Turn the sample workbook into an XLSX to pivot chart
An XLSX to pivot chart starts from a workbook, not from a typed-in table. Click Sales, 12 rows on this page and the browser fetches pivot-sales-sample.xlsx, then reads it with the same parser your upload uses. The file has one worksheet, named Sales. Row 1 is the header. Rows 2 through 13 are the records. The columns are Region, Category, Sub-Category, Sales, Quantity, and Order Date. There is no second sheet to discover, which is the situation you want when you are testing the tool.
The chart opens on Sum of Sales by Region and Category. East totals 333,187, West 404,815, Central 246,650, and South 209,430. The grand total is 1,194,082. Technology across regions is 528,627, Office Supplies is 386,005, and Furniture is 279,450. Those figures are the check on the parser. If your own workbook's first sheet is the same extract, you should land on the same cross-tab. If the totals differ, the header row, a hidden total row, or a different first sheet is the first place to look.
You can download the sales sample workbook and open it in Excel before uploading it again. The Order Date cells are stored as the text 2026-01-15 and similar ISO dates, not as Excel date serials. That is deliberate. The detector recognizes that text, shows the date grouping control, and can bucket the column by day, week, month, quarter, or year. A workbook you save from Excel after converting those cells to dates may display 1/15/2026 instead. Override the field type to Date if the grouping list does not appear.
What the XLSX reader keeps and what it ignores
The reader takes the first worksheet and turns its used range into rows. Formatting, column widths, colors, comments, defined names, and charts that already live in the file are not imported. Formulas come across as their displayed values when the file stored a cached result, which is normal for an .xlsx saved by Excel. A formula with no cached value can arrive blank. PivotTables that already exist in the workbook are not reused. This XLSX to pivot chart builds a new summary from the flat rows. If the workbook only contains a finished PivotTable and not the original transactions, upload that source range instead.
File formats that are supported in Excel distinguishes .xlsx from older .xls files and from CSV text. This page accepts .xlsx and .csv. It does not accept .xls, .xlsb, or .ods. If you still have a binary .xls workbook, open it in a spreadsheet and save a copy as .xlsx, or save the sheet as CSV and use the CSV input. The CSV to Excel tool can also wrap a CSV in a new .xlsx file, with values stored as text so leading zeros survive. That text storage is a different choice from this chart page, which parses numbers so it can sum them.
Merged cells, a title in row 1, and subtotal rows are the usual reasons an XLSX to pivot chart looks wrong. Unmerge the headers so each column has one name. Delete the title so row 1 is Region, Category, and the rest. Delete subtotal rows so Sum does not add a number that is already a sum. Blank header cells are renamed Column 1, Column 2, and so on. Duplicate header names get a numeric suffix. The sample workbook does not need those repairs: all six headers are unique and filled.
Two more workbooks: traffic and survey
Traffic, 8 rows fetches pivot-traffic-sample.xlsx. Its first sheet has Month, Channel, and Sessions. Month is an ISO date, so the chart can be a line of monthly sessions. The monthly totals are 6,000, 6,650, 7,060, and 7,380. Organic plus Paid is 27,090. Use this workbook when the Excel file you care about is a time series and you want to see the date grouping control after a real .xlsx parse, not only after an in-memory sample.
Survey, 9 rows fetches pivot-survey-sample.xlsx, with Team, Topic, and Responses. The cross-tab totals 396 responses. Support is 191, Success is 130, and Sales is 75. It is a clean small workbook for checking that the first sheet's header became the field list. The chart axis is labeled in thousands, and 396 is a small number on that scale, so confirm Survey in the pivot table. The sales and traffic workbooks are the ones that match the axis.
Each sample button replaces the previous rows. The file name shown in the tool becomes the .xlsx name, and the row count should match the button: 12, 8, or 9. If a fetch fails, the page falls back to the same rows held in the script and shows that fallback in the status area. The totals stay comparable either way. Your own upload never uses that fallback. A local file that fails does so with the parser's error, and the previous sample remains until a file parses.
