The Build vs Buy Question Nobody Answers Honestly
Every business considering software investment faces the same question: build something custom, or buy an off-the-shelf product? The typical answer you'll find online is "it depends" — which is true but unhelpful. In this post, we're going to be specific with real numbers drawn from our client engagements over the past five years.
The short answer: custom software almost always costs more in year one. But the break-even point is typically between year two and year three, after which custom software becomes dramatically cheaper — and more strategically valuable.
Understanding the True Cost of SaaS
SaaS pricing is designed to feel affordable. $99/month sounds cheap until you multiply it out and add the hidden costs that rarely appear in the headline number.
Per-seat pricing compounds quickly. A CRM at $100/seat/month for a 50-person sales team is $60,000/year. Add the 20 people in customer success, marketing, and operations who also need access and you're at $84,000/year. In three years: $252,000.
You'll need multiple tools. That CRM doesn't do project management. The project management tool doesn't do invoicing. The invoicing tool doesn't integrate with your warehouse system without a middleware subscription. Every tool you add increases the monthly spend and the integration complexity.
Annual price increases are contractual. Most SaaS contracts allow 5–10% annual price increases. On a $100,000/year spend, that's $5,000–$10,000 more per year, every year, without any additional value delivered to you.
Implementation and migration costs. Enterprise SaaS implementations are expensive. Salesforce implementations typically cost 1–3x the first year's licence fee in consulting costs. You also pay these costs again when you eventually switch platforms.
Feature limitations force workarounds. Every time the software doesn't quite fit your process, your team creates a workaround — a spreadsheet, a manual step, an extra email loop. These workarounds accumulate into significant hidden labour costs.
A Real Comparison: CRM
Let's run a specific comparison for a 60-person company choosing a CRM.
SaaS Option (Salesforce Professional):
- Year 1: $75,000 (licence) + $60,000 (implementation) = $135,000
- Year 2–5: $75,000/year (+ ~8% annual increase) = $324,000
- 5-year total: ~$459,000
Custom CRM (built by us):
- Year 1: $85,000 (build cost) + $12,000 (hosting) = $97,000
- Year 2–5: $12,000/year (hosting) + $15,000/year (maintenance retainer) = $108,000
- 5-year total: ~$205,000
The 5-year saving is approximately $254,000. And the custom CRM does exactly what the business needs, without the limitations of Salesforce's data model.
When Off-the-Shelf Wins
Custom software is not always the right answer. Off-the-shelf wins decisively when:
- Your process is standard. If your workflow matches the tool's default workflow, there's little to gain from custom.
- You're under 10 people. The fixed costs of custom development don't amortise well at small scale.
- Speed to market is paramount. If you need something working in two weeks, SaaS is the only option.
- The domain is complex and regulated. Payroll, accounting, and HR compliance tools embed years of regulatory expertise. Reproducing that in custom software is expensive and risky.
The Hidden Strategic Value of Custom Software
The TCO comparison above ignores an important dimension: strategic differentiation. Custom software can become a genuine competitive advantage in ways SaaS cannot.
When your software is built around your unique process, it encodes institutional knowledge that competitors cannot simply replicate by subscribing to the same tool. It can also create network effects — if your platform connects suppliers, partners, or customers in a proprietary way, switching costs for all parties increase over time.
Several of our most successful clients have found their custom software becoming a product in its own right — something they could white-label and sell to their industry peers, creating an entirely new revenue stream.
How to Make the Decision
Use this framework: start with your five-year TCO calculation. If custom is cheaper over five years AND your process is sufficiently differentiated to benefit from custom, build. If the cost advantage is marginal or your needs are standard, buy. If you're somewhere in the middle, consider a hybrid: buy a platform with a strong API (like HubSpot or Airtable) and build custom integrations and automations on top rather than a full replacement.