Most study apps show you a feature list. Very few show you what happens on a Tuesday night when you have four hundred cards due, a lecture PDF you never opened, and about ninety minutes before you give up.
So here is the honest version. This is what each part of GoodOff does, why it exists, and how the pieces talk to each other.
The dashboard is a decision, not a summary
Open GoodOff and the first thing you see is a number. Cards due today. Not a motivational quote, not a chart of your week, just the size of the pile.
That is intentional. The hardest part of studying is rarely the studying. It is the ten minutes you spend deciding what to study, which usually ends with you reopening the notes you already know well because they feel comfortable.
Under that number are three things worth understanding:
Due today is what the scheduler thinks you are about to forget. It is not everything you have ever made. It is the subset that is sitting at the edge of your memory right now.
Retention is your actual recall rate on reviews you have already done. It is a feedback signal. If it drops, the scheduler tightens up. If it climbs, intervals stretch out and you see those cards less often.
Streak is the least clever metric on the page and probably the most useful one. Consistency beats intensity with spaced repetition, and the streak is there to make skipping feel slightly annoying.
The weekly strip below it shows which days you actually showed up. No commentary, just squares. You can read your own pattern in about two seconds.
Then there is the button. Start study session. One click and the app picks the order for you. If you want to override that, Pick a subset lets you narrow down to a single deck or topic, but the default assumes you would rather start than plan.
FSRS is the part doing the quiet work
Behind the dashboard is FSRS, a modern spaced repetition scheduler. Here is the short version of what it does.
Every time you review a card, you tell the app how it went. FSRS uses that to model three things about that specific card: how likely you are to recall it right now, how stable the memory is, and how difficult that card is for you personally.
From those, it picks the next review date. Cards you find easy drift further and further apart. Cards you keep fumbling come back tomorrow. You never schedule anything yourself.
The practical effect is that your review load stops growing linearly with your library. You can have thousands of cards and still have a finite daily queue, because most of them are quietly sitting on long intervals waiting their turn.
This is also why the "By Subject" breakdown on the dashboard matters. It shows due count, total cards, and accuracy per subject, so you can see which topic is actually dragging. If one subject has a low accuracy percentage and a large due pile, that is your evening.
Quick Capture: getting material in without friction
The dashboard has a box that asks what you read today. You can paste a link, upload a PDF, drop in an image, or record audio.
This exists because of a specific failure mode. You find a good article, you think "I should make cards from this," and you never do. The gap between finding material and turning it into something reviewable is where most study systems die.
Quick Capture closes that gap to about a minute. Paste the URL, and Sage drafts a deck from it. Upload the lecture slides and you get cards. Record a voice note walking out of class and it becomes material you can actually revisit.
You still review what it generated. You should. But the blank-page problem is gone, and that is usually the whole battle.
The Library is where sources become things
The Library page is a running inventory of everything you have added and everything it turned into. Sources, decks, quizzes, podcasts.
The distinction matters. A source is raw input: the PDF, the article, the pasted notes. A deck is what you review. A quiz is what tests you. A podcast is that same material converted to audio you can listen to while doing something else.
One source can become all of them. And on the Library page, any source that has not been converted yet shows a "Generate study set" link right on its card. That is your backlog, visible at a glance.
Items group themselves by recency: earlier this week, this month, older. Search and sort sit at the top. It is a boring page by design, because the job of a library is to let you find one specific thing without thinking about it.
The counters at the top (sources, decks, quizzes, podcasts) are there for a slightly sneaky reason. Seeing "111 sources, 85 decks" tells you something real about your own habits. If your source count runs far ahead of your deck count, you are collecting rather than studying.
Sage Tutor: an AI that read your material, not the internet
This is the feature most people misunderstand at first, so here is the distinction clearly.
A general chatbot answers from everything it has ever seen. Ask it about a concept from your lecture and it will give you a competent, generic answer that may or may not match what your professor said, what your syllabus covers, or how your exam will phrase it.
Sage answers from your sources. You pick a document from your library, start a tutor session, and every answer comes back grounded in that material with citations pointing to where it came from.
That changes what you can ask. Instead of "explain the cardiovascular system," you can ask "what does my lecture say about preload versus afterload, and where in the slides is that." Instead of a summary you cannot verify, you get an answer you can trace back to page four.
It also handles the awkward questions you would not ask in a lecture hall. The ones where you have to admit you did not follow the last twenty minutes. Sage does not care, and it will keep rephrasing until something lands.
Practical use: after a lecture, start a tutor session on the slides and ask it to quiz you on the parts you flagged. Before an exam, ask it what you seem to be weakest on based on the material. It is a study partner that has actually done the reading.
Visual Explanations: for the concepts that refuse to sit still in text
Some things do not survive being written down as bullet points. Processes with branches. Systems with feedback loops. Anything where the relationships matter more than the definitions.
Visual Explanations turns a concept into a diagram. You get four formats, and each one is suited to a different shape of information:
Flowcharts for processes with decisions and branches. Anything where "if this, then that" is the actual content.
Mind maps for a central concept that fans out into subtopics. Good for structuring a chapter or seeing how one idea connects to six others.
Timelines for anything that happens in sequence. Historical development, procedural steps, the evolution of a field.
Diagrams for relationships and systems. Components, connections, and the direction things flow.
You type a concept, pick a format, and get something you can download and keep. The point is not decoration. Visual and verbal encoding create separate retrieval paths in memory, which is a fancy way of saying you get two chances to remember instead of one.
The saved visuals live on their own page, labelled by type and date, so you can pull the same diagram back up the night before an exam.
Pomodoro: the timer that actually knows what you studied
Nearly every study app has a timer bolted on. This one is wired into the rest of the platform, which makes it worth using rather than reaching for your phone.
The core is standard: a focus block, a short break, a long break. Twenty-five minutes by default, with cycle lengths you can change. Start, and the timer floats in the corner so you can carry it across pages while you review cards or talk to Sage.
What is less standard is Focus Radar underneath. It tracks session progress and interruptions live. That second number is the interesting one. Every time you break focus, it goes up. Nobody is grading you, but after a few sessions you start to notice which times of day and which environments produce a clean run and which ones produce eleven interruptions.
There is a Deep Focus mode for when you need the rest of the interface out of the way, and a history view so the sessions accumulate into something you can look back at.
How the pieces fit together
Individually, none of these features is unusual. Together they form one loop, and the loop is the actual product.
You capture something (a PDF, a link, a voice note). It becomes a deck, a quiz, a podcast, or all three. FSRS decides when you see each card. Sage explains whatever the cards expose as weak, using your own material. Visual Explanations handles the concepts that need a shape. Pomodoro holds the session together. The dashboard tells you where to start tomorrow.
The thing that breaks most study systems is friction between steps. You have notes in one place, flashcards in another, a chatbot in a third tab, and a timer on your phone. Every handoff costs you a little willpower, and willpower is the resource you have least of at 11pm.
Closing those gaps is most of what GoodOff is trying to do.
If you want to see the loop on your own material, start with a single lecture PDF at goodoff.co and watch what it becomes.
Top comments (0)