How to bulk convert Publisher (.pub) files to PDF

Which method fits your situation?

Short version: if someone in the office still has a working copy of Publisher, Microsoft’s script gives the most faithful PDFs. If not, LibreOffice is free but takes some scripting, and our converter uses the same importer, adds layout repairs and needs no setup.

Before you start: a planning checklist

When you need to mass convert an archive, it goes smoothly if you plan it like a records project, not a one-off click. Work through this list first.

1. Take an inventory. List every .pub file with its size and date. On Windows, in PowerShell:

   Get-ChildItem "D:\Shared" -Recurse -Filter *.pub -ErrorAction SilentlyContinue |
     Select-Object FullName, Length, LastWriteTime |
     Export-Csv "$HOME\Desktop\pub-inventory.csv" -NoTypeInformation

   On Mac or Linux: find ~/Shared -iname '*.pub' > pub-inventory.txt. Microsoft also suggests searching OneDrive and SharePoint for “.pub” (Microsoft’s FAQ).
2. Remove files that aren’t Publisher documents. Files like id_rsa.pub in .ssh folders are login keys, not documents.
3. Find duplicates. The same bulletin often exists in five places. In PowerShell, this groups identical files:

   Get-ChildItem "D:\Shared" -Recurse -Filter *.pub | Get-FileHash |
     Group-Object Hash | Where-Object Count -gt 1 |
     ForEach-Object { $_.Group.Path; "" }

   Convert one copy of each. Don’t delete the others until you’ve decided on a retention policy.
4. Keep the originals. Convert a copy of the archive, never your only originals. Afterward, store the .pub files read-only next to the PDFs.
5. Decide formats and names. PDF for the archive and printing, ODG if you’ll edit in LibreOffice Draw, text-only Word if you’ll reuse the words. Keep the same base name (2019-04-21 Easter Bulletin.pub becomes 2019-04-21 Easter Bulletin.pdf) so the two are easy to match.
6. Install your fonts (LibreOffice route). If your office used particular licensed fonts, install them on the converting computer first. Missing fonts are the biggest cause of changed line breaks.
7. Run a pilot. Convert 20 to 50 files from different years and staff members. Compare them with printed copies and time the run. Files from the 1990s behave differently from newer ones; see old Publisher files.
8. Run the full batch, then verify. Compare the number of .pub files with the number of PDFs, list the missing ones, and open a sample (about 5%, at least 20 files).
9. Handle failures. Retry them, then send the stubborn ones to someone with Publisher.
10. Write down what you did. A short note in the archive folder (date, tool, settings, known issues) will help whoever comes after you.

Route 1: Microsoft’s PowerShell script (needs Publisher)

Microsoft’s Publisher retirement FAQ links to a sample script called Convert-PubFileToPDF.ps1. It drives your installed copy of Publisher to export each file as PDF. Microsoft calls it a sample to adapt and test for your own needs.

Requirements (Microsoft Q&A):

• a Windows PC with a working, licensed copy of Publisher. Microsoft 365 copies were removed from October 1, 2026, so in practice that means Publisher 2016, 2019 or 2021
• the script, downloaded from the link in Microsoft’s FAQ
• PowerShell with an execution policy that lets the script run

Running it. Microsoft’s examples use -Filter and -Recurse. Change to the top folder of your archive copy and run:

# Allow scripts in this PowerShell window only, and unblock the downloaded file
Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass
Unblock-File "C:\Scripts\Convert-PubFileToPDF.ps1"

cd "D:\PubArchive-copy"
& "C:\Scripts\Convert-PubFileToPDF.ps1" -Filter "*.pub" -Recurse

Microsoft only says to use an appropriate execution policy. Limiting it to the current window, as shown here, is our suggestion.

What to expect (Microsoft Q&A):

• It keeps going when a file fails, then shows which files converted and which didn’t.
• If a PDF with the same name already exists, that file is skipped with an error. Re-running is safe; to redo a file, delete its PDF first.
• It produces PDFs only. There’s no direct route from Publisher to PowerPoint, and the scriptable path to Word goes through PDF (LazyAdmin).
• Test it on your pilot folder first to see where the PDFs are written.
• Publisher is now unsupported software. Run the batch on a PC used only for this job, and only on your own files.

Route 2: LibreOffice from the command line (free)

LibreOffice can convert files without opening a window. The --convert-to option implies headless mode, and --outdir sets where the output goes.

