Dataset transform tools for reshaping tables
Choose a tool by the table change you need: build a grouped summary, rename CSV headers, swap rows and columns, or turn wide columns into long attribute/value records. The dataset is reshaped locally in your browser, with no upload.
Choose by the output you need
Each tool changes a different part of a tabular dataset. Pick the one that matches the result you need.
- For grouped calculations, choose Pivot Table Generator. It groups rows by row and column fields, then calculates values with count, sum, average, minimum, or maximum. You can also set missing-value and subtotal behavior before exporting to CSV or a spreadsheet.
- For header changes, choose Rename Columns. It applies an explicit old-to-new mapping to CSV headers, preserves column order and every cell value, and reports missing source headers or name collisions before export.
- For a different orientation, choose Transpose Table. It swaps rows and columns in a rectangular or ragged table. Set header treatment, row-label handling, padding for uneven rows, and blank-cell display before previewing the result.
- For wide-to-long conversion, choose Unpivot Table. Select identifier columns and the columns to unpivot, name the output attribute and value columns, set blank-value handling, and check the record count before export.
Group rows or create long records
Use Pivot Table Generator when rows need to become a grouped summary. In a table with product, month, and sales fields, set product as the row field, month as the column field, and sales as the value field. The selected aggregation controls the calculation in each result cell.
Use Unpivot Table when each selected attribute needs its own record. Identifier columns are carried into the output, while the selected wide columns become the attribute and value fields. You choose which columns to unpivot and how blank values are handled.
Change a table's orientation or headers
Transpose Table changes the table's orientation. A rectangular input with three rows and two columns becomes two rows and three columns, while its options determine how headers, row labels, uneven rows, and blank cells are treated. Ragged input needs a padding choice so shorter rows have a defined shape.
Rename Columns changes the CSV headers while leaving column order and cell values in place. Enter an old header and its new name in the mapping. The preview reports a missing source header or a destination name collision before you export the CSV.
Typical jobs
- A reporting table needs totals by product and month. Choose the row fields, column fields, and value aggregation in Pivot Table Generator, then export the summary.
- A CSV needs new header names while its data stays unchanged. Add the old and new names in Rename Columns, then check the preview for missing headers or collisions.
- A spreadsheet stores repeated measures in separate columns, while the output needs attribute/value records. Select the identifier and measure columns in Unpivot Table, then review the record count.
- A table needs its rows and columns swapped. Open Transpose Table, choose the header and padding behavior, and preview the new layout.
Find related dataset tools
For profiling, cleaning, reshaping, merging, and generating tabular data, visit Dataset Tools.
Frequently asked questions
What are dataset transform tools used for?
They change the arrangement or labels of tabular data. Use them to summarize rows, rename CSV headers, transpose a table, or convert wide columns into long attribute/value records.
Which tool makes a pivot table from CSV data?
Pivot Table Generator makes a pivot table from tabular data. You choose row fields, column fields, and value fields, then select count, sum, average, minimum, or maximum.
How do I convert a wide table to long format?
Use Unpivot Table and select the identifier columns plus the columns to unpivot. You can name the output attribute and value columns, control blank values, and preview the record count.
Can I change CSV column names without changing cell values?
Yes. Rename Columns uses an explicit old-to-new mapping, preserves column order and every cell value, and reports missing source headers or name collisions before export.