The verdict in three sentences
An off-the-shelf SaaS costs little upfront but its subscription grows with your users; custom software requires an initial investment but no recurring license. In 2026, break-even usually sits between 24 and 42 months depending on user count and how specific your process is. Decide on total cost of ownership and vendor lock-in risk, not on the headline first-month price.
The real cost over 3 and 5 years
The SaaS trap is its per-user cost: painless at 5 people, heavy at 40. Here is a comparison for a 40-user team at a median 40 EUR/user/month.
| Horizon | Off-the-shelf SaaS (40 users) | Custom | Gap |
|---|---|---|---|
| Upfront investment | 0 - 5,000 EUR | 45,000 - 150,000 EUR | +custom |
| Recurring annual cost | 19,200 EUR | ~9,000 EUR (run+maint.) | -custom |
| 3-year TCO | 57,600 - 62,600 EUR | 72,000 - 177,000 EUR | varies |
| 5-year TCO | 96,000 - 101,000 EUR | 90,000 - 195,000 EUR | ~even |
| Recurring license | Yes, tied to users | No | -custom |
Under 40 users SaaS is often cheaper; above that, and especially above 60-80 users, custom catches up and rarely exceeds SaaS on cost.
The criteria beyond price
Cost is only one input. These qualitative criteria often swing the decision as much as TCO.
| Criterion | Favours SaaS | Favours custom |
|---|---|---|
| Process specificity | Standard process | Unique process, competitive edge |
| Vendor lock-in | Acceptable | To avoid (strategic data) |
| Evolvability | Vendor roadmap suffices | Need to steer the roadmap |
| Integrations | Standard API available | Deep, ERP, custom integrations |
| Timeline | Immediate | 3-6 months of build |
| User count | < 40 | > 40-60 |
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Mini case study
Nathalie is CFO of a 55-person distribution SMB, 40 of whom use the management tool. A vertical SaaS costs 42 EUR/user/month, i.e. 20,160 EUR/year and 100,800 EUR over 5 years. Custom development is quoted at 95,000 EUR, run and maintenance included at 9,000 EUR/year, i.e. 131,000 EUR over 5 years gross, but custom removes 4 side tools (12,000 EUR/year) and automates re-keying (8 h/week, ~14,000 EUR/year). Factoring in those gains, custom drops below SaaS by month 32 and the gap widens after. Verdict: custom justified here by process specificity and user count.
FAQ
From how many users does custom become worth it? Often above 40-60 users, because SaaS cost grows linearly with seats while custom has a near-fixed run cost.
Is vendor lock-in a real risk? Yes: price hikes, a discontinued feature, difficulty exporting your data. For a strategic process, that risk often justifies custom.
Can the two approaches be mixed? Absolutely. Best practice is to buy SaaS for standard functions (payroll, email) and build custom where your competitive advantage lives.
How to avoid underestimating SaaS TCO? Project the cost over 5 years with your headcount growth, add paid modules and integrations: the real price often doubles the headline.
Is an Odoo-type SaaS a good compromise? For semi-standard needs, yes, with configuration. As soon as specifics pile up, customisation cost approaches that of true custom.
Let's scope your project. Give us your user count, your key processes and your current tools, and we'll build a costed 3- and 5-year TCO comparison. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.