Simple case: one folder (Mac or Linux; on Mac use the full path shown below):

soffice --headless --convert-to pdf --outdir ./pdf ./pubs/*.pub

Whole archive, PDF next to each .pub, folders kept. Run these on a copy of the archive.

Windows (PowerShell). soffice.com is LibreOffice’s console version. Unlike soffice.exe, it prints errors to the window:

$soffice = "C:\Program Files\LibreOffice\program\soffice.com"
Get-ChildItem "D:\PubArchive-copy" -Recurse -Filter *.pub | ForEach-Object {
    & $soffice --headless --convert-to pdf --outdir $_.DirectoryName $_.FullName
}

Mac (Terminal):

SOFFICE="/Applications/LibreOffice.app/Contents/MacOS/soffice"
find "$HOME/PubArchive-copy" -iname '*.pub' -print0 |
  while IFS= read -r -d '' f; do
    "$SOFFICE" --headless --convert-to pdf --outdir "$(dirname "$f")" "$f" < /dev/null
  done

Linux (the timeout stops a damaged file from stalling the run):

find /srv/pub-archive-copy -iname '*.pub' -print0 |
  while IFS= read -r -d '' f; do
    timeout 180 soffice --headless --convert-to pdf --outdir "$(dirname "$f")" "$f" < /dev/null
  done

Find what didn’t convert. Windows:

Get-ChildItem "D:\PubArchive-copy" -Recurse -Filter *.pub |
  Where-Object { -not (Test-Path ([IO.Path]::ChangeExtension($_.FullName, ".pdf"))) } |
  Select-Object -ExpandProperty FullName

Mac or Linux:

find ~/PubArchive-copy -iname '*.pub' | while IFS= read -r f; do
  [ -e "${f%.*}.pdf" ] || echo "MISSING: $f"
done

Things that trip people up:

• Close LibreOffice first. If a LibreOffice window is open, a headless command can hand its job to that window and seem to do nothing.
• Running in parallel needs one profile per process. Two LibreOffice processes can’t share a user profile, so give each worker its own with -env:UserInstallation (details):

  soffice -env:UserInstallation=file:///tmp/lo-worker-1 --headless --convert-to pdf --outdir out/part1 part1/*.pub &
  soffice -env:UserInstallation=file:///tmp/lo-worker-2 --headless --convert-to pdf --outdir out/part2 part2/*.pub &
  wait

  On Windows the profile looks like -env:UserInstallation=file:///C:/Temp/lo-worker-1. Each worker uses its own memory, so two to four workers is plenty for an office PC.
• Other formats. --convert-to odg gives editable LibreOffice Draw files. --convert-to png gives only the first page. For images of every page, convert to PDF first, then split the PDF with a tool such as pdftoppm -r 150 -png file.pdf page.
• Stuck processes. If a run is interrupted and LibreOffice then refuses to start, look for a leftover soffice.bin process and end it.

Route 3: Our converter (no install, folders kept)

If scripting isn’t your thing, or you’d like a report you can hand to your supervisor, the converter on this site lets you convert multiple .pub files to PDF in one upload, from a handful to a whole archive. It runs the same libmspub importer as LibreOffice, then repairs some of its layout problems: a story in linked boxes on the same page flows in order, and text wraps around pictures. On top of that you get no setup, the preview and the paperwork.

• Upload a folder, a set of files, or a ZIP. Up to 2,000 .pub files per batch, 25 MB per file, and 1 GB uncompressed per ZIP. Other files in a ZIP are skipped and listed.
• Preview first. You get a free, watermarked preview of every page for the first 10 files in the batch, so you can judge quality on your own files.
• Pay once, only for what converts: $5 for up to 10 files, $15 for up to 60, $29 for up to 300, $59 for up to 2,000. Files that fail aren’t counted. See pricing.
• Download one ZIP that mirrors your folders. Each file comes with a print-ready PDF, an editable ODG, a text-only DOCX and 150 dpi PNG and JPG images of every page. A report.csv lists each file’s path, status, page count, warnings (such as substituted fonts or a page that looks blank) and a plain-English reason for any failure.
• No account. Each file is converted in a locked-down sandbox with macros disabled, and everything is deleted after 24 hours.

Start on the converter page to convert your .pub files to PDF in bulk.

Related guides

• Microsoft Publisher end of life: dates and options
• How to open a .pub file without Publisher
• How to convert a Publisher file to Word
• What is a .pub file?