SaaS vs. Custom Software: When to Build, When to Buy

We built Billwaze — our own billing SaaS — because we tried four off-the-shelf tools and none of them actually modelled an MSP contract. But we also tell clients regularly: don't build custom when a €50/month SaaS solves the problem. The question isn't which is better. It's which is right for where you are.

The bias toward building (and why it's expensive)

Technical founders default to building. It's what they know, it's what they enjoy, and it feels like control. Non-technical founders often default to building too — because they've been told by a development agency that their problem is unique and requires a custom solution.

The reality: most operational problems a growing business faces have already been solved. CRM, project management, invoicing, HR, time tracking, email marketing — the off-the-shelf market for these is deep and mature. Building custom versions of solved problems is one of the most reliable ways to burn budget without creating competitive advantage.

The three questions that decide it

Question 1: Is this a differentiator or a commodity?

• If the software is how your business works differently from competitors — build it
• If the software is just how your business operates (payroll, HR, email) — buy it
• ChiliAuto's dispatch engine is a differentiator. ChiliAuto's email provider is not.
• Your competitive advantage should live in custom software. Your operations should run on SaaS.

Question 2: Does the off-the-shelf solution actually fit your process — or just approximately fit?

• Test the SaaS tool with real data for two weeks before committing
• "We'll adapt our process to fit the software" works for commodities, not differentiators
• If you spend more time fighting the tool than using it, you've answered the question
• We tried four billing tools before building Billwaze. Each required a workaround for the same MSP-specific feature.

Question 3: What is the real cost comparison?

• SaaS costs: monthly fee × months × seats + integration cost + migration cost
• Custom costs: build + maintain + update + the ongoing engineering overhead
• Most companies underestimate the ongoing cost of custom software by 40–60%
• Most companies overestimate the monthly SaaS cost by ignoring the productivity it delivers

The real cost of SaaS is the monthly fee. The real cost of custom software is the monthly fee plus the engineers who maintain it.

When to buy (SaaS)

• The problem is a commodity — solved many times, in a mature market
• Your usage matches the tool's model — you're not fighting defaults constantly
• The vendor's roadmap is moving in a direction that works for you
• You're early-stage and need to move fast — integration, not engineering
• The switching cost is low — you can leave if the tool stops working

When to build (custom)

• The workflow is core to your competitive differentiation — it's how you're different
• No off-the-shelf solution models your domain accurately — you're always working around it
• You've outgrown SaaS tools and the workarounds are costing you more than a build would
• You need integration depth that SaaS APIs can't support
• You operate at a scale where SaaS licensing costs exceed the cost of ownership

The hybrid approach most companies miss

The best architecture for most growing businesses isn't all-SaaS or all-custom. It's a custom core with SaaS periphery. Build the thing that makes you different. Buy the things that keep the lights on.

MoonBeauty is a custom booking and CRM platform. But it sends SMS via Twilio, processes payments via Stripe, and sends emails via SendGrid. The differentiation is the booking engine and the client CRM. The commodity is the message delivery. We bought the commodity and built the differentiator.

The lifecycle question

The right answer changes as the business grows. At €0 ARR, almost nothing should be custom. At €1M ARR, the things that drive that revenue might need to be. At €10M ARR, the tools that were good enough at €1M are probably the ceiling.

The mistake is applying the €10M answer to a €0 business. Build what you need today, with a clear view of where the limits are — so you know when the SaaS ceiling is approaching before you hit it.

Buy until it hurts. Then build exactly what hurts. Don't build what might hurt someday.