CSV vs XLSX: what you lose when saving a spreadsheet as CSV

Choose the deliverable before choosing the extension

Ask what the next person or system must do. If they need to change quantity and have totals recalculate, a text table is insufficient. If an import system asks for sku, quantity and price fields, sending an entire workbook can introduce extra sheets and formatting the importer does not understand.

The sample order has a SKU, two numeric inputs and a calculated total. Its workbook also includes a Notes sheet. Those are separate requirements: the order values can travel as a table, while the calculation and notes need another representation if the recipient must retain them.

Keep an editable master and export a copy

Save the working workbook as XLSX before exporting. Give the CSV a name that identifies its purpose and period, rather than reusing the master filename without an explanation. Work from a copy if the export requires deleting columns or replacing formulas with approved values.

Select the intended worksheet and confirm its first row contains the importer headings. Remove decorative title rows from the export copy if the recipient expects the first record to be the header. Do not remove meaningful data merely to make the table look cleaner.

1. Write down the required output headings and the intended worksheet.
2. Save the full workbook in its editable format.
3. Create an export copy with only the required table and approved values.
4. Use Save As or the export command to select the CSV format required by the recipient.
5. Read any feature-loss warning, complete the export and keep the master workbook separate.

Inspect the text before reopening it as a workbook

Open the exported CSV in a text editor or a parser that leaves the fields as strings. In the illustrative result below, the formula has become a total value of 25.00. That string will not recalculate if someone later changes quantity to 3.

The appearance of the total depends on the export workflow and cell representation. Do not assume that every workbook exports exactly two decimal places. Check the actual bytes you created and compare them with the destination specification.

sku,quantity,unit_price,total
00123,2,12.50,25.00

Separate export loss from import interpretation

Suppose the exported text still contains 00123, but double-clicking it in Excel displays 123. In that case the original CSV field may be intact; the opening application interpreted it as a number. Inspecting raw text distinguishes that situation from an export that already contains only 123.

A filename change from .csv to .xlsx does not create a workbook. Likewise, changing the displayed number format after an identifier has been altered is not a reliable recovery method. Reimport preserved source text with an explicit identifier type.

Related help: Tell CSV export loss apart from Excel zero removal

Reconcile a small set of important fields

For the sample, confirm one data record, four fields and the SKU 00123. Compare quantity 2, unit price 12.50 and total 25.00 with the approved source. Also verify that the Notes content has a separate destination if the recipient needs it.

For a real export, choose representative examples rather than only the first row: a leading-zero ID, a non-ASCII name, a blank field and text containing punctuation. Record the exported row count and any deliberate exclusions so another person can repeat the check.

Be precise about conversion claims

This guide demonstrates the difference between a workbook and a text table. It does not recreate XLSX features from a CSV, and it does not promise that one export option behaves identically across all Excel editions. The sample CSV was parsed independently; no Excel export was executed for this article.

Column Harbor currently inspects UTF-8 CSV text and selected risks. It does not read or convert XLSX files. Use the workbook application or a format-aware converter for that step, and compare the resulting table with the source before sending it onward.