First, decide what deserves a card
The most common mistake is trying to turn a whole PDF into cards. A 40-page lecture PDF contains maybe 30–80 facts worth remembering, and the rest is context, examples and filler. Skim the PDF first and mark what you would actually be tested on. Every method below works better when you start from that shortlist.
Method 1: Make the cards by hand (copy, paste, cloze)
This is the classic way, and it gives you total control.
- Open Anki and click Add.
- Set the note type to Cloze.
- Copy a sentence from the PDF and paste it into the text field.
- Select the key term and press Ctrl+Shift+C (Cmd+Shift+C on a Mac). Anki wraps it as
{{c1::term}}. - Click Add and repeat.
Good for: small amounts of material and people who like to think while they type. Not so good for: long PDFs. Text copied from a PDF often arrives with broken line breaks, and the time adds up fast.
Method 2: Import from a text or CSV file
Anki can import plain-text files where each line is a card and the fields are separated by tabs or commas. So you can extract the text of the PDF, reshape it into one card per line, and import it with File → Import. Anki lets you choose the note type and map each column to a field.
Good for: material that's already structured, like vocabulary lists or tables. Not so good for: normal lecture notes, because a PDF's text isn't organized into question/answer pairs, so you still have to write the cards yourself before importing.
Method 3: Use an AI generator
An AI tool reads the text and writes the cards for you. You still need to review them, because AI can misread a sentence or hide the wrong word. With Citecards the flow looks like this:
- Upload your PDF (DOCX, PPTX and TXT also work, or paste text) and pick the pages you want.
- Click generate. Citecards writes cloze cards from the text.
- Review each card next to the original page, with the source passage highlighted. Click a card to jump to where it came from.
- Edit or delete anything: the card text, the extra notes, the tags and even the order. You can also select a passage in the PDF and generate a card from just that text.
- Export an .apkg file and open it in Anki (double-click it, or use File → Import).
Good for: long notes and tight schedules. Honest limits: it works with text-based files, so scanned pages and images aren't read yet, and complex math formulas may not come through perfectly. Always check the cards before you study them.
Which method should you use?
| Method | Time | Control | Best for |
|---|---|---|---|
| By hand | Slowest | Total | Short material, careful learners |
| Text / CSV import | Medium | High | Structured lists and tables |
| AI generator | Fastest | High, if you review | Long lecture notes |
Many people mix them: generate a first draft with AI, then rewrite the cards they care about most by hand.
Tips for better cards, whatever method you use
- One fact per card. If a card needs three blanks to make sense, split it.
- Hide the key term, not a random word. The blank should be the thing you need to recall.
- Check numbers, dosages and directions. These are easy to get subtly wrong, whether a person or an AI wrote the card.
- Keep a link to the source. When a card confuses you a week later, you'll want to find the original sentence quickly.
- Tag by lecture or chapter so you can study or clean up a whole batch at once.
Frequently asked questions
Can Anki import a PDF directly?
No. Anki imports its own deck files (.apkg) and text files such as CSV, so a PDF has to be converted first.
Does it work with scanned PDFs?
Not yet with Citecards. It reads the text inside the file, so a scan that is only a picture of a page can't be read. A PDF exported from Word, Google Docs or PowerPoint works.
Is Citecards free?
You can try it for free: the free plan covers 3 files and up to 50 cards a day, and you don't need an account to generate and review your cards. You add your email when you export.