise the business logic — not just the UI
• Store data on-premise for compliance (healthcare, legal, finance)
• Train an AI on your proprietary data without it leaving your servers
• Sell the software as a product to your own clients

The right question to ask

Not "is SaaS cheaper?" — but "is this software core to my competitive advantage?" If the answer is yes, you should own it. If it's truly commodity (email, payroll), SaaS is fine. The mistake most businesses make is treating core operational software as commodity.

Custom software is not for everyone. But if your business depends on a specific workflow, data structure, or integration — and you're paying ₹5L+ per year in subscriptions for it — the build case is almost always there.

Vendor lock-in sounds like a technical problem. It's actually a business risk. When your operations depend on software you don't own — and can't modify — you've handed control of your business continuity to a third party.

The three types of lock-in

Data lock-in is the most dangerous. Your customer records, transaction history, and operational data live in a format only the vendor can export — or they export it in CSV files that are useless without rebuilding the relationships. Getting your data back is a ₹5–20L consulting project.

Feature lock-in happens when your process is shaped around the tool's limitations. You stop asking "what's the best way to do this?" and start asking "what's the best way to do this inside this software?" Your business adapts to the vendor's roadmap.

Price lock-in is the long game. Vendors acquire customers at low prices, then raise them once switching costs are high enough. Per-seat SaaS pricing is particularly effective at this — it escalates automatically as your team grows.

How to audit your current exposure

• Can you export 100% of your data in a usable format, today, for free?
• What would it cost (in time and money) to move to a different system?
• What happens to your business if this vendor raises prices 40% next year?
• What happens if they shut down, get acquired, or discontinue your plan?

"The vendor was acquired. New owner discontinued our plan. We had 90 days to migrate 6 years of client data. It cost us ₹18L and three months of engineering time."

The escape strategy

The goal is not to never use SaaS — it's to ensure that any SaaS you use for core operations is replaceable within 30 days. The way to achieve this: own your data model, keep all critical business logic in your own code, and treat the SaaS as a replaceable UI layer.

For core operational software — the systems your business runs on — custom development with full source delivery is the only true protection against lock-in.

Most businesses hire developers based on two things: portfolio and price. Both are unreliable signals. A beautiful portfolio tells you about past work, not about whether your project will be delivered on time and on budget. Low price tells you almost nothing except that someone is willing to underbid to get the work.

Here are the 9 questions that actually matter.

1. "Can I see the source code of a past project?"

If they hesitate, that tells you something. A studio that regularly delivers clean, documented code has nothing to hide. Ask to see the actual code structure — not just screenshots.

2. "Will you provide a fixed-price, itemised scope document before we sign?"

Change orders are how margins get made in software. A studio that can't commit to a scope hasn't understood the project well enough to build it.

3. "What happens if you deliver something that doesn't match the scope?"

The answer should be: "We fix it at no cost." Any other answer — "we'll discuss," "it depends on the hours" — is a red flag.

4. "Who owns the code after delivery?"

You should own 100% of it, unconditionally. Some studios retain IP rights or require ongoing "licence fees" for code they wrote on your budget.

5. "Can you show me your project management process?"

How do they communicate? Daily standups? Weekly reports? Do they use a project tool you have access to? Opacity during a project is a reliable predictor of surprises at delivery.

6. "What is your warranty policy after delivery?"

Industry standard is 30–90 days of bug fixes included. Anything less is a shortcut on quality assurance.

7. "Do you have a signed contract and GST invoice as standard?"

If they prefer informal payment or can't provide a proper invoice, you have no legal recourse if something goes wrong.

8. "Can I talk to a past client?"

References that are willing to take a 10-minute call are worth more than any number of portfolio screenshots.

9. "What do you do when the scope needs to change mid-project?"

Scope change is inevitable. A professional studio has a clear change-order process — not "we'll figure it out" or silent scope creep that results in a surprise invoice.

The studio that asks the most questions before the project starts is usually the one that delivers. Uncertainty in the planning phase becomes cost in the build phase.