No-code development: what you can build, where the limits are and when to switch
No-code development lets people build applications, websites and automations with visual tools instead of program code. A form, a database, a few screens and a workflow can be live in days, built by the people who know the process. No-code development is a real option for many business needs, and a trap for some. This guide explains no-code, low-code and custom code, what each does well, the limits that show up after launch, the data protection and exit questions to settle first, and how to move to custom software when the time comes.
Legal texts referred to were read on 12 November 2026: Regulation (EU) 2016/679 (GDPR) and Regulation (EU) 2023/2854 (Data Act). Platforms are named as examples only; we are not affiliated with any of them.
No-code, low-code and code: the definitions
| Approach | Who builds | How | Where it runs |
|---|---|---|---|
| No-code | Business users, citizen builders | Visual editors, templates, drag and drop, workflow builders | On the platform only |
| Low-code | Technical staff and IT teams | Visual building blocks plus custom code where needed | On the platform, sometimes self-hosted |
| Custom code | Software engineers | Program code in standard languages and frameworks | Anywhere: your cloud account, your servers |
The boundaries blur: many no-code tools allow small scripts, and low-code platforms market themselves to business users. The practical question is not the label but where the logic lives and who can change it.
The main no-code platforms by type
- Web and mobile app builders: Bubble, Glide, Softr, Adalo, FlutterFlow. Screens, user accounts and a database without code.
- Website builders: Webflow, Wix, Squarespace, Framer. Marketing sites and simple stores.
- Databases and internal tools: Airtable, Notion, Microsoft Power Apps, Retool (closer to low-code). Spreadsheets that grew into systems.
- Automation: Zapier, Make, n8n, Microsoft Power Automate. Moving data between SaaS products when something happens.
- Forms and portals: Typeform, Jotform, Tally and portal builders on top of a database.
What no-code development does well
- Speed to a first version. An idea can be tested with real users in a week rather than a quarter.
- The process owner builds it. The person who knows the work designs the screens, so fewer things are lost in translation.
- Internal tools for small teams. Request forms, approval flows, simple registers, dashboards over existing data.
- Glue between SaaS products. New lead in the CRM creates a task, a signed contract creates a folder and an invoice draft.
- Prototypes as a specification. A clickable, working version is the best requirements document there is.
The limits of no-code
The limits rarely show on day one. They appear when the application succeeds and more people depend on it:
- Complex rules. Pricing logic, scheduling with many constraints, calculations with exceptions. Visual editors become hard to read and harder to test.
- Volume and performance. Platforms have limits on records, workflow runs and requests, and pages slow down as data grows.
- Testing and change control. Few platforms offer automated tests, separate test environments or a reviewable history of changes. A change goes straight to production.
- Integrations. Ready connectors cover popular SaaS products; an ERP with its own interface, a machine, or a national registry often needs code.
- Mobile and offline. App builders produce usable apps, but native features, offline work and store requirements are where they struggle.
- Accessibility. Generated pages do not always meet WCAG, and from 28 June 2025 the European Accessibility Act applies to many consumer-facing digital services. See our European Accessibility Act guide.
- Cost at scale. Per-user and per-run pricing that is small for ten users can exceed the cost of running custom software for a thousand.
- Dependency. The application exists only on the platform. If the vendor changes pricing, removes a feature or closes, the application goes with it.
Personal data and hosting
A no-code platform that stores customer or employee data is a processor under the GDPR, and in many cases its subprocessors include hosting and email providers in several countries. Article 28 GDPR requires a contract with the processor that, among other things, obliges it to process only on your documented instructions, to apply the security measures of Article 32, and, at your choice, to delete or return all personal data at the end of the service. Check before building:
- Is there a data processing agreement, and where is the data stored and processed?
- Is there a list of subprocessors, and do any transfers outside the EU rely on a valid basis?
- Can you set roles and permissions finely enough, and is access logged?
- Can you delete individual records and whole datasets when retention periods end?
Our data processing agreement guide lists what the contract should contain, and the sovereign cloud guide covers the questions of jurisdiction.
Exit rights: what the Data Act changes
Many no-code platforms are data processing services in the sense of the Data Act, whose switching rules have applied since 12 September 2025. Article 23 requires providers to remove obstacles that prevent customers from switching to another provider or porting their exportable data and digital assets, and Article 25 requires the contract to allow switching or porting within a maximum transitional period of 30 calendar days. Under Article 29, switching charges are prohibited from 12 January 2027.
That helps with data. It does not turn a no-code application into portable code: the screens and workflows are built in the platform's own format, which no other platform reads. Plan the exit at the start: keep the data model clean and documented, export regularly, and write down the business rules the platform implements so they can be rebuilt.
No-code or custom software: a decision grid
| Signal | Points to no-code | Points to custom code |
|---|---|---|
| Users | One team, up to a few dozen people | Customers, many teams, hundreds or thousands of users |
| Rules | Simple forms and approvals | Calculations, schedules, pricing, exceptions |
| Lifetime | A test or a tool for a year or two | A core system for many years |
| Integrations | Popular SaaS with ready connectors | ERP, machines, registries, custom interfaces |
| Compliance | Low-risk data | Health, finance, audit trails, regulated products |
| Differentiation | A standard process | The process is how you compete |
Moving from no-code to custom code
A no-code application that has outgrown its platform is the best possible starting point for custom software: the users exist, the screens have been tested in daily work and the data model is known. The move usually runs in four steps. First, document what the application does today, including the workflows nobody remembers building. Second, design the new system around the real process rather than the platform's limits; our process mapping guide helps. Third, build it and migrate the data with repeatable scripts and reconciliation; see data migration. Fourth, run both side by side for a short period and switch when the numbers match.
Where we stand
We recommend no-code when it fits, and we say so: a small internal tool on a no-code platform can be the right answer. When a system needs custom rules, integrations, scale or a long life, we build it as internal tools or a web app in code, hosted in the EU, with the source code in your company's repository from the first day. Either way, your data and your process stay yours.
Questions people ask
What is no-code development?
No-code development means building software with visual tools instead of writing program code: you drag components onto screens, define data tables and connect steps in a workflow editor. The platform runs the result for you. It suits prototypes, internal tools, simple apps and automations between existing systems.
What is the difference between no-code and low-code?
No-code platforms are designed for people who do not write code at all; everything is configured visually. Low-code platforms also use visual building blocks but let technical staff add code where the visual tools stop, which extends what can be built and also requires people who can maintain that code.
Is no-code good enough for a real business application?
For some, yes: internal tools for a team, simple customer portals, booking forms, approval workflows, early versions of a product. It becomes difficult when the application needs complex business rules, high volumes, strict performance, deep integrations, offline use, or must meet security and audit requirements the platform cannot show.
Can we move a no-code app to custom code later?
The data can usually be exported; the application itself cannot, because the logic lives inside the platform. Moving means rebuilding the application in code, using the no-code version as a living specification. That is often a good path: the no-code version proves what users need, and the custom version is built once, properly.
Who owns a no-code application?
Your company owns its data and has rights under the platform's terms, but the application runs on the platform and depends on its continued existence, pricing and rules. With custom code built for you, the code is in your repository and can run anywhere. That is the main trade-off to weigh.
Sources
- Regulation (EU) 2016/679 (GDPR), Articles 28 and 32 (read 12 November 2026)
- Regulation (EU) 2023/2854 (Data Act), Articles 23, 25, 29 and 50 (read 12 November 2026)
- Directive (EU) 2019/882 (European Accessibility Act), Article 31 (read 12 November 2026)