Why your PDF turns into a widescreen slide with your page shrunk in the middle
You convert a PDF to PowerPoint, open the result, and your A4 report is sitting in the middle of a widescreen slide with a thick empty band down each side. Nothing is broken. Nothing failed. It just looks wrong, and no setting in the converter seems to fix it.
This is not a bug in your file. It is a decision the converter made before it had even read your document, and it is worth understanding, because once you know what is happening you can tell which converters are doing the work and which are taking the easy route. You can see the alternative with the free PDF to PowerPoint Converter, which reads the page size out of your PDF and makes the slides that shape.
A PowerPoint file has to commit to a size before anything is on it
Open a .pptx and you will find it is a zip file full of XML. One of those files, presentation.xml, contains a single element called sldSz that sets the size of every slide in the deck. It carries two numbers, a width and a height, and in a deck made by most converters they read 12192000 and 6858000.
Those are English Metric Units, 12,700 to a point, and they work out to 13.333 by 7.5 inches: standard widescreen. That element has to be written before any slide content is added, because every shape and every text box on every slide is positioned relative to it.
So a converter has to answer "how big is a slide?" as its first question, before it has looked at a single page of your PDF. Answering "13.33 by 7.5" is free, it requires no thought, and it produces something that unmistakably looks like PowerPoint. Answering "whatever shape the pages actually are" means reading the file first and then building around the answer.
Most converters take the free option. That is the whole explanation for the empty bands.
Your PDF knows exactly how big it is
Here is what makes the shortcut unnecessary. Every page in a PDF carries its own MediaBox: an explicit rectangle, in points, saying how large that page is. A4 is 595 by 842. US Letter is 612 by 792. A tall newsletter might be 480 by 906.
That number is not inferred, estimated or guessed at. It is written in the file, it is the first thing any PDF reader looks at, and it is available before a single character of text has been extracted. A converter that puts your A4 page on a widescreen slide did not fail to work out the page size. It chose not to ask.
Read it first and set sldSz from it, and an A4 report becomes an A4 slide. The deck looks like the document you started with, because it is the shape of the document you started with.
What the format genuinely will not let you do
Being straight about the limits matters as much as fixing the easy part, because there are two real constraints here and they are not the converter being lazy.
One slide size per file. That sldSz element appears once, in presentation.xml, and applies to the whole presentation. There is no per-slide override. So if your PDF mixes portrait and landscape pages, no converter and no amount of cleverness can give both their native shape in a single .pptx. Something has to give: the first page sets the size, and pages of a different shape are scaled to fit inside it and centred, with their proportions kept. If that matters for your document, split the PDF by orientation and convert each part separately. A converter that does not mention this is hoping you will not notice.
Fifty-six inches. PowerPoint will not open a slide larger than 56 inches on a side. Posters, plans and long scrolling pages can easily exceed that, so they have to be scaled down. Keeping the proportions means the deck still looks right; it is just smaller than life size.
What else survives the trip, and what does not
Slide size is the most visible problem but it is not the only one worth asking about.
Text should be text. Each line of the PDF can become a real PowerPoint text box at the position it had, keeping its size, its bold and italic and its colour. You click into it and type. What you do not get is reflow: because every line is positioned separately, typing a long sentence into one box will not push the lines below it down the slide. That is the honest trade for a layout that matches the page, and it is the right trade for a brochure or a report. If you intend to rewrite the thing from scratch, you were always going to retype it.
Colour is harder than it looks. The obvious way to find the colour of a piece of text is to look at the darkest pixel around it. On a white page that works. On the navy banner across the top of a newsletter it is exactly backwards: the darkest thing in that area is the background, so white heading text gets painted navy on navy and vanishes. Finding the colour that covers most of the area and calling that the paper, then taking the ink as the colour furthest from it, handles both cases.
A circle is not its bounding box. Vector shapes in a PDF are paths, and the lazy way to redraw one is to fill the rectangle it occupies. Do that and every round play button, numbered badge and avatar in your document turns into a square. Measuring how close the outline comes to the corners of its box tells you whether you were looking at a circle, a rounded card or a square corner, and PowerPoint has a preset shape for each.
Letter-spaced headings lose their word gaps. This one cannot be fixed by anybody. When a heading is tracked out wide, as mastheads and eyebrow text usually are, a PDF records the gap between two spaced-out words in exactly the same way as the gap between two spaced-out letters. By the time any program reads the file, WATCH EPISODE 12 and WATCHEPISODE12 are indistinguishable. Copy such a heading out of Acrobat or Preview and you will get the same thing. The honest response is to say so and let you add the spaces back, which takes seconds.
A scan has no text in it to convert. If a page was scanned, photocopied or photographed, it is a picture of a page and there is nothing to extract. The useful thing to do is make that slide a picture of the page, which at least looks right, and say plainly that it happened rather than handing back a blank slide. If you need the words out of a scan, that is a job for OCR and for the PDF to Word Converter, which reads scanned pages in twelve languages.
If you actually wanted widescreen
Sometimes you do. You are putting the pages into an existing deck, or presenting on a projector, and the shape of the original document is beside the point.
That direction is easy. In PowerPoint, go to Design, then Slide Size, and choose Widescreen; in Google Slides it is File, then Page setup. Your content gets repositioned, which takes a little tidying, but the option is there and it takes a few seconds.
The reverse is not easy. Once a converter has flattened your A4 report into the middle of a widescreen slide, getting the page shape back means resizing the slide and then moving every object on every slide by hand. That asymmetry is the argument for converting to the real page size and letting you change it afterwards if you want to, rather than the other way round.
What to check in any converter
Three questions will tell you most of what you need to know about a PDF to PowerPoint tool.
Does it tell you the slide size it is going to use before you convert, or do you find out afterwards? Does it say anything about pages it could not handle properly, such as scans or pages of a different shape? And does it ask you to upload your document to a server you know nothing about, for a job that modern browsers are perfectly capable of doing on your own machine?
The documents people convert to slides are pitch decks, brochures, internal newsletters and reports. Those are rarely things you want to hand to a stranger in exchange for a download link.
Our PDF to PowerPoint Converter runs entirely in your browser, shows you the page size as soon as you choose a file, and tells you about anything it could not carry across. Nothing is uploaded, nothing is stored, and there is no watermark, no account and no limit on how many files you convert.
Ready to turn your pages into slides?