Do flashcards actually work, or does it just feel productive?

They work, and the reason is the retrieval rather than the card. Pulling an answer out of your head strengthens the memory far more than reading the same answer again, an effect measured repeatedly since Roediger and Karpicke published on it in 2006. Rereading notes feels more productive because the material feels familiar, which is exactly the illusion the test is measuring. A flashcard is simply the cheapest way to force a retrieval, which is why the index card has outlived every app built on it.

How many cards should a deck have?

Small enough that a full pass takes under twenty minutes, which in practice is 50 to 150 cards. Past that, the cards you already know eat the session and the ones you do not get five seconds each. Split a big subject by topic instead of building one enormous deck. Twenty cards you review six times beats two hundred you review once.

Should the answer side have a picture on it?

For anatomy, chemistry, maps and any vocabulary with a concrete referent, yes: a picture gives the memory a second route in and is often faster to recognise than a sentence. For an abstract definition it usually adds nothing but decoration and cost. The honest test is whether you could draw the picture yourself. If you could, it is doing work; if you found it because the card looked empty, it is not.

Are paper or digital flashcards better?

Paper wins on writing, digital wins on reviewing. Writing a card by hand is slower, and the slowness is the useful part: you have to decide what the card is actually asking. Digital wins the moment the deck is bigger than about fifty cards, because software can put the ones you keep missing in front of you and a stack of card cannot. Plenty of people do both, drafting on paper and typing the keepers in.

How often should I review a deck?

Daily while the material is new, then let the spacing stretch. The evidence for spacing over cramming is about as settled as anything in learning research, and the exact intervals matter far less than the habit. A short session every day beats a long one every week, and a deck reviewed twice at ten minutes beats the same deck reviewed once at twenty.

Is Quizlet still free?

Partly. Making and studying a set is free, and it always has been. Learn mode, test mode and studying offline moved behind Quizlet Plus, and a free account sees adverts. You also need an account to save anything, and your sets sit on their servers. That is a fair trade for the sync and the shared library, and it is worth knowing you are making it.

Can ChatGPT make flashcards for me?

Yes, and the useful prompt is a boring one. Ask for one card per line, term first, separated by a tab, and nothing else in the reply. You get a list you can paste into any tool that takes a plain list, including this one. Check the cards before you learn them: a model will produce a confident, wrong definition in exactly the same format as a right one, and a flashcard is the worst possible place to put an error, because you are about to memorise it.

How to make flashcards that actually work

Six ways to make them compared, the exact table settings for Word and Google Docs, and the four rules that decide whether a card is worth writing at all.

Free toolFlashcard MakerPaste a list to build a deck, study it with spaced repetition, and print it double-sided. Your deck stays in your own browser.Open the flashcard maker

The short answer

Write one question per card, then review it by answering out loud before you turn it over. Everything else is logistics. The tool you choose changes how fast you build the deck and how well the reviews are scheduled, and it changes nothing about whether the cards teach you anything. That is decided by how they are written.

If you want the fastest route: get your terms into a plain list, one per line, with the term first and the answer after a dash or a tab. Every tool below imports that shape, including the browser one on this site.

The four rules for a card worth making

One fact per card. A card holding three things is a card you will get two-thirds right forever, because you can never mark it wrong without also re-learning the parts you knew. Split it.
Ask a question, do not name a topic. "The French Revolution" is not a card. "In what year did the Bastille fall?" is. If the front does not end in a question mark or read as one, it is a heading.
Make it produce, not recognise. Recognising the right answer among four is far easier than producing it from nothing, and only one of those is what an exam or a conversation asks for. This is also why multiple-choice practice flatters you.
Write it in your own words. A definition copied out of a textbook is a sentence you can recite without understanding. Rewriting it is where most of the learning in making a deck actually happens, and it is the reason a deck someone else made is worth less than the one you built.

A useful test for a finished card: could a friend read the front aloud and mark your answer right or wrong without having to interpret anything? If not, the card is still vague.

