Machine-readable content: why a scanned table fails the exam

Machine-readable content means data that software can interpret directly—think of a typed table in a Word document versus a photo of that same table in a PDF. The photo is just an image; you cannot search for a number, copy a cell, or let a screen reader announce the rows. Structured text, on the other hand, lets you index every word, translate the whole document, transform it into a spreadsheet, and run accessibility checks. The core value is that machines can act on the content, not just display it.

A common trick in digital-skills questions is to present a scanned invoice or a screenshot of a chart and ask whether it is machine-readable. The answer is no—unless it has been run through optical character recognition (OCR) and saved as selectable text. Another hint: if you can highlight and copy individual words, it is machine-readable; if you can only copy the whole image, it is not. Also remember that assistive tools like screen readers rely on machine-readable text to function, so a PDF that is just a stack of scanned pages fails accessibility standards.

To test yourself, imagine you are given a PDF of a handwritten form. Ask: can a search tool find the word 'total' in it? If not, it is not machine-readable. A quick mental check: if the content looks like a photograph, it probably is not machine-readable; if it looks like a document you can edit, it probably is.