PDF Format Explained: Everything You Need to Know in 2026
PDF — Portable Document Format — was created by Adobe in 1992 and became an ISO open standard in 2008. It is the dominant format for sharing documents that must look the same on every device, regardless of operating system, installed fonts, or screen size. A PDF preserves the exact visual layout of its creator's intent, which is why it is used for everything from tax forms to legal contracts to design portfolios. Despite being ubiquitous, the PDF format is deeply misunderstood. Most people know what a PDF looks like but not why the same PDF might be searchable on one device and not another, why some PDFs can be edited and others cannot, or why a PDF created from a Word document behaves completely differently from a PDF created by scanning a paper page. These differences are not arbitrary — they reflect fundamental choices in how the PDF was created. This guide demystifies the PDF format: what is in a PDF file, why PDFs behave differently, and how to work with each type effectively.
What a PDF File Actually Contains
A PDF file is a container — it holds different types of content depending on how it was created. At the most basic level, a PDF contains a page description: instructions for rendering each page's visual appearance. These instructions can be vector-based (mathematical descriptions of shapes and text) or raster-based (pixel images), or a combination. A PDF also contains resources shared across pages: embedded fonts (so text looks identical on any device), embedded images, color profiles, and structural metadata. Advanced PDFs contain additional layers: annotation objects (sticky notes, highlights, comments), interactive form fields, JavaScript for calculations and validation, bookmarks and navigation structures, and digital signature objects. The presence or absence of these layers determines what a PDF can do — and why two PDFs that look identical on screen may behave completely differently when you try to edit, search, or fill them.
- 1Check a PDF's type by trying to select text — if text selects, the PDF contains vector text
- 2If clicking on text selects the whole page as an image, the PDF is scanned (raster-only)
- 3Check for form fields by clicking apparent form boxes — active fields allow typing
- 4Check for bookmarks in Adobe Reader's sidebar — bookmarked PDFs have structured navigation
The Four Main Types of PDFs and How They Differ
Understanding PDF types explains why the same tools produce different results on different PDFs. Type 1: digitally created text PDFs — exported from Word, Google Docs, LaTeX, or code. These contain vector text and embedded fonts. Text is perfectly searchable, selectable, and sharp at any zoom. File sizes are small relative to content. Type 2: scanned image PDFs — created by scanning physical documents. Each page is a raster image. Text is not searchable unless OCR has been applied. Quality is fixed at scan resolution. Type 3: interactive form PDFs — contain text PDF content plus interactive form fields. Designed to be filled electronically. Fields may be fillable by default or locked. Type 4: PDF/A archive PDFs — a standardized subset of PDF designed for long-term archiving. Fonts are mandatorily embedded, color spaces must be device-independent, and external dependencies are prohibited. Required by many government and legal systems for official submissions. Most PDFs you encounter are Type 1 or Type 2, often without the creator's explicit choice.
- 1Identify your PDF type before deciding how to process it
- 2For scanned (Type 2) PDFs: apply OCR before searching or converting to make text accessible
- 3For form PDFs: check security settings — some are fillable, others are locked
- 4For archiving submissions requiring PDF/A: verify compliance or convert using Acrobat Pro's PDF/A export
Why PDFs Look Different in Different Viewers
PDF is a specification, not a single implementation. Adobe's specification defines what a valid PDF contains, but different viewer applications implement the specification with varying completeness. Adobe Reader implements the full specification. Browser-based PDF viewers (Chrome's built-in viewer, Firefox's viewer) implement most of the specification but omit some advanced features — certain XFA forms, some JavaScript interactions, and advanced transparency modes may not render correctly. MacOS Preview renders PDFs well for most content but is known to have occasional issues with complex transparency and some color spaces. On mobile, both iOS and Android render basic PDFs correctly but may not handle complex interactive elements. For most business documents — reports, letters, invoices — any modern viewer renders the PDF identically. For complex interactive forms, multimedia elements, or 3D content, Adobe Reader is the only fully compatible viewer.
- 1If a PDF looks wrong in your browser, download it and open in Adobe Reader for comparison
- 2If a form does not work in Chrome's viewer, try Adobe Reader — XFA forms require it
- 3If colors look different across viewers, the PDF may use CMYK color with an embedded ICC profile
- 4For critical print production, preview in Adobe Reader or Acrobat — the most accurate PDF renderer
PDF Versions and What They Mean
The PDF specification has evolved through multiple versions, each adding capabilities. PDF 1.0 (1993) through 1.7 (2006) introduced progressively more features — forms, JavaScript, AES encryption, layers, 3D content. PDF 2.0, published in 2017, deprecated many legacy features and clarified ambiguities in the specification. PDF/A, PDF/X, PDF/E, and PDF/UA are standardized subsets for specific use cases: archiving, print production, engineering documents, and universal accessibility respectively. For most users, PDF version numbers are invisible in daily use. Where they matter: PDF/A is required for legal and government submissions in many jurisdictions. PDF 1.4 or earlier uses RC4 encryption (weak). PDF 1.6+ with AES encryption is necessary for genuinely secure password protection. Ghostscript compression, used by LazyPDF, re-encodes PDFs as PDF 1.5 by default — compatible with all modern viewers but an older compatibility level than PDFs exported from current Office software.
Frequently Asked Questions
Why can I copy text from some PDFs but not others?
There are two distinct reasons. First, if the PDF is scanned, the pages are images — there is no text layer to select or copy. Applying OCR creates a selectable text layer. Second, if the PDF was digitally created but has a permissions password restricting text copying, you can see the text but cannot copy it. LazyPDF's unlock tool removes permissions restrictions (if you have the password). A third case: some PDFs use custom font encoding where the copy-paste output appears as gibberish — this is a font encoding issue in the PDF, not a permissions restriction.
Is a PDF always smaller than the same content in Word format?
Not always. A simple Word document exports to a PDF of comparable size. An image-heavy PowerPoint exported to PDF at print quality produces a much larger PDF than the PPTX. A PDF compressed with Ghostscript can be significantly smaller than either format. File size depends primarily on embedded content — images, fonts, media — not the format itself. DOCX files are ZIP-compressed containers, which often makes them smaller than uncompressed PDFs of the same document. Compressed PDFs using efficient JPEG image streams can be smaller than the equivalent DOCX with the same embedded images.
Why do merged PDFs sometimes have inconsistent page sizes?
PDFs can contain pages of different physical sizes — A4, Letter, Legal, or custom sizes. When you merge PDFs created from different source documents or different region settings, the merged result preserves each page's original size. Pages from an A4 document and pages from a US Letter document will appear slightly different in the merged output. Most viewers handle mixed-size PDFs correctly. For printing, the printer driver may scale pages to match the selected paper size. LazyPDF's merge tool preserves each page's original dimensions without normalization.