How PDF Text Extraction Works
PDF files do not naturally understand paragraphs; they store character glyphs with coordinates (X, Y). When an ATS extracts raw text, it sorts glyphs by coordinates. If a page has two columns, the parser frequently reads line 1 of column 1 followed immediately by line 1 of column 2, destroying your sentence structure and dropping your qualifications.