The verdict in three sentences
No-code (Zapier, Make, n8n) is unbeatable for getting started and for simple, low-volume flows. It becomes expensive beyond roughly 50,000 tasks per month, or as soon as critical scenarios (orders, invoicing, stock) break several times a month. The right strategy is rarely a full rewrite: move the 5 to 10 critical flows to code and keep the rest in no-code.
Why your scenarios break
With 40 scenarios built ad hoc by 3 or 4 people, the causes are always the same:
- a third-party API changes format or expires an access token;
- a renamed field in the CRM or form breaks the mapping;
- rate limits hit during a spike (import, campaign);
- chained scenarios with no error handling: one step fails, the next ones run empty;
- nobody knows who built what, or why.
| Symptom | Typical frequency (40 scenarios) | Estimated cost per incident |
|---|---|---|
| Expired token, reconnection | 2 to 4 per month | 30 min, low |
| Broken mapping after a field change | 1 to 3 per month | 1 to 3 h + data to fix |
| Order or invoice not created | 1 to 2 per month | 200 to 1,500 EUR (delay, credit note, unhappy customer) |
| Duplicates created on automatic retry | 1 per month | 2 to 6 h of cleanup |
| Task quota exceeded | During spikes | Plan upgrade or flow stopped |
Comparing costs: no-code or code
| Criterion (2026 estimate) | Zapier | Make | Custom development |
|---|---|---|---|
| Subscription | 20 to 700 USD per month by tasks | 10 to 300 USD per month by operations | Hosting 20 to 150 EUR per month |
| Initial cost | Internal time | Internal time | 5,000 to 30,000 EUR |
| Maintenance | Internal time, often invisible | Internal time | 15 to 20% per year of build |
| Error handling | Basic retries | Error routes | Tailored replays, log, alerts |
| Automated tests | No | No | Yes |
| Versioning and review | Limited | Limited | Git, code review |
| Cost at 100,000 tasks per month | High plan, several hundred USD | 100 to 300 USD | Nearly the same as at 10,000 tasks |
The subscription is rarely the problem. The real cost is internal maintenance time and the cost of incidents. An ops team spending 8 to 15 hours a month fixing scenarios already spends 400 to 900 EUR a month in time.
Signals that justify the switch
- Volume: more than 50,000 tasks per month, or a Zapier plan above 300 USD per month.
- Criticality: the flow touches orders, invoicing, payroll or stock.
- Complexity: more than 15 steps, loops, nested conditions.
- Incidents: more than 2 failures per month on the same flow.
- Compliance: personal or financial data that must stay in the EU with an audit log.
If at least two signals apply to a flow, moving it to code (a serverless function, a small Node.js or Python service with a queue) usually pays back in under 12 months.
Mini case study
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Illustrative scenario: Julie, head of operations at a distribution SME in Nantes (45 staff), runs 40 Zapier scenarios for 70,000 tasks per month. Subscription: 450 USD per month, about 415 EUR. Internal maintenance: 12 hours per month at 45 EUR, or 540 EUR. Incidents on orders and invoices: 2 per month at 600 EUR, or 1,200 EUR. Real monthly cost: about 2,155 EUR. Chosen plan: move 8 critical flows to code for 18,000 EUR, hosting 80 EUR per month, maintenance 3,000 EUR per year; keep 32 simple flows on a 100 USD Zapier plan. Estimated new monthly cost: 92 + 80 + 250 + 3 h maintenance (135 EUR) + 0.3 incident (180 EUR) = about 740 EUR. Saving: 1,415 EUR per month, payback in 13 months, with much better reliability.
FAQ
Should we drop Zapier or Make entirely?
No. Simple, low-stakes flows (Slack notifications, adding to an email list) stay cheaper in no-code. In practice, 70 to 80% of scenarios can stay there.
Is self-hosted n8n an alternative?
Yes, it removes per-task pricing and allows EU hosting. Count 20 to 60 EUR per month for a server, but technical maintenance becomes yours.
How much does migrating one flow cost?
As a 2026 estimate, 1,500 to 4,000 EUR per critical flow with error handling, tests and documentation, less if several flows share the same connectors.
What about dependency on third-party APIs?
It remains, but code allows contract tests, alerts within 5 minutes and automatic replays. Mean time to recovery often drops from several hours to under one hour.
Where do we start?
With an inventory of your scenarios: volume, criticality, failures over 3 months. This audit takes 2 to 4 days and ranks the flows to migrate first.
Let's scope your project. Send us your list of scenarios with volumes and recent failures: we will identify the flows to migrate, the budget and the timeline. 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.
