A no-code database built on a real SQL engine

AxisIQ is a no-code database: custom tables with typed fields, validation and relationships on a real SQL engine, plus forms, an API and SQL reporting.

A no-code database lets you design tables, fields, rules and relationships through an interface instead of writing code — and the ones worth using store your data in a real database rather than a decorated spreadsheet. AxisIQ is that second kind: every object type you create through the UI is a real table in a real SQL database, with typed columns, validation enforced on every write, relationships that resolve, and SQL you can actually run against it.

That last part is the dividing line in this category. Most online database builders give you a grid that looks like a table and behaves like a spreadsheet: any cell can hold anything, rules are suggestions, and querying means learning a proprietary formula language. This page explains what you get when the thing underneath is a database engine — and, honestly, what you give up.

Types are tables, fields are typed columns

Creating an object type in AxisIQ — properties, students, repairs, warranty-claims, whatever your business counts — creates a table. Each field you add is a typed column: text, long text, number, decimal, date, timestamp, yes/no, single choice, multiple choice, money with its currency attached, email address, link, or a reference to another record. The type is not a display hint; it is what the column will accept.

Rules travel with the field. Mark a field required and a write without it is refused — on later edits, not just the first save. Mark it unique and the second record repeating a value is rejected, not flagged for someone to notice. Choice fields refuse options that are not on the list; number fields enforce ranges; formats are checked at the moment of writing. A field can also carry an auto-generation pattern like JOB-{seq:4:yearly}, so every new record gets the next number — reset yearly, monthly or never — while imported historical values keep their own.

A new workspace comes with types most businesses need already defined — customers, orders, products, stock, invoices — and your custom types are exactly as first-class: same lists, same search, same reports, same permissions. Operations covers the built-ins.

Relationships that hold

A reference field points at a record — one or many — not at a text label that happens to match. An order line references a product; a repair references a customer; rename the customer once and everything pointing at it still resolves. Because relationships are real, joins are real too: a report can join repairs to customers to invoices with ordinary SQL rather than a chain of lookup formulas.

One honest note on enforcement: AxisIQ checks references when they are written, and the data quality report re-checks stored records later — a reference whose target was deleted, a value a since-tightened rule would now refuse — so rule changes do not silently strand old rows.

Getting data in

• Forms. A form publishes selected fields as a hosted page or an iframe on your own site, and every submission becomes a validated record. Spam is handled with a honeypot and rate limiting, and a public form can only create — never read or edit — so it is safe to face the internet. The intake forms guide is a worked example.
• CSV. Import up to 500 rows per file, mapped onto your fields and validated before anything is written — a bad column tells you what is wrong instead of half-importing. Export up to 10,000 rows out. The migration guide covers moving an existing dataset.
• API. Every type is reachable programmatically. You create a service account, issue it an API key, and grant it exactly the permissions it needs — a key is a principal in the same permission system as your people, scoped to your workspace, never an all-access token. Revoke it and it is gone.
• By hand. The records UI — lists, search, filters, editing — exists for every type the moment you define it, with zero page-building.

Getting answers out

Because a type is a table, reporting is SQL — real SELECT, JOIN, GROUP BY over live records, read-only, with results in seconds and no export or warehouse in between. If you do not write SQL, describe the question in plain English and review the generated SQL before it runs; the query is shown, not hidden. Saved queries become reports and dashboards — tables, charts, metric tiles — and the reports guide goes deeper. Field permissions follow you into SQL: a column you are denied comes back masked, not leaked.

Logic without a server

When clicking is not enough, two escape hatches sit inside the permission system rather than beside it:

• Flows — visual automation triggered by record events, schedules or a button: conditions, record writes, loops over lists, outbound HTTP calls. A flow runs under its author's permissions, so it can never do what its author could not.
• Functions — your own JavaScript, run in isolation on the server, reading and writing records under the permissions of whoever invoked it. A function can also be exposed as a public HTTP endpoint, which is how a no-code database quietly becomes a small backend.

Permissions down to the field

Grants are written per action, per type and per field: one team edits repairs but only reads customers; the cost column is invisible to everyone outside finance; an API key can create records in one type and touch nothing else. The same grants bind people, keys, automations and the AI assistant alike — there is no side door where a tool sees more than its user. The permissions overview and team setup guide cover the model.

Against Airtable and Notion

Both deserve a straight answer. Airtable is faster from idea to shared grid, friendlier to people who have never thought about a schema, and backed by a far larger template and integration ecosystem — for project trackers and editorial calendars it is the better tool, full stop. Notion is the better place to write, and its databases are pages with properties, which is exactly right for thinking. The full arguments are in AxisIQ vs Airtable and AxisIQ vs Notion.

The structural difference is enforcement. In a flexible base, a rule is a convention the team agrees to keep; here, a rule is checked on every write and a value that breaks it is refused. If your database is a convenience, conventions are fine. If your database is becoming the business — the thing that decides what ships and who gets charged — refusal is the feature, and that is the case where AxisIQ is the right pick. One step earlier in that story is moving off spreadsheets.

What this is not

Three limits, stated plainly. There is no drag-and-drop interface designer. The records UI is generated from your schema — define a type and the list, search, filters and editing screens exist — but you do not design custom layouts, and if pixel-level control over an internal tool's screens is the requirement, an app-builder product serves you better. This is not a platform for building arbitrary consumer apps. Public surfaces exist — forms for collecting data, HTTP functions, hosted content — but you are not building a marketplace or a social app here. CSV import runs 500 rows per file, so a very large first migration arrives in batches.

What you get in exchange is the part app builders skimp on: a real database underneath, validation that cannot be clicked past, SQL, an audit trail on every change, and permissions that reach individual fields.

Where it fits

The same builder is what makes the rest of AxisIQ fit odd businesses: the CRM is customer types plus forms plus flows, inventory and order management are built-in types you extend, invoicing adds money that balances, and together they make up the small-business ERP. Start with the one table that hurts most; the platform overview shows what the rest connects to.

FAQ

What is a no-code database?

A no-code database lets you create tables, typed fields, validation rules and relationships through an interface instead of code. In AxisIQ each type you define is a real table in a SQL database, so the structure you click together is enforced by the engine, not by convention.

Is AxisIQ a real database or a spreadsheet-style grid?

A real database. Fields are typed columns, required and unique are enforced on every write, references point at records rather than matching text, and you can query it with ordinary read-only SQL. The grid you see is a view of the table, not the storage itself.

Can I query a no-code database with SQL?

In AxisIQ, yes — read-only SELECT with joins and aggregates over live records, powering saved reports and dashboards. If you do not write SQL, you can ask in plain English and review the generated query before it runs.

Does AxisIQ have an API for programmatic access?

Yes. You create a service account, issue it an API key, and grant it precise permissions — down to which types and fields it can touch. Keys live in the same permission system as your team, are scoped to your workspace, and can be revoked at any time.

How is AxisIQ different from Airtable as a database builder?

Airtable is a flexible base where rules are conventions; AxisIQ enforces them — typed columns, required and unique checked on every write, per-field permissions, SQL querying and an append-only audit trail. Airtable is faster for light trackers; AxisIQ is built for data a business depends on.

Can I build a customer-facing app on AxisIQ?

Partly. Public forms collect data into your tables, functions can serve HTTP endpoints, and published content is deliverable — but there is no interface designer for building arbitrary consumer apps. AxisIQ is a system of record with public edges, not an app-builder.