Six ways to make them

Method	Best at	Costs you
Index cards and a pen	Thinking. Writing by hand forces you to decide what the card asks.	Time, and no scheduling. A stack cannot know which cards you keep missing.
Microsoft Word	Printing on a specific card size, with full control of the layout.	Fiddly table work, and you lay out the answer page by hand.
Google Docs	The same job, shared with a study group.	Fewer layout controls than Word, and the same manual answer page.
Quizlet	Finding a deck someone has already made for your course.	An account, adverts, and a paid tier for Learn, test and offline.
Anki	Long courses. The scheduling is the best there is, by a distance.	A steep first hour, and a desktop install to get the full thing.
A browser tool	Starting in thirty seconds with no account, and printing properly.	No sync between devices, so moving the deck is a file you save.

The split most people actually want is this. For a term-long subject you will revisit for months, learn Anki and accept the first hour. For a test in two weeks, a vocabulary list, or anything you want on paper, the setup cost of Anki is the whole job again, and a browser tool or a printed sheet gets you studying immediately.

How to make flashcards in Microsoft Word

Word has no flashcard feature. What it has is tables, and a table with tall rows is a sheet of cards. On a portrait A4 or Letter page:

Insert then Table, and pick 2 columns by 4 rows. That is eight cards a page at roughly index-card size.
Select the table, open Table Properties, then Row, and set Specify height to 6 cm with the row height set to Exactly. Without "Exactly" the rows collapse to fit the text and your cards come out different sizes.
On the Table Design tab set the borders to a thin solid line, so there is something for the scissors to follow.
Type one term per cell. Centre the text vertically from Alignment, or every card reads from the top edge.
Copy the page, paste it below, and replace the terms with their answers. Then swap the two columns over. See the section on printing below for why.

Word can also print onto pre-cut index card stock through Mailings, then Labels, then Options, where several 3 by 5 inch card stocks are listed. That is neater than cutting, and it costs more per card than a sheet of paper.

How to make flashcards in Google Docs

The same shape, with fewer controls. Insert, then Table, and drag out 2 by 4. Click into the table, open Format, then Table, then Table properties, and set a minimum row height of about 6 cm. Docs has no exact row height, only a minimum, so a long answer will push its row taller and shunt the rest of the page. Keep the answers short or accept that the cards vary.

Vertical alignment lives in the same panel, under Cell. Set it to Middle. Then duplicate the page for the answers and swap the columns, exactly as in Word.

Google Slides is worth knowing about as the other half of this: one slide per card, speaker notes for the answer, and presenter view shows you one side at a time. It is a serviceable study mode built out of parts that were meant for something else.

Printing on both sides without misaligning everything

This is where nearly every printable template on the web falls over, and the fix is one sentence: the answer page has to be mirrored left to right.

Turning a sheet over on its long edge, which is what "flip on long edge" means in a print dialogue and what your hands do naturally, reverses the columns and leaves the rows alone. So the card printed at the top left of the front page ends up backing onto the top right of the reverse. Print the answers in reading order and every card on the sheet gets the wrong answer, in a way that looks fine until you cut them out.

Two things make this survivable. Number both sides of every card, so a misalignment is obvious rather than silent. And print a single test sheet before you commit forty. If you would rather not think about it at all, the tool on this site mirrors the answer page for you and numbers both sides.

How often to review

Spacing beats cramming, and the effect is large enough that the exact intervals barely matter next to the habit. A workable schedule with no software at all is the Leitner system: five boxes, a card you get right moves up a box, a card you get wrong goes back to the first, and you review box 1 every day, box 2 every other day, and so on out to weekly.

That is what the browser tool here implements, and it is a deliberate simplification. Anki computes an interval per card from your whole review history, which is genuinely better and genuinely needs a database. Five boxes need nothing but the boxes, which is why the paper version has outlasted most of the software.

Word and character counterCheck a summary or an essay against a limit, counted as you type.CSV to JSON converterReshape a spreadsheet of terms before you turn it into a deck.