Ultorg puts a spreadsheet-style interface on relational databases, no SQL required

Ultorg, a desktop application featured on Hacker News's front page on November 10, markets itself as the user interface that relational databases have been missing. Instead of writing SQL or commissioning a custom CRUD app, users work in a spreadsheet-like grid where they can query and edit data spread across multiple related tables.

A spreadsheet front end for database work

According to the company's website, the tabular view on screen represents a live database query rather than a static copy of the data. Ultorg generates SQL behind the scenes — executed either remotely on the database or locally on your machine — and the query can be refreshed or modified without risking changes to the underlying raw data. The site claims that most SELECT queries expressible in core SQL have a visual equivalent in Ultorg, including arbitrary combinations of joins, filters, calculations, and subqueries.

The app runs on macOS, Windows, and Linux, with a 14-day free trial on offer.

Wide connector coverage

Ultorg connects directly to PostgreSQL, Microsoft SQL Server, MySQL and MariaDB, DuckDB and Parquet, Google BigQuery, Microsoft Access files, Excel workbooks, Google Sheets, CSV and TSV files, and other JDBC-based sources. Beta support exists for Oracle, Snowflake, SQLite, Athena, and Exasol, per the site. That breadth means a single view can join a production database table with a local spreadsheet or CSV file.

Visual joins and one-to-many layouts

The tool's headline capability is how it renders joins. For SQL-based sources, Ultorg reads existing foreign key relationships and suggests relevant joins; users can also join on arbitrary fields of their choosing.

One-to-many relationships — where each row in one table can match several rows in another — get particular attention. Ultorg draws them with what the site describes as a purpose-built system of visual data layouts, with sorting, filters, and formulas applicable at any level. Joins can be chained or placed side by side without dropping or duplicating rows, and the company claims to be alone among visual database tools in handling one-to-many relationships generically.

The same layouts apply inside queries. When grouping fields to compute sums or averages, or nesting one query within another, users can see the individual values feeding a group-wise total and adjust filters and calculations while auditing the aggregated results visually.

Automatic layout, manual editing

For datasets with hundreds of fields spread across dozens of tables, Ultorg analyses the data on the fly and derives defaults for label placement, column widths, text field dimensions, decimal precision, and pagination. Its typographical engine renders records and relationships as nested tables, forms, and crosstabs — layouts the company compares to hand-designed screens in bespoke database applications, but generated instantly for any data.

Editing works the way spreadsheet users would expect: cell-level edits, dropdown value selection, row insertion, and fill-down, applied to relational databases, Excel workbooks, or Google Sheets. Changes are reviewed before being committed, and multiple users can edit concurrently. The company positions this as an alternative to custom CRUD interfaces for internal business operations.

The site also carries a testimonial from Tudor Achim, co-founder and CEO of AI research firm Harmonic, who calls Ultorg the best tool available for visualising and investigating the company's data, and links conference talks by Eirik Bakke on the underlying design.

Why it matters

The gap between spreadsheets and databases is a longstanding pain point. Spreadsheets handle a single table well but fall apart when data must be combined and queried in many different ways, while databases demand SQL skills or bespoke internal tooling. A general-purpose GUI that treats joins and one-to-many relationships as first-class, visual citizens could lower that barrier from both directions — giving spreadsheet-native analysts access to real databases, and giving engineers a faster alternative to hand-typed SQL.

The claims around generality and one-to-many handling are the vendor's own, so the free trial is the fastest way to judge fit. But the underlying idea — a multidimensional interface for a set of database tables rather than a flat grid — echoes ambitions dating back to Lotus Improv, and the traction on Hacker News suggests the technical audience finds the pitch credible.