Convert a PDF to PDF/A for long-term archiving, court filing and records retention.
Free tier supports files up to 50MB.
Processed in your browser — files never leave your device. How it works
Courts, government agencies, universities and records departments often insist on PDF/A because a normal PDF can depend on things that may not exist in twenty years: fonts on your computer, colour settings, scripts or external files. PDF/A (ISO 19005) rules all of that out, so the file carries everything it needs. PDFDoco converts your PDF in your browser — embedding missing fonts, adding an ICC colour profile and archival metadata, and removing forbidden features — then tells you exactly what it changed, including anything it couldn't guarantee.
Any PDF you need to archive or submit: contracts, filings, reports, theses, records.
PDF/A-2b suits most uses; pick 1b or 3b if a portal or process asks for it.
Fonts, colours, metadata and forbidden features are fixed on your device.
See exactly what changed and anything that couldn't be guaranteed, then download.
A checklist of every fix, plus anything we couldn't guarantee — no blanket promises.
Missing Helvetica, Arial, Times and Courier fonts are embedded with exact-width substitutes.
Conversion runs in your browser. Sensitive filings are never uploaded.
Choose PDF/A-2b unless told otherwise: it's widely accepted and allows modern features such as transparency, so your pages almost always stay as real text. Pick PDF/A-1b when a portal specifically asks for it; because it forbids transparency, transparent pages are flattened to images. Pick PDF/A-3b to keep attachments, for example the XML inside a ZUGFeRD or Factur-X e-invoice. Files are saved at conformance level b (basic), the level most requirements ask for, which guarantees reliable appearance.
A PDF that merely names a font relies on the reader's computer having it, so PDF/A requires every font to be inside the file. When Helvetica, Arial, Times or Courier aren't embedded, PDFDoco embeds Arimo, Tinos or Cousine — open-source fonts with exactly the same character widths — so lines and page breaks don't move. A page that uses any other missing font is rendered as a high-quality image at the resolution you choose, which keeps its appearance but gives up selectable text. The report names every font and page involved.
PDF/A needs an output intent, which defines what the colours mean. PDFDoco embeds an sRGB ICC colour profile and, if the file uses CMYK print colours, a CMYK profile so they're device-independent too. It writes XMP metadata declaring the PDF/A part and conformance, keeps the title, author, subject and dates in sync between XMP and the document properties, and adds the required file identifier.
Encryption, JavaScript, launch and media actions, automatic triggers and XFA forms are removed, as are attachments unless you choose PDF/A-3b. Annotations are set to print and show their normal appearance; ones with no visible appearance are removed. Links, bookmarks, and form fields with an appearance stay. Files that need a password to open must be unlocked first.
Only a dedicated validator can state whether a file conforms, and we don't pretend otherwise. PDFDoco re-reads every saved file and runs its own structure checks: header, XMP identification, ICC profile, embedded fonts, and the absence of scripts and encryption. It doesn't re-verify every detail inside fonts that were already embedded. For legal, court or regulated submissions, run the file through veraPDF, the free industry-supported validator, or the validator your court or portal uses.
Pro tip: convert from the original file whenever you can. A PDF exported directly from Word or your design app usually embeds its fonts, so every page stays as real, searchable text. Keep your original too: PDF/A removes features you might want to edit later.
PDF/A is the ISO 19005 family of standards for long-term archiving. A PDF/A file contains everything needed to display it the same way decades from now: all fonts are embedded, colours are defined by an ICC profile, metadata is standardised, and features that depend on outside resources — such as JavaScript, encryption, or links to external content — are not allowed.
PDF/A-1 (2005) is based on PDF 1.4 and forbids transparency and JPEG 2000. PDF/A-2 (2011) is based on PDF 1.7 and allows both, along with layers and embedding other PDF/A files. PDF/A-3 (2012) adds arbitrary file attachments, used for e-invoices such as ZUGFeRD and Factur-X. The letter b means basic conformance: the file will look right, though its text isn't necessarily mapped for extraction. If you have no specific requirement, choose 2b.
Many courts and agencies require PDF/A, and files from this tool are built to meet the standard's requirements, but we can't certify any file. For legal or court submissions, check the result with a validator such as the free veraPDF (verapdf.org) or the validator your court or portal uses. Requirements also vary: some portals want 1b specifically, or limit file size.
Normally yes. Text, images and layout are kept. Fonts that weren't embedded are replaced by metric-compatible faces (Arimo for Helvetica/Arial, Tinos for Times, Cousine for Courier), so words stay in the same position, though letter shapes can differ slightly. Hidden annotations are made visible or removed, because PDF/A doesn't allow hidden content in annotations.
Some things can't be repaired in place: a page may use a font that isn't embedded and has no compatible substitute, or, for PDF/A-1b, transparency or JPEG 2000. Rather than produce an invalid file, those pages are rendered as high-quality images inside the PDF/A. They look the same, but their text can't be selected or searched. The report lists the affected pages and why; PDF/A-2b usually needs far fewer of them.
JavaScript, launch and media actions, automatic triggers, encryption, XFA forms, audio and video, and (for 1b and 2b) file attachments. Regular links, bookmarks, and form fields with a visible appearance stay. Choose PDF/A-3b if you need to keep attachments.
No. The conversion runs entirely in your browser, which matters for legal, medical and financial records. The only things fetched from the network are open-source fonts and helper data when a file needs them; your PDF is never sent anywhere.