Skip to content
Getting Digital

Programming and Software Development

No-code and low-code development

No-code and low-code tools let people build working software by configuring it rather than writing it: visual app builders such as Bubble, site builders such as Webflow, automation platforms such as Zapier and Make, and spreadsheet-like databases such as Airtable. They are a real route to shipping a product or automating a process, with limits that surface once logic, data volume or custom integrations outgrow what the platform exposes.

Why this topic exists: Building software without writing most of it (Bubble, Webflow, Zapier, Make, Airtable) is a recognised route into shipping products; Microsoft's Power Platform is the office-suite version of the same idea.

No-code and low-code platforms let people assemble working software from configured parts instead of writing it line by line. Underneath the visual editors the familiar pieces are still there: a data model, rules that act on it, screens that display it, and connections to other systems. The platform supplies the code for each piece, and the builder decides how they fit together. Low-code tools add an escape hatch where a developer can write a formula, a script or a custom component when configuration runs out.

Four families of tool

  • Visual app builders such as Bubble, where pages, workflows and a database are defined in one editor and deployed together.
  • Site builders such as Webflow, which start from a web template or a blank canvas and produce production-ready pages and content collections.
  • Automation platforms such as Zapier and Make, which chain triggers and actions across services, usually by calling each service's REST API on your behalf.
  • Spreadsheet-style databases such as Airtable, which give a team relational tables behind a familiar grid, with forms and views on top.

Where the ceiling appears

The limits arrive in a predictable order. Logic first: a workflow with many branches and exceptions becomes harder to follow in boxes and arrows than it would be in code. Then volume, when records or automated runs grow past what the plan or the platform handles comfortably, and pricing tied to usage grows with them. Then integration, when the system you need exposes nothing the platform can connect to. Underneath all of these sits portability: logic built in one vendor's editor usually cannot be exported and run anywhere else.

We have not found a vendor-neutral body of knowledge or exam for this work; the platforms train their own users. The useful frame is therefore the one programming already has: gather requirements, design the data, test changes before they reach real users, keep track of versions and control who can edit what. Microsoft's Power Apps brings the same approach into organisations that run Microsoft 365, and flows built by office staff for their own routines are covered under workplace automation.

The data model still decides

Visual tools make it easy to start with screens. Start with the tables and how they relate instead; a poorly shaped data model limits a no-code app exactly as it limits a coded one, and it is the part hardest to change later.

The beginner's error is building a whole business process on a single platform with no plan for leaving it. Before committing, check how data can be exported, what happens if the account lapses, and who else can maintain the build. Those who later want more control move on to programming fundamentals, databases and SQL and web development; for sites where the design matters more than the logic, see web design.

Next to this topic

Concepts to know

Glossary entries with the reason each one matters here.

  • REST API

    No-code tools glue APIs together.

  • Template

    Templates are no-code's raw material.

Tools of the trade

Frequently asked

Is no-code real software development?
It produces real software, and the hard parts are the same: understanding the problem, modelling data, handling errors and testing. What changes is the medium. Builders who think like developers get far more out of these platforms than those who treat them as drawing tools.
Zapier or Make?
Both connect services through triggers and actions. They differ in how they present multi-step logic and how they count usage, so the sound approach is to build one real process in each on a free tier and compare how easy it was to read and debug afterwards.
When should a no-code project move to code?
When the workarounds cost more than the platform saves: logic that cannot be expressed cleanly, performance problems at your data volume, integrations the platform cannot reach, or running costs that grow faster than the business. Moving is easier if the data model was designed carefully from the start.

Courses in the directory

86 courses are filed here; the top 6 by our ranking, details and the provider link on each course page.

Browse the directory shelf

Last reviewed 26 September 2026 · Getting Digital