Contents
No-code and low-code tools (Zapier, Make, Airtable, Bubble and AI app builders) have changed the game: a small business can automate a task or set up a first tool without writing a line of code. Does that mean giving up on custom development? No. But you need to know when to move from one to the other.
What no-code does very well
- Speed: a first automation or a prototype in a few days.
- Entry cost: a monthly subscription rather than an investment.
- Autonomy: a non-technical person can create and adjust their own workflows.
- Testing: ideal for validating a need before investing further.
To connect two off-the-shelf tools, send a notification or feed a spreadsheet, no-code is often the best answer.
Where it reaches its limits
Problems appear as the need grows:
- Platform dependency: your processes and data live with a vendor, on its terms and at its prices. Leaving often means rebuilding everything.
- Costs that scale with usage: the number of users, operations or records pushes the bill up as your business grows.
- Customization: as soon as the need goes beyond the built-in templates, workarounds start piling up.
- Performance: with large volumes or complex processing, response times degrade.
- Data control: hosting, often outside the European Union, makes GDPR compliance harder.
- Invisible maintenance: a chain of ten automations that nobody understands in detail any more becomes fragile.
What custom development brings
- A tool that fits your processes, with no compromises imposed by a template.
- Ownership of the code and data, and the choice of hosting.
- Predictable costs that do not depend on the number of users.
- Unlimited integrations with your existing tools.
- Controlled performance, even with a lot of data.
The trade-off: a higher upfront investment and a lead time of a few weeks to a few months.
The decision grid
| Your situation | Lean toward no-code | Lean toward custom |
|---|---|---|
| Processes | Standard, covered by existing tools | Specific to your business |
| Urgency | Needed within the week | Project over a few months |
| Users | Few, stable headcount | Many or growing |
| Data | Not very sensitive, low volumes | Sensitive, large or strategic |
| Integrations | Two or three off-the-shelf tools | Internal systems, APIs, complex logic |
| Time horizon | Test or one-off need | Core tool for several years |
The best approach is often hybrid
You do not have to choose once and for all. A three-step approach works well:
- Automate simple tasks quickly with no-code tools, and measure the time saved.
- Spot what gets stuck: fragile workflows, soaring costs, needs the tools do not cover.
- Build the core of your business as custom software, where a tailored tool makes a real difference, and keep no-code for the rest.
My advice: no-code is an excellent starting point and a poor foundation. When a tool becomes central to your business, it deserves to be built for you.