From the first worksheet to a chart you can export
After the sheet loads, check the detected types before you trust the picture. Sales and Quantity should be numbers. Order Date should be a date. Region, Category, and Sub-Category should be text. Then set the row, column, and value. The sample chart uses Region, Category, and Sum of Sales. Change the value to Quantity and the grand total becomes 5,675, which is 640+830+220+710+930+260+410+510+185+340+470+170. That second total is a useful sanity check that you are not still looking at Sales.
Filters apply to the parsed rows, not to an Excel slicer. Filtering Category to Furniture leaves 71,654, 94,721, 63,812, and 49,263, which sum to 279,450, from 4 of 12 rows. The chart follows the filter. Clear it before you export a chart that is supposed to represent the whole sheet. The on-screen pivot table shows up to 10 row groups; this sample has 4 regions, so the screen and the export match. A workbook with more than 10 row groups still exports every group in the CSV and in the XLSX summary.
Export PNG from the Chart tab once the 1,194,082 total is the total you expected. Export CSV if you only need the cross-tab. Export XLSX if you want the parsed rows and the summary together. The exported workbook is a new file, pivot-chart-maker-export.xlsx. It contains the raw rows and the calculated pivot values. It does not contain an Excel PivotChart, a PivotCache, or a Refresh button. Create a PivotChart remains the way to build that native object from Sheet1 Raw Data after you open the export in Excel.
Size, privacy, and the first-sheet rule
Stay under 10 MB, 200 data rows, 30 columns, and 6,000 cells on the sheet that is first in the workbook. A financial model with many formula columns can pass the row limit and still fail the column or cell limit. Copy the columns you intend to group onto a new first sheet, or save that range as a separate .xlsx, and upload the smaller file. The ignored sheets do not count toward the limit because they are not parsed, but they do count toward the 10 MB file size. A 9 MB workbook whose first sheet is tiny can still be too large to open here.
The sample .xlsx files are requested from this site when you click a sample button. That request is the sample file, and its contents are the public rows listed above. Choosing your own workbook should not add a request that contains those private bytes. The file input reads the disk through the browser. Confirm that in the network panel: select a local .xlsx and watch for a new upload. Parsing and the chart update should proceed without one. The Excel-writing library may download the first time you export XLSX. That payload is code. The file you save is generated locally.
When the workbook is wrong for a chart, fix the sheet rather than the picture. One header row, ISO dates or an explicit Date override, numeric measures, no subtotal rows, and the intended sheet in front. Then build the XLSX to pivot chart again. The online pivot table shows the same engine with the table opened first, which is the better check when the workbook is messy. This page keeps the chart in front because the search that lands here is asking for the picture.
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
An .xlsx file is the Office Open XML workbook Excel writes by default. Microsoft lists it among the workbooks Excel opens in File formats that are supported in Excel. Create a PivotChart is the desktop command this page does not replace.
Questions about this page
Which sheet becomes the XLSX to pivot chart?
The first worksheet only. Later sheets are not listed and are not combined. In the sample workbook the first sheet is named Sales and has 12 data rows. Move the sheet you want to the front before uploading a larger workbook.
Will the exported workbook still be a live Excel pivot chart?
No. The download has a raw-data sheet and a sheet of calculated values plus a static chart image. Excel will not refresh those values when the source rows change. Build a native PivotChart in Excel if you need refresh.
Why did my date column stay as text?
The detector looks for YYYY-MM-DD, a month name plus a year, or a day plus a month name plus a year. A cell that Excel displays as 1/15/2026 can arrive as that text and stay a text field. Set the type override to Date, or write the sample's ISO dates before you save the workbook.
What are the size limits?
10 MB, 200 rows, 30 columns, and 6,000 cells, counted on the first worksheet after the header. The sales workbook is 12 rows and 6 columns. Files over those limits need a smaller extract.