> ## Documentation Index
> Fetch the complete documentation index at: https://www.landbase.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Transform Data

> Run custom SQL transformations — joins, aggregations, derived fields — across any Landbase dataset.

# Transform Data

Most GTM tools let you filter. Transform Data lets you *compute*.

Joins, aggregations, derived fields, deduplication, custom calculations — anything you can express in SQL, you can run across Landbase's entire dataset. This is the power tool for GTM engineers who need to answer questions that filters can't.

## What it unlocks

* **Derived attributes** — "companies where >30% of employees are in engineering" (compute the ratio, filter on it)
* **Cross-dataset joins** — combine company data with job posting data to find companies hiring for specific roles in specific locations
* **Deduplication** — merge overlapping lists, resolve conflicts, pick the best record per company
* **Custom scoring** — build a composite score from multiple signals and rank your TAM by it
* **Keyword intelligence** — built-in functions for matching keywords and job titles with boundary awareness

**What's happening under the hood**

The query runs on a distributed compute cluster. Datasets are read from cloud storage (pre-partitioned files). The system validates your query with a dry-run before committing compute. Custom functions for keyword and job title matching are registered automatically. Output is written to cloud storage and optionally copied to Google Drive. A status page tracks execution progress.

## Available datasets

Transform Data can query across Landbase's core datasets including company, person, and job posting data — as well as any custom datasets you've registered.

## Output

Results are written to Google Drive or Google Sheets — ready for download, analysis, or import into your next workflow.

## Constraints

* Maximum 200,000 rows per query
* Uses Spark SQL syntax


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.