Build an online pivot table from the 12-row sales sample
An online pivot table answers a grouped question without scrolling every source row. This page opens that table first. The chart is available on the next tab, using the same fields, filters, and aggregation. Load the sales sample to see the layout before you replace it with your own CSV or XLSX file.
The sample has 12 data rows and 6 columns: Region, Category, Sub-Category, Sales, Quantity, and Order Date. Dates run from 2026-01-09 through 2026-04-11. There is no title row, no blank row, and no subtotal row. That shape is what the parser expects: the first row supplies field names, and every later row is one record.
Set Region as the row field, Category as the column field, and Sales as the value, with Sum. The online pivot table then has four row labels and three category columns. East is 333,187, West is 404,815, Central is 246,650, and South is 209,430. The grand total is 1,194,082. West is the largest region. You can check any cell against the source: West Technology is the single Accessories row, 183,656, because each region has only one row per category in this file.
Use the survey and traffic samples on the same table
Two more samples sit above the tool so you do not have to invent a file just to learn the controls. The survey sample has 9 rows and 3 columns: Team, Topic, and Responses. Sum of Responses with Team on rows and Topic on columns totals 396. Support is 191, Success is 130, and Sales is 75. Onboarding is 162, Billing is 111, and Product is 123. Support is about 48.2 percent of 396. Switch Aggregation to Count and the grand total becomes 9, one count per source row, because every Responses cell is filled.
The traffic sample has 8 rows: Month, Channel, and Sessions. Months are stored as 2026-01-01, 2026-02-01, 2026-03-01, and 2026-04-01 so the date control can group them. With Month grouped by month, Channel across the columns, and Sum of Sessions, the four monthly totals are 6,000, 6,650, 7,060, and 7,380. The grand total is 27,090. Organic is 18,830 and Paid is 8,260. Stay on the Pivot Table tab while you read those totals. Open Chart afterward if you want the line.
Choose one aggregation and read the row count
Sum adds numeric values. Average divides that sum by the number of numeric rows in the group. Count counts non-blank value cells. On the sales sample, Average of Sales by Region, with the column field cleared, is 111,062.33 for East, 134,938.33 for West, 82,216.67 for Central, and 69,810 for South. Each of those averages uses 3 source rows. Count of Sales by Region is 3, 3, 3, and 3, and the grand count is 12.
The status line above the result reports how many source rows are included. With no filter it should read 12 of 12 for sales, 9 of 9 for the survey, and 8 of 8 for traffic. If you add a filter, that fraction drops. The on-screen table lists at most 10 row groups. These samples have 4, 3, and 4 groups, so nothing is hidden. A larger file can have more groups than the screen shows; the CSV and XLSX exports include every group, not only the first 10.
Group dates, then filter the online pivot table
Order Date is detected as a date because the sample uses YYYY-MM-DD. The grouping list offers the original day, week, month, quarter, or year. Month labels for this file are 2026-01, 2026-02, 2026-03, and 2026-04. Sum of Sales by month is 528,627, then 386,005, then 215,638, then 63,812. Those four numbers add back to 1,194,082. Quarter grouping collapses January through March into 2026-Q1 and leaves April in 2026-Q2. Year grouping returns a single 2026 row. Week labels use an ISO week such as 2026-W02. The tool does not compute the change from one month to the next. April minus March is 63,812 minus 215,638, which is a drop of 151,826, and you calculate that from the exported column.
Filters keep or drop dimension values before the aggregation. Add a Category filter and leave only Technology. The four Technology amounts are 152,805, 183,656, 101,734, and 90,432, and the filtered total is 528,627 from 4 of 12 rows. Add a Region filter for West only and the table shrinks to West's three category rows, totaling 404,815. Clear the filter pills to return to 12 of 12. A filter is a checkbox list, not a formula. If a value is missing from the list, it was blank or the field type did not match the text you expected.
