LUCERIA
Sign in
Free tool · runs in your browser · no account

What a parser reads: extracted text preview

Drop in a PDF or Word file and the page shows the extracted text line by line, marks the headings and bullets Luceria's engine recognises, and flags some artefacts that usually come from tables, columns, text boxes or repeated headers; it does not catch every reading-order problem (in Luceria's benchmark it flagged none of three column layouts that read out of order), so read the text it shows. The file is read in your browser and never uploaded. It shows what one parser reads; employer systems vary, so treat it as a plain-text view of your document rather than a verdict.

Everything you type or upload is processed in your browser by this page's own code. Nothing is uploaded, stored or logged by Luceria, and analytics record only that the tool was used, never its contents.

Use the tool

Pick the symptom you saw in an application system. The explanation names the layer it belongs to: file accepted, text extracted or fields interpreted.

Upload a PDF or Word file

Up to 5 MB · read in this browser, never uploaded

Useful for checking copied text before pasting it into an application form.

How it works

  1. Text is extracted by Luceria's browser-side parser (PDF text layer or Word document body), the same extractor the free resume check uses.
  2. Each line is classified: a recognised section heading, a bullet, body text or blank, using the engine's own heading vocabulary.
  3. Layout heuristics look for tab characters, wide spacing inside lines, many one-word lines, very long lines, upper-case body text and repeated contact details, each explained with its likely cause.
  4. Email, phone and link presence is checked with simple patterns; the values are shown only inside the preview text and never recorded.
  5. When the preview finds a layout flag, missing contact details or no text layer, it shows the one-change fix measured in Luceria's document repair lab (with the before and after files); those figures come from Luceria's server-side extractor on synthetic files, not from any employer's system.

What it does not do

This shows what Luceria's own extractor reads. Other parsers may read a file differently, and no preview can show how an employer's system scores or stores it. A clean preview means the text is readable in order, not that the content fits any job.

Your data

Everything you type or upload is processed in your browser by this page's own code. Nothing is uploaded, stored or logged by Luceria, and analytics record only that the tool was used, never its contents.

Worked example

Illustrative example with invented details, not a real person’s result.

A fictional warehouse team leader resume: inputs

  • A single-column Word document: name and contact line, then Profile, Experience with two roles, Education and Skills.
  • Bullets typed with the bullet-list button; skills separated with middle dots.

What the tool shows

  • Four headings recognised: Profile, Experience, Education, Skills.
  • Three bullet lines marked; the contact line found with an email and a phone number; no portfolio link.
  • No layout flags: no tabs, columns or repeated headers.
  • The line-by-line view shows the roles in the intended order, so a parser reading top to bottom sees the same story a person does.

How to use it well

  1. Preview the exported file you will submit, because the export is what a parser receives.
  2. Read the line list from top to bottom as if you knew nothing about the layout; that is the order a parser sees.
  3. If a flag mentions tables or columns, rebuild that part as plain single-column text and preview again.
  4. Check that your name, email and phone appear in the body text, not only in a header or image.

Common questions

Why does my sidebar text appear mixed into the main text?

Most extractors read a page in text-stream order. Two-column layouts, sidebars and text boxes are stored as separate streams, so their lines interleave. A single column avoids the problem.

My PDF shows almost no text. Why?

It is probably a scanned image or an exported design with text converted to outlines. There is no text layer to read, so save a text-based PDF or paste the text.

Does an employer's ATS read files the same way?

Not necessarily. Parsers differ, and vendors document their own limits. This preview shows one real extractor's view so you can remove the problems that affect most of them.

Keep going

Published 6 November 2026 · Drafted with AI assistance and checked against Luceria’s editorial standards. Spotted a problem? Tell us.