Contents5
An operations manager builds, in one afternoon, the inspection form her team had been waiting months for. A salesperson connects his spreadsheet to the CRM without opening a ticket. Low-code tools (you assemble blocks and write a little logic) and no-code tools (you write nothing) made this possible, and that is good news. The question is not whether they replace developers, but how far to push them, and when code takes over.
In short
Low-code tools excel at internal tools, forms, tracking dashboards, automations and prototypes.
Their price is often counted per user per month: it grows with the team.
An app built on a platform lives on that platform; the data comes out more easily than the logic.
They do not replace developers: they shift their work towards integration, security and what must hold at scale.
The hybrid route, low-code around a coded core, is often the most economical.
What tools without developers do very well
Their strength comes down to one thing: the person who knows the business builds it themselves, without translating the need to someone else. The cases where they are at their best:
- Internal tools: an inspection form, a request tracker, a dashboard on existing data.
- Automations: connecting software together, triggering a reminder, copying an order into the CRM.
- Prototypes: checking that a journey makes sense before investing, just like a clickable prototype.
- Needs that change fast: the team adjusts a field or a step itself, without waiting for a release.
From the field
We use a low-code tool ourselves
We run n8n workflows in production, an automation tool where you assemble connected blocks: the daily import of new properties from a real estate portal, or an expense-entry agent on Telegram. For connecting software together, it is often faster and more readable than a program written from scratch. The choice of tool is compared in n8n, Make or Zapier.
Where they reach their limits
A price that grows with the team
The major platforms are often paid per user per month, which is reasonable for a team of ten and becomes a significant line for a company of three hundred.
per user per month, billed yearly, for Microsoft’s Power Apps Premium, according to its pricing page
per user per month for AppSheet’s Starter and Core plans, Core being included in most paid Google Workspace plans, according to its pricing page
These prices, noted in October 2026, do not say a tool is expensive: a subscription that avoids a development project can be an excellent deal. They say the calculation must be made over three years, with the real number of users, as in custom or off-the-shelf.
The logic stays on the platform
You can easily take your data out of a no-code platform; the app’s logic, much less so. Bubble, for example, says so plainly in its FAQ: apps there are built visually, so there is no traditional codebase to export. Bubble states that the app and its data belong to you, offers a raw export of the logic as JSON to help migrate, data export as CSV, and commits to releasing its source code under an open-source licence if it ever shut down. That is honest and reassuring, but leaving the platform remains a rebuild.
Complex rules and security
Nested business rules, calculations that must be exact, fine-grained access rights between several types of users: this is where visual logic becomes hard to read and test. And security cannot be fully delegated to the tool: data access rules still have to be written and checked, as app security explains. Tools that generate a whole app from a description, another face of “no developers”, have had flaws of this kind, recounted in vibe coding.
Public apps and the stores
An app for your customers, published on the App Store and Google Play, with its notifications, its offline mode, its subscriptions and its listing under your own developer account, needs a level of control that no-code platforms offer unevenly. The more exposed the app, the more reasonable code becomes.
Low-code does not replace developers
No-code tools have widened the circle of people who build, and that is progress. They have not made development unnecessary: they have shifted part of developers’ work. When a business team builds its own tools, someone is still needed to connect the platform to the company’s software, check who accesses which data, monitor what runs, and rewrite in code what has outgrown the tool. The setup that works is a business team that builds fast, and developers who make those builds solid.
Low-code or no-code tool
Built by the business team, in days rather than weeks
Per-user subscription, logic hosted by the platform
Ideal for internal use, automations and prototypes
Coded app
Built by developers, from a requirements document
Development cost, then operations in your name
Suited to public apps, complex rules and the long term
The hybrid route
The two combine well, and it is often the most economical choice.
- Validate in low-code.
A first tool built quickly by the team shows what is really useful.
- Code the core.
What must be exact, secure or public-facing moves into a coded app, with a real database.
- Keep low-code around it.
Automations, dashboards and small internal tools stay in the team’s hands, connected to the core through its programming interface.
This split goes into the app’s requirements document, and weighs on the price of an app as well as its maintenance budget. The coded core often rests on a back end like those compared in Supabase, Firebase or your own server. The other decisions on the journey each have their article: PWA or native app, the field team app, notifications, the client portal, AI in an app and taking over an abandoned app.
Our app development starts with written scoping, which also says what can stay in low-code.
Frequently asked questions
What is the difference between low-code and no-code?
A no-code tool is configured entirely through screens, without writing logic. A low-code tool lets you assemble blocks visually, but also write some code for cases the blocks do not cover. In practice the line is blurred, and many tools do both.
Will low-code replace developers?
No. It lets business teams build useful tools themselves, which frees developers from small requests. But developers are still needed to connect those tools to the company’s software, secure the data, and build what must hold at scale or be released to the public.
Can you move from a no-code tool to a coded app?
Yes, but it is a rebuild: data can generally be exported, logic much less so. The work done on the tool is not lost for all that: it has validated the journeys and rules, which become the requirements for the coded version.
How much does a low-code tool cost a company?
Most often, a subscription per user per month: in October 2026, Power Apps Premium lists US$20 per user per month billed yearly, and AppSheet US$5 or US$10 depending on the plan. The calculation is made over three years with the real number of users, and compared with the cost of a development project and its operation.
Sources
- Microsoft, Power Apps pricing (new tab)microsoft.com
- Google, AppSheet pricing (new tab)about.appsheet.com
- Bubble, FAQ (new tab)bubble.io
Jim · founder of Brain Plus AI
Also in web and mobile apps
Stepping back22 min
Vibe coding: AI codes fast, the product stays human
AI has removed real technical barriers, but it has not removed what makes a product. What is true and what is false in "anyone can do anything with computers", with examples in both directions.
Choose and price6 min
Business app: custom-built or off-the-shelf software?
Off-the-shelf software costs less as long as your way of working fits its boxes. Here are the signs that tip the balance towards custom, and the middle path that is often forgotten.
Choose and price6 min
How much an app costs: what makes the price vary
A price range says nothing about your project: what makes an app quote vary is a handful of specific decisions, plus a monthly running cost to be priced separately. Here they are.