A home office at day's end, a brass lamp above a notebook full of block diagrams connected by arrows

· 6 min read

No-code in 2026: what lasted, what died, and what comes next

The 2026 no-code picture has three observations: the visual promise held, three tools died in a summer, and two platforms started running the same agents we direct. Seven platforms reviewed with prices at source.

The 2026 no-code picture is visible on pricing pages. Seven platforms checked at source September 2, 2026 (Softr, Bubble, Glide, FlutterFlow, Airtable, Notion, Retool) list paid tiers from $19 to $549 monthly and six incompatible credit systems, with none writing a specification before building.

$19–$549 per monthpaid tiers across seven no-code platforms on September 2, 2026
6 credit systemsfor 7 tools, none comparable
23 monthsFirebase Studio's lifespan from launch to announced closure

What lasted

Visual editors do their intended job. Bubble builds complete web and mobile applications with databases, roles, payments, and store submission, managed by one person without a server or desire for one. Softr and Glide put clean portals over existing data in hours with team permissions. FlutterFlow produces a real downloadable Flutter project from $39 monthly. Airtable and Notion began elsewhere but became applications for teams asking no more. For an internal tool three colleagues will use Monday, these remain the shortest route, and custom software never beats their first day.

Three disappearances in one summer

Firebase Studio closed registrations June 22, 2026, with shutdown and data deletion announced for March 22, 2027 (Google migration documentation): twenty-three months. GitHub Spark stopped new users and applications August 4, 2026, maintaining access until August 31 for code export (GitHub changelog). Databutton, renamed Riff, left the field: its domain points to a supply-chain platform. These belonged to description-generated applications rather than classical visual no-code, irrelevant to users: what a closing tool owes you is decided when retrieving belongings, not at announcement.

The bill changed units

Displayed price stopped stating what you pay. Bubble counts workload units measuring application server resources, alerts at 75% and 100%, and permits overage cutoff. Softr, Airtable, and Retool use differently defined AI credits, Retool at $100 per thousand. Glide adjusts credit intensity. FlutterFlow combines Lite credits and AI Generation requests allocated per seat. Notion charges $10 per thousand monthly or $13 annually. Seven tools, six counting methods, one consequence: your bill rises when the product succeeds. Amounts were recorded in dollars except Notion's euro-served page; euro prices for the other six were not verified.

What remains when you stop paying

Settle this before choosing. FlutterFlow leaves a compiling Flutter project with one-way return to the editor. Retool leaves versionable files its documentation declares unexecutable outside Retool. Softr, Airtable, and Notion leave exported data without the application. GlideOS requires a custom-quote plan just to see AI-written code. Bubble leaves nothing and explicitly says migration away 'is not possible'. It has the complete lock, explaining why the wall arrives on an access rule rather than a missing feature, and the shop blocked forty-eight hours without export.

What comes next has begun

Since July 27, 2026, FlutterFlow runs Claude Code and Codex inside its desktop editor with per-message model choice (FlutterFlow changelog). Since April 21, 2026, Glide sells Fling, presented as Claude Code's missing publish button. Two no-code platforms embedded the assistants we direct. They took the engines; the remaining work waits. None of the seven writes a specification before building, writes tests before code, or asks document-by-document agreement. An owner describing needs to Fling or FlutterFlow's agent gets screens, never the product-definition document or accepted-decision record. The unoccupied ground of 2026 concerns method rather than learning an editor.

Assess your case tonight

Take your platform and write two answers: what remains on your hard drive when payment stops, in which form, and what your busiest month costs including credits rather than an average month. If both suit you, stay; migration gains nothing. If either causes hesitation, settle it before renewal, never afterward.

Compare alternatives to Lovable: pricing, code and local workflows

Back to the journal

Take the baton.

Leave your email to try Maestro in the first waves.

The beta is open by invitation on macOS 13 and later. Leave your email for an upcoming wave of access. Windows is in development.

The beta is currently available on macOS 13 or later. Your answer helps us plan other versions.

Your email is only used to let you know when access opens. Nothing else, we promise.