What to do with scanned PDFs and photos of slides
Some PDFs on your course page are really just pictures of pages. They look normal on screen, but there's no actual text inside for CourseFeed to pull out.
Is it a scan?
Open the PDF and try selecting a sentence. If you can highlight individual words, there's text in it. If you can only drag a box over the whole page, it's an image.
Searching works too. Press Ctrl+F (Cmd+F on a Mac) and look for a word you can see on the page. No match usually means a scan. Sometimes you can tell just by looking, from a crooked page or the dark shadow where a book was pressed against the glass.
Library readings, photocopied chapters and handwritten notes are nearly always scans. Slides exported from PowerPoint or Google Slides almost never are.
Why CourseFeed says "No text found"
CourseFeed reads text that's already stored in a file. It doesn't do OCR (optical character recognition, which means reading letters out of an image). When a file has no stored text, CourseFeed marks it "No text found" and leaves it out, so you know right away that it's missing.
Watch out for PDFs that are only partly scanned: typed pages with a few scanned ones mixed in. Those still convert, but the scanned pages won't be in the result. If a section seems short, compare it with the original.
Getting the text out
Ask for the original
If it's a scan of slides or a handout your instructor made, the PowerPoint or Word file exists somewhere. Check the course page again, or email and ask. Nothing gets misread this way, so it's worth trying first.
Give the scan to the chatbot
ChatGPT, Claude and Gemini can read images and most scanned PDFs on their own. Attach the scan in the same chat as your CourseFeed notes. Short scans work better. A 40-page reading may run into the chatbot's file limits.
Copy text with your phone
Recent iPhones (Live Text) and Android phones (Google Lens) can copy text out of a photo or a page on your screen. It's handy for a page or two, like a definition or the labels on a diagram.
Run OCR on the PDF
Some PDF apps, including free ones, can add a text layer to a scan. Look for an option called "OCR" or "Recognize text", save the new file, and drop it into CourseFeed. Then check it over. OCR gets letters wrong, and math and tables are where it slips the most.
Handwritten notes
Handwriting is the hardest case. Chatbots can often read neat handwriting from a photo, but go through what they give back carefully. Typing your notes up yourself takes longer, though you'll remember more of them.