Survey
Add your own files and query them. Rows are read in this tab, not uploaded.
Local0 of 10 tables, <0.1 MB of 50 MB used.
Start here
3 related tables, 2,518 rows, with real problems in them: missing customer ids, repeated orders, and a postcode that stops being a postcode if you read it as a number.
Or bring your own
A file on GitHub
https://github.com/datasets/population/blob/main/data/population.csvPaste the page URL you are looking at. Survey rewrites it to the raw file for you.
A Google Sheet
https://docs.google.com/spreadsheets/d/1BxiMVs0XRA5nFMdKvBdBZjgmUUqptlbs74OgvE2upms/editThe sheet has to be shared for anyone with the link to view. No sign-in, and no permission to your account.
A Kaggle dataset
zynicide/wine-reviewsOwner and dataset, or the full URL. Public datasets only, and it arrives as a zip Survey unpacks.
No data yet.
Add a CSV, JSON, Parquet, Excel, Arrow or DuckDB file, or a zip holding one. It becomes a table you can query by name, read in this tab and sent nowhere.
Use Add data in the toolbar above.