Export the table, then decide whether the chart helps
Export CSV downloads the pivot matrix only: row labels, column labels, and the aggregated numbers. It does not include the 12 source rows. Export XLSX builds a workbook with two worksheets. Sheet1 Raw Data contains the source columns and rows. Sheet2 Pivot Chart contains the field names, the aggregation, the filtered row count, a chart image when the chart canvas is available, and the pivot values starting at row 33. Those pivot cells are stored numbers. They do not refresh when you edit Sheet1 inside Excel. Create a PivotTable to analyze worksheet data explains Excel's refreshable PivotTable, which this export is not. If you need that object, use the raw sheet as the source inside Excel.
The Chart tab can show a clustered bar, a line, or a pie of the same aggregation. Pie is a reasonable look only when you have one value column and a short list of slices. The sales cross-tab has three category columns, so a bar comparison is the clearer default. Export PNG from the Chart tab. On the Pivot Table tab the chart canvas is not displayed, so the PNG button is the wrong control for a table-only check. The vertical axis labels are scaled in thousands. Sales figures in the hundreds of thousands are readable that way; the survey's 396 responses are easier to audit in the table than on that axis.
Limits and a network check for this online pivot table
The file input accepts .csv and .xlsx up to 10 MB. Parsing stops above 200 rows, 30 columns, or 6,000 cells. The sales sample is 12 by 6, which is 72 cells, so it is far inside the cap. An XLSX workbook contributes only its first worksheet. A second sheet is ignored, even if the pivot you wanted lives there. Move that sheet to the front, or save it as its own workbook, before you upload it. CSV files use the first row as headers. A report title in row 1 becomes the field names and the real header becomes a data row.
To check privacy, open the browser's network panel, refresh this page, and then load a sample or choose a file. You should see the page's scripts and styles, not a request whose body is the spreadsheet. Choosing a file uses the local file API. The first time you export XLSX, the page may fetch the Excel-writing library if it is not already in memory. That response is script, and the workbook is assembled on the device and saved through a download. Refresh the tab when you want the rows discarded. Nothing in this online pivot table writes your file to an account, because there is no account.
Checked result
Sum of Sales by Region and Category in the 12-row sales sample. Grand total is 1,194,082.
| Region | Technology | Office Supplies | Furniture | Total |
|---|---|---|---|---|
| East | 152,805 | 108,728 | 71,654 | 333,187 |
| West | 183,656 | 126,438 | 94,721 | 404,815 |
| Central | 101,734 | 81,104 | 63,812 | 246,650 |
| South | 90,432 | 69,735 | 49,263 | 209,430 |
| Grand total | 528,627 | 386,005 | 279,450 | 1,194,082 |
West is 404,815 of 1,194,082 sales, about 33.9 percent. Technology is 528,627, about 44.3 percent. These shares are calculated from the table; the tool does not add a percent-of-total column.
Source note
Microsoft documents the desktop version of this layout in Create a PivotTable to analyze worksheet data. This page uses the same row, column, and value idea, then calculates the summary in the browser instead of in Excel.
Questions about this page
What is an online pivot table here?
It is a summary built in this browser from a CSV or XLSX file. You pick a row field, an optional column field, a value field, and Sum, Average, or Count. The sales sample produces a 4 by 3 table whose grand total is 1,194,082.
Does the chart replace the table?
No. This page opens on the Pivot Table tab. The Chart tab draws a bar, line, or pie view of the same aggregation. Export the table as CSV or XLSX when you need the figures.
Can it calculate percent of total, median, or distinct count?
No. Aggregations are Sum, Average, and Count. You can divide a cell by the grand total yourself. The sales table's Technology column is 528,627 of 1,194,082, which is about 44.3 percent, but that percentage is not a built-in field.
Where does the file go?
The browser reads it locally. Refreshing the page clears the rows. A first XLSX export may request the Excel-writing script; that request is application code, not your spreadsheet.