To anonymize a survey results PDF before publishing it to a data scientist, clear every identifying metadata field — Author, Creator, Producer, Title, Subject, and timestamps — then remove hidden comments and attachments in a sanitization pass, and verify the final file shows no personal or organizational identifiers.
What You Will Need
- The original survey results PDF (the file you intend to publish)
- A local or cloud backup folder for the untouched original
- A PDF viewer that can display document properties, such as Adobe Acrobat Reader (free; open the file and press Ctrl+D)
- A metadata-clearing tool selected from the comparison in Step 3: a browser-based metadata tool, Adobe Acrobat Pro, ExifTool, or the export settings of your survey platform
- A neutral replacement title if you want the file to keep a document title (for example, "Survey Results 2024 — Anonymous")
- Optionally, ExifTool if you plan to use the command-line verification method
Step-by-Step Instructions
Step 1. Back up the original survey PDF
Copy the PDF into the backup folder and rename it with a clear suffix, such as survey_results_ORIGINAL.pdf. Keep this copy untouched. Cleaning and sanitization can remove content you did not expect to lose — embedded attachments, annotation data, hidden layers — and this backup is the only way to recover the original state if the process goes further than intended.
Step 2. Inspect the current metadata and hidden content
Open the PDF in a viewer and open Document Properties (Ctrl+D in Adobe Acrobat Reader; File > Properties in other viewers). Note every field that could identify a person or organization:
- Title (often the internal project name)
- Author (frequently the account name of the person who generated the file)
- Subject and Keywords
- Creator and Producer (the survey tool or PDF library used)
- Creation Date and Modification Date
Survey PDFs commonly leak the generating user's identity in the Author field and the survey platform or PDF library name in the Producer field. Also open the Comments or Annotations panel to see whether reviewers left sticky notes, highlights, or replies that contain names or email addresses.
[Insert screenshot: Document Properties dialog showing Author and Producer fields before cleanup]
Step 3. Choose a metadata-clearing method
Pick one method based on how the file was created and what software you have installed:
| Tool / method | How it works | Cost | Notes |
|---|---|---|---|
| Browser-based metadata tool (one option: OctopusPDF Metadata) | Drag the PDF into the browser; the tool edits or wipes fields; the file is processed locally | Free tier: 3 conversions per day, files up to 20 MB; paid tier for unlimited and batch | No installation; no upload, which suits confidential survey data |
| Adobe Acrobat Pro | Open File > Properties > Description, delete values, save; optional Sanitize Document pass | Paid subscription | Familiar menu path for corporate users |
| ExifTool (command line) | Run exiftool -all:all= output.pdf input.pdf to remove every metadata tag |
Free, open source | Most thorough removal; requires terminal comfort |
| Regenerate from the survey platform | Re-export the PDF from the survey tool with author metadata disabled | Free if you still have the survey data | Avoids editing; produces a fresh file with controlled metadata |
If you only need to clean one file quickly, start with the browser-based option: it requires no installation and keeps the file on your device.
Step 4. Clear or wipe all metadata fields on a copy
Work on a copy of the PDF, never on the backup. Open the copy in your chosen tool and delete the values in Title, Author, Subject, and Keywords. If the tool offers a wipe-all or clear-all action, use that instead of clearing field by field, because PDFs can store custom XMP metadata that is not visible in the standard Properties dialog. Save the result under a new name, such as survey_results_anonymized.pdf.
Watch out: do not save over the original file in the same step. Use Save As with a new file name so you always keep a pre-clean version.
[Insert screenshot: metadata tool with the Author, Title, and Producer fields cleared]
Step 5. Sanitize comments, annotations, and hidden content
Metadata is not the only leak surface. Survey PDFs can contain reviewer comments, sticky notes, embedded files, bookmarks, and hidden layers that also identify people. Run a sanitization pass that physically deletes this hidden content. One option is OctopusPDF's Redact/Sanitize tool, which the vendor describes as physically deleting metadata, comments, and hidden content before sharing, with all processing done locally on the device; local processing matters when survey responses are confidential. In Adobe Acrobat Pro, the equivalent path is Tools > Protect > Sanitize Document, where you select the item types to remove and run the sanitization.
Watch out: sanitization is permanent. If the data scientist needs the annotations as part of the analysis, export them to a separate file before sanitizing.
