A PDF represents displayed glyphs and text encodings through its font and document data. A visible word is not proof that every extraction tool will recover the intended characters. Ligatures can be represented correctly; they are not inherently an ATS failure.
Appearance and extraction are separate checks
Adobe’s text-extraction documentation describes unmapped glyphs, characters that cannot be represented in the output encoding, and problems when a font has an unknown or custom encoding without a ToUnicode table. A correct-looking page therefore does not guarantee usable extracted text. Check the exported words directly.
Test words that matter
Select and copy your name, email, job titles and important skills from the final file. Look for missing or substituted characters. A failed search can have more than one cause, so do not diagnose every problem as a ligature issue.
Repair the export when needed
If copied text is wrong, compare another export setting or a simpler font and check again. Keep a readable source document available for portals that request it. No named font guarantees correct parsing across all software.
- Inspect the final exported PDF
- Compare visible words with copied text
- Retest after changing fonts or export settings
- Review captured application fields