What types of information Flora works best on

Most learning apps would rather not tell you what they’re bad at. We think it’s the most useful thing we can tell you, because the people who get the most out of Flora are the ones who point it at the right sort of thing — and the people who bounce off it are usually the ones who pointed it at the wrong sort and concluded the whole idea doesn’t work.

So: Flora is built on two findings, and those two findings have a shape. Here’s where the line falls.

What it’s very good at

Discrete facts. Things that have an answer, where the answer is short, and where knowing it quickly is the point:

• Vocabulary in a language you’re learning
• Anatomy, drug names, taxonomies, the parts of a system
• Dates, names, and who did what to whom
• Definitions and terminology in a field you’re new to
• The names of everyone at your new job

This isn’t a modest claim. When researchers reviewed ten common study techniques for Psychological Science in the Public Interest, only two came out rated “high utility”: spacing your practice out over time, and testing yourself instead of reviewing. Highlighting and rereading — the two things nearly everyone actually does — came out bottom.

For facts like these, spaced retrieval is genuinely among the best-evidenced methods there are. It isn’t a hack or a productivity fashion; it’s a century of converging evidence, starting with Ebbinghaus in 1885 and still holding up in present-day classroom studies.

What it won’t do

It won’t hand you an understanding of a subject.

The evidence that spaced retrieval improves conceptual understanding, reasoning, or transfer to new problems is thin and mixed. One large meta-analysis found the spacing benefit was strong for simple subject matter and disappeared entirely for genuinely complex tasks. Recalling a definition of “opportunity cost” on demand is not the same as noticing one in the wild, and no amount of reviewing the card will carry you across that gap.

It also won’t make you fluent in a language. Recall of a word pair isn’t the same faculty as understanding speech in real time or finding a word while you’re talking. Vocabulary is genuinely useful scaffolding, and it is scaffolding.

And it can’t do much about the atomisation problem. Break a subject into small enough pieces and you can end up able to recall every fragment while being unable to assemble them into anything. Andy Matuschak, who has thought about this more carefully than most, names it directly: the biggest problem with flashcards is cards becoming disconnected from each other.

How to get the most out of it

Point it at the layer of specifics underneath the understanding. The facts are the part that quietly goes missing while the gist stays. You’ll still remember roughly what the paper argued; you’ll have lost the number, the mechanism, and the name of the author. That’s the layer worth defending.

Read the thing properly first. Flora is not a substitute for reading — it’s what stops the reading evaporating. Cards made from something you never really understood are just noise you’ve committed to reviewing.

Edit the cards. We generate them because writing good ones is slow and genuinely skilled work, and that difficulty is the main reason people give up on spaced repetition before it pays off. But you know what you actually want to keep. Delete the ones that don’t matter. Rewrite the ones that ask you to recognise something when they should ask you to retrieve it.

Then just keep turning up. This is the part that decides everything. Across the research, the difference between people who end up remembering and people who don’t is overwhelmingly not which scheduling algorithm they used — it’s whether they kept going. A well-scheduled app you abandon loses to a rough one you open. That’s why Flora is built to make the days short: a queue that shrinks as you learn is the only feature that addresses the thing that actually goes wrong.

The honest summary

If you want to durably know a large number of specific things, spaced repetition is the best-evidenced tool available, and Flora is a straightforward way to use it without making the cards by hand.

If you want to understand a field, argue about it, or speak a language, you’ll need other things too — reading, writing, problem-solving, conversation. Flora will hold the facts steady underneath while you do.