Trigger 
Describe Converting a document
Preparing your document
Best results come from a PDF that already contains real text — the kind produced by Word, InDesign, your ESEF tagging software, or any other tool that writes text directly onto the page. DynDyn.Pdf reads the text, fonts and layout that are already inside the file; it does not read a scanned image and turn it into words — there is no OCR step. A page that's really a photograph or a scan, with no text underneath it, converts to a page with no selectable text either, because there was none to find in the first place.
What you get back
A single HTML file, self-contained: fonts, images and layout are all bundled into it, so it opens in any modern browser without anything else installed. The text on the page stays live and selectable, and the page is laid out to match the look of the original PDF as closely as possible.
Checking it against the original
Open the HTML file and the source PDF side by side and compare them page by page:
- Does the text read the same, in the same order?
- Do the numbers in tables match?
- Can you select and search the text in the HTML version — try searching for a word you know is on the page?
- Do headings, paragraphs and page breaks look right?
Small differences in spacing are normal; the two are produced by different software. Wrong words, missing sections, or text that looks like a picture instead of text are not, and are worth reporting.
Common problems
Text looks like it's using the wrong font, or the spacing looks slightly off. Usually means the original PDF didn't embed that font, or embedded it in a way DynDyn.Pdf isn't allowed to reuse. The software substitutes the closest available font automatically — the text stays live and correct, but its exact shape may not match the original pixel for pixel.
A page, or the whole document, has no selectable text. The source PDF is a scan or an image with no text layer underneath it. DynDyn.Pdf can't recover text that was never there. You'll need a version of the document that has real text in it, or a separate OCR pass before conversion.
The layout looks jumbled, or columns run into each other. Usually a genuinely unusual page — dense multi-column tables, rotated text, very small print. This is exactly the kind of case worth sending to support rather than working around.
What to send support
If a conversion looks wrong, send us:
- The original PDF, or tell us if it's confidential and can't be shared.
- The page number(s) where the problem shows up.
- A short description of what you expected versus what you got — a screenshot of the HTML output helps.
Reach us at /contact.