Datasets

Datasets are collections of tabular data that power the platform. Each dataset contains indicators, dimensions, and observations. Accessible from the management top navigation under Data → Datasets.

Creating a dataset

1. Go to the Datasets page from the management top navigation
2. Click Create
3. Fill in the fields:
   • Name (multi-language)
   • Description (multi-language)
   • Indicators — select one or more indicators
   • Data geographic granularity — the finest geographic level for which data is available
   • Fact table — select the database fact table that will store the data
   • Dimensions — select applicable dimensions
     • The year dimension is required.
4. Click Save

Editing a dataset

1. Click Edit next to a dataset
2. Update the fields
3. Click Update

Downloading an import template

Click Template next to a dataset to download a CSV template. The template is generated from the dataset's configuration — it includes columns for each selected indicator, dimension, and geographic level, with codes pre-filled in the correct format. This ensures your import file matches exactly what the system expects, reducing column-mapping errors.

If your source data is in a wide format (e.g. one column per year), use the Tidy Data Maker to reshape it into the long format that matches the template. The Tidy Data Maker is accessible from the datasets index page.

Importing data

1. Prepare your data using the template (see above)
2. Click Import next to a dataset (manage/dataset/{dataset}/import)
3. The dataset importer guides you through uploading an Excel or CSV file
4. Map columns and click Import

Truncating data

Click Empty to remove all observations from a dataset. This action requires confirmation.

Publishing / unpublishing

Toggle the publish switch on a dataset row to control whether it appears on the public Datasets page.

Deleting a dataset

1. Click Delete next to a dataset
2. Confirm in the dialog — this also removes associated data