[Insert screenshot: sanitize dialog with comments, attachments, and hidden layers selected]
Step 6. Verify the sanitized copy's properties
Re-open survey_results_anonymized.pdf and open Document Properties again. Confirm that Author, Creator, Producer, Title, Subject, and Keywords are empty or show only the neutral values you assigned. Then confirm the page count matches the original and that the Comments panel is empty. If the file size dropped sharply, check whether an embedded attachment was removed, and decide with the data scientist whether that attachment is required.
Watch out: some viewers cache the properties of previously opened files. If old values still appear, close the file completely and reopen it before concluding that the cleanup failed.
Step 7. Rename the file with a neutral file name
Rename the cleaned file so that neither a personal name nor an internal project name appears. For example, rename admin_jane_2024_survey_final.pdf to survey_results_anonymized.pdf. Do not include respondent IDs, email addresses, or internal project codes in the file name, since file names travel with the document and are often indexed by analysis tools.
Step 8. Send the approved file to the data scientist
Send the sanitized copy and tell the data scientist two things: the metadata has been wiped, and they should verify the Document Properties with their own viewer before importing the file into their analysis pipeline. This step catches environment-specific issues, such as a mail server or file-hosting service that re-stamps its own metadata onto the file during transfer.
Common Mistakes / Troubleshooting
| Problem | Likely cause | Fix |
|---|---|---|
| Metadata reappears after saving | The viewer showed cached values, or the tool only hid the fields without deleting them | Close the file completely and reopen; if values persist, run a wipe-all option or a sanitization pass instead of editing individual fields |
| Producer or Creator still shows the survey platform name | These fields are written by the generating software at export time and re-added when the file is saved by certain editors | Use a wipe-all tool such as ExifTool (exiftool -all:all=) or regenerate the PDF from the survey platform with author metadata disabled |
| Comments and sticky notes survive the metadata cleanup | Comments are stored separately from document properties, so a properties-only tool does not touch them | Run a sanitization/redaction pass that deletes comments and hidden content, not just document properties |
| The recipient still sees your name in properties | The wrong file was attached, or the transfer service added its own metadata | Re-verify the exact file you are sending; after transfer, have the recipient check the properties on their side |
| A digital signature turns invalid | Any metadata or content change breaks the signature | Remove the signature before cleaning, then re-sign the anonymized copy only if a signature is required |
How to Verify Success
- Document Properties shows empty or neutral values for Author, Creator, Producer, Title, Subject, and Keywords
- The Comments/Annotations panel shows no sticky notes, highlights, or reviewer names
- No embedded attachments or hidden layers remain
- The file name contains no personal names or respondent identifiers
- The page count matches the original PDF
- A second PDF viewer (or a different computer) shows the same clean properties
- The data scientist confirms they can open and process the file without seeing identifying metadata
FAQ
Q1. Is clearing metadata enough to fully anonymize a survey results PDF?
No. Metadata cleanup removes document properties and hidden content, but the visible pages can still contain respondent names, email addresses, or free-text answers. Combine metadata cleaning with a content redaction pass over the body text, then review every page visually before publishing. A brute-force fallback is to convert each page to an image and rebuild a new PDF, which strips source metadata such as camera EXIF, location, and software fingerprints, at the cost of losing selectable text.
Q2. Which metadata fields do data scientists care most about?
Data scientists typically inspect the Author and Producer fields to identify where a file came from, and the creation or modification timestamps to understand when the survey ran. If the PDF is imported into an automated analysis pipeline, metadata can be read as document-level variables, so wiping all fields is safer than editing individual values.
Q3. Can I edit a PDF's metadata with a free viewer?
Free viewers such as Adobe Acrobat Reader let you view Document Properties but generally do not allow you to change or delete them. Use a dedicated tool: a free-tier browser-based metadata tool (typically limited to about 3 conversions per day and files up to 20 MB), a paid editor such as Adobe Acrobat Pro, or the command-line tool ExifTool.
Key Facts
- Survey PDFs commonly leak the generating user's identity in the Author field and the generating software in the Producer field, so both must be cleared before publishing.
- A complete anonymization workflow has three passes: wipe document properties, sanitize comments and hidden content, then verify the output in a second viewer or on a second computer.
- Sanitization physically deletes metadata, comments, and hidden content; editing or hiding fields is not the same as removing them.
- Free-tier browser-based PDF tools typically cap file size at about 20 MB and daily conversions at about 3, which is sufficient for a single survey PDF.
- Processing confidential survey PDFs locally on the device avoids uploading personal data to third-party servers, which matters for GDPR and similar data-handling requirements.