The verdict in three sentences
Stay on SaaS while your need is standard, your users few and your annual cost controlled. Switch to custom when at least two signals stack up: specific process, more than 20 users, strategic dependency, or a SaaS cost above 15,000 EUR/year. Custom isn't a luxury: it's an economic trade-off that becomes rational above a certain volume.
The four tipping criteria
No single criterion justifies a build. Their combination is what counts. A specific process without volume never pays back; high volume on a standard need is fine with a negotiated SaaS.
| Criterion | Stay on SaaS | Move to custom |
|---|---|---|
| Process specificity | Standard | Unique, uncovered |
| Number of users | < 20 | > 20-30 |
| Annual SaaS cost | < 10,000 EUR | > 15,000 EUR |
| Strategic dependency | Low | Core business |
| Integration need | Standard connectors | Proprietary systems |
| Usage horizon | < 3 years | > 4 years |
Warning signals and switch examples
Some signals show SaaS is costing you more than it serves: repeated manual workarounds, systematic Excel exports, paid-for but unused features, or a vendor refusing to evolve. Here are concrete 2026 thresholds.
| Situation | SaaS cost/year | Verdict |
|---|---|---|
| 12 users, standard need | 8,000 EUR | Stay SaaS |
| 30 users, 3 workarounds | 21,000 EUR | Study the build |
| 50 users, unique process | 36,000 EUR | Build |
| 25 users, sensitive data | 18,000 EUR | Build (sovereignty) |
| 40 users, vendor blocked | 28,000 EUR | Build |
The build risk is real: 2-4 month lead time, upfront cost, maintenance need. It's only justified if recurring savings and process gains cover the investment within 3-4 years.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Mini case study
Karim, manager of a 60-person logistics firm in Casablanca, pays a SaaS TMS at 32,000 MAD/month (about 3,000 EUR) for 38 users, i.e. 36,000 EUR/year. His teams work around the tool with three Excel files because the SaaS can't handle his specific routes. A custom TMS at 55,000 EUR + 7,000 EUR/year maintenance would cost 90,000 EUR over 5 years, versus 180,000 EUR of SaaS. With two strong signals (volume + unique process) and a 90,000 EUR saving, the switch is justified; break-even lands in year 2.
FAQ
Is one criterion enough to justify custom? Rarely. You usually need at least two stacked signals, e.g. high volume plus a specific process, for the build to pay off.
At what SaaS cost should I study a build? Above 15,000 EUR/year the study becomes relevant. Above 30,000 EUR/year with a specific process, custom often wins.
How many users to make it worthwhile? In practice, the switch gets interesting between 20 and 30 active users, depending on SaaS price and need specificity.
What are the build risks? Lead time (2-4 months), upfront cost, and dependency on a vendor for maintenance. Limit them with tight scoping and a code-transfer clause.
Can we do a SaaS + custom hybrid? Yes: keep SaaS for standard needs and build only the strategic specific module. Often the best cost/risk ratio.
Let's scope your project. List your current workarounds and annual SaaS cost, and we'll assess whether switching pays off and how fast. 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.
