**Size:** The dataset has 5 tables and 446 rows in total (95 + 96 + 115 + 98 + 42). The table-overview query returned each table's row count, as shown in the rendered result. - **Columns:** Each table has 9 business columns: Company Name, Founded, Location, Sector, What it does, Founders, Investor, Amount and Stage. The raw tables also carry 4 ingestion-metadata columns (`_ingestion_timestamp`, `_batch_id`, `_source_file`, `_source_system`), so they show 13 columns in the dataset card. The tables don't join to each other, so there is no single combined table. - **Rows per table:** January has 115, March 98, February 96, April 95 and May 42. **Time period:** The data covers January to May 2022, based on the table names (one table per month) and the dataset's title. - **No funding-date column:** None of the tables has a column giving the date of each funding round. The month is known only from which table a row sits in. - **`Founded` is not a funding date:** It is the year the company was founded (about 1871 to 2022 across the tables), so it doesn't describe the funding period. - **Ingestion timestamp:** `_ingestion_timestamp` records when the data was loaded, not when the funding happened. I did not check its values. - **May table:** It is the smallest at 42 rows. According to the dataset card, some of its rows are misaligned or blank, and its `Founded` column is stored as text rather than a number.