Hidden Excel Data Can Leak Through PDFs: Our Test and a Safe Redaction Guide

当ページのリンクには広告が含まれています。
Anime privacy analyst detecting hidden sensitive information in exported PDF
ノートンストア

White text is not redaction. If you hide names, telephone numbers or identifiers in an Excel workbook by changing the font to white and then export that workbook to PDF, the information may remain as selectable, searchable or extractable text. Our original hands-on test demonstrated exactly this: a telephone number that was invisible on the page still appeared when the resulting PDF was converted to a text file.

This English article adapts our Japanese-language practical test using a fictional staff contact list. No real employee identity or telephone number is disclosed in the example. We also cross-check Adobe’s published PDF sanitization guidance. Reviewed November 8, 2026.

広告・PR
目次

The mistake: hide a cell by changing font color

Imagine preparing a staff list for an external partner. The source Excel table contains names, department names and emergency telephone numbers. The recipient needs only the first two columns. Instead of removing the phone numbers, the sender changes their text color to white so the cells look blank and exports the sheet to PDF.

The PDF appears to have no phone numbers, but visual concealment does not necessarily remove text from the PDF structure. If the number remains represented as a text object, it may be selected, copied, searched, exported or extracted by automated tools.

Illustrative sample Excel staff list with fictional contact information
Fictional sample spreadsheet used for the original experiment, not a genuine private staff list.
Excel sample shows telephone number cells formatted in white font
In the example, changing font color hides numbers visually but does not remove cell values.
PDF exported from Excel visually appears to have hidden the telephone number column
The exported PDF looks clean at first glance; that is not evidence of safe redaction.

Test result: the hidden telephone number remained extractable

In our test, the numbers were still present after Excel-to-PDF conversion. Using PDF text extraction revealed data that the normal page view had concealed. This occurs because PDF may store the characters, drawing positions and visual properties separately: an invisible-looking white glyph can remain an ordinary text object.

Original screenshot showing a PDF text-export workflow used to test for hidden content
Actual Japanese-language application screenshot from the test. Menu names may differ in English software versions.
Text extraction result from the sample PDF showing numbers that were hidden visually
Extracted text exposed the sample values, proving that visual concealment had not removed them.

Why the PDF still contains invisible data

An Excel file stores cell values separately from their formatting. Setting a font color to match the background changes presentation, not data. Exporting to PDF usually paints the document for display, but that process may preserve text objects for search and selection. Covering text with a white box, black rectangle, drawing or highlight can similarly leave the original text underneath. Exact results depend on the conversion engine and selected export method—do not assume an image conversion, print driver or other workflow removes all hidden data without testing it.

Hiding methodWhy it is unsafe as redaction
White font on white backgroundThe characters may still be in the document and copied or extracted.
Black or white rectangle over textThe original underlying text might still be selectable.
Comment, annotation or highlightAn annotation can overlay text without removing its underlying content.
Hidden rows, columns or sheetsThe original workbook still contains the information and some exports/workflows may expose it.
Custom number format that visually hides valuesDisplay formatting does not delete the cell’s value.
Flattened screenshot aloneIt may remove selectable text in that image but is not a complete way to sanitize metadata, hidden pages or other content.

Safer method: delete unnecessary data before making the PDF

  1. Make a working copy of the Excel file specifically for the external recipient. Preserve the source workbook securely for internal use.
  2. Remove sensitive columns, rows, worksheets, comments and embedded objects that the recipient must not receive; do not just change font or visibility.
  3. Use a test copy to confirm that formulas or linked sheets do not reconstruct sensitive values.
  4. Export the sanitized copy to PDF using the normal, approved workflow.
  5. Open the resulting PDF and select/copy text from areas where sensitive data used to appear; use the viewer’s search function to check known sample strings.
  6. If organizational policy requires thorough sanitization, use a dedicated PDF redaction/sanitization tool and inspect hidden metadata, attachments, layers and other document properties.
  7. Review the final output separately and send only the checked, sanitized PDF.
Excel sample after deleting sensitive telephone number values rather than hiding them
The safer approach removes the source values from the export copy before PDF conversion.
Text extraction after deleting sample private values demonstrates absence in output
In the revised sample, the deleted phone numbers did not appear in the extracted text.

If you must redact an existing PDF instead

Use a genuine redaction feature, not a white rectangle, highlighter or text-color change. Adobe’s Acrobat Pro sanitization guide describes tools to remove hidden document content, including metadata, comments, hidden layers and sensitive material. Apply and save redactions permanently, then verify the saved file again. Acrobat Reader alone may not provide the full Acrobat Pro redaction/sanitization workflow.

A practical release checklist for finance, HR and school offices

  • Confirm the recipient and the minimum fields they actually need.
  • Use a new export-only copy rather than modifying a master payroll, contact or HR workbook.
  • Check every sheet, table, chart, print range, note, comment, hidden field and embedded object that may enter the output.
  • Inspect the resulting PDF for selectable hidden data, document properties, attachments and unexpected pages.
  • Use fictional sample identifiers in training materials and test files.
  • Require a second reviewer when the document contains particularly sensitive information.
  • Record the correct final filename and avoid accidentally attaching the original Excel workbook.

FAQ

Is blacking out PDF text with a rectangle safe?

Not on its own. A graphic overlay may not remove underlying text; use proper redaction that deletes the contents and verify it.

Can I trust a PDF because search does not find the hidden value?

A failed search is one useful check but not a guarantee; PDFs can carry hidden layers, embedded files, metadata or images. A risk-appropriate sanitization workflow is stronger.

Is the problem unique to Excel?

No. The general principle applies to document editors, presentation files and PDFs: altering visibility is not the same as removing confidential information.

What should I do if a PDF containing hidden sensitive data has already been sent?

Treat it as a possible data disclosure. Follow your organization’s incident reporting and containment process; contact the recipient through an appropriate channel, assess the actual content and access scope, and determine any notification duties with privacy/legal staff. Avoid assuming the data was unreadable because it was not visually displayed.

Sources and original experiment

Sample data in screenshots is fictitious. English adaptation reviewed November 8, 2026. The article documents the original export method’s observed behavior; different PDF conversion paths need separate verification. Visual concealment should never be treated as a security control.

よかったらシェアしてね!
  • URLをコピーしました!
  • URLをコピーしました!

この記事を書いた人

はじめまして、「ぴろりん」です。
このブログでは、IT運用管理・PC修理・情報セキュリティ対応・資格学習など、実務で直面した課題とその解決策を発信しています。
同じような悩みを抱える方のお役に立てれば幸いです。

目次