When to Build Custom Software vs Using Off-the-Shelf SaaS

There’s a moment in the life of every growing business where the spreadsheet stops working. Or Airtable starts creaking. Or three different SaaS tools each hold a piece of the same data and nobody can tell which is correct.

At that point, the question comes up: do we buy another SaaS, stitch them together better, or build something custom? It’s one of the most consequential technology decisions a growing business makes, and there’s no universal right answer.

The Real Cost Comparison

The naive comparison looks like this:

• SaaS: $50/user/month, 20 users = $12,000/year
• Custom: $80,000 to build, $20,000/year to maintain

Looked at that way, SaaS wins for five years. But this misses the real math.

The SaaS cost you forgot

• Annual price increases: most SaaS raises prices 5–15% per year, compounded.
• Seat inflation: you’ll have 35 users in year 3, not 20.
• Feature gating: the tier you’ll eventually need is usually 2–3x the basic tier.
• Integration tax: connecting 8 tools often needs Zapier (~$300/month) or middleware.
• Workflow compromises: you adapt to the tool — a real cost in ops efficiency.

The custom cost you should include

• Build cost (upfront)
• Hosting ($50–$500/month)
• Maintenance (~15–25% of build cost per year)
• Feature additions (ongoing, but only what you actually need)
• Bus-factor risk — can someone else take over if your developer disappears?

The Five Signs You’ve Outgrown SaaS (and Spreadsheets)

1. You’re maintaining a “shadow system”

The SaaS does 70% of what you need; the other 30% lives in Excel, Notion, or someone’s head. Every customer request means checking three places. This is the #1 signal.

2. You pay for features you don’t use to access ones you need

You only need custom reporting, but it’s gated behind the Enterprise tier at $180/user/month alongside twenty features you’ll never touch.

3. Your competitive edge lives in the workflow itself

If the way you deliver service is a differentiator and better than the standard SaaS model, the SaaS forces you to look like your competitors. Custom lets you operationalise your difference.

4. Data is trapped in silos

Your CRM has the client, your PM tool has the project, your accounting system has the invoice, and nothing talks to anything else.

5. The vendor’s roadmap isn’t aligned with yours

You need a feature; they don’t care. You wait two years while competitors who built custom iterate monthly.

When Custom Is the Wrong Answer

• The use case is commodity. Nobody should build a custom accounting system, CRM, or email platform.
• You can’t fund ongoing maintenance. Custom software is a child you’ve committed to feed for ten years.
• The workflow is still in flux. Prototype in Airtable first.
• The team is too small. For a 5-person company, the overhead often outweighs the benefit.

The Build-vs-Buy Decision Framework

Go through these in order. The first “no” is your answer.

1. Is there a mature SaaS that does 85%+ of what we need? If yes, start there.
2. Can the missing 15% be filled with integration? If yes, do that.
3. Is the missing 15% strategic — where our advantage lives? If yes, build just that part.
4. Do we have 5+ years of runway on this business model? Custom pays off over time.
5. Can we afford to maintain it? Plan for 20% of build cost annually.

A Practical First Step

Before spending a cent on a custom build, map every tool your business uses, what data lives in each, and where the gaps are. Sometimes the answer isn’t “build custom,” it’s “consolidate” — fewer tools, used better. If you’d like a free review of your current stack, get in touch — we’ll tell you the truth, including when the answer is “you don’t need us to build anything.”