How it works
A timetable, a small batch and check-ins that arrive at exactly the right moment.
1. Pick a class
Each class follows one year group, from kindergarten to 12th grade (Year 1 to Year 13). The syllabus uses Khan Academy's grade-level courses for maths and English, with science and history from their middle and high school courses. Classes start on set dates and run for a term, six months or a year.
2. Join a batch
Classes are split into batches of up to 12. Start one and send the invite code to friends, or join a batch with space. Batchmates follow the same timetable and can see each other's progress.
3. Study parties, live
Each 40-minute period is a study party: a scheduled live lesson where the batch watches a short playlist on one topic in sync. Everyone starts together, and some parties are led by a teacher or guide. The video pauses for check-ins at moments chosen from the narration, and everyone gets the same time to answer:
- Predict: commit to an answer just before the video reveals it.
- Try it first: solve a problem like the one on screen before the narrator does.
- Quick checks: multiple choice, true or false, or pick all that apply. Every wrong answer gets its own explanation.
- Your own words: short answers and open "think about it" questions, with a model answer to compare.
- Notes and tips: key ideas and common mistakes, shown without stopping the video.
After each video come recap questions and a question for the batch to talk about. Then it's practice on Khan Academy.
4. Catch up any time
Missed a party? Catch-up mode plays the same videos and check-ins at your own pace. The video waits for each answer.
5. Parents stay in the loop
Parents see parties done and missed, progress by subject, what's next and who is in the batch, for every child on one account.
Why this works
Most online courses are watched alone — and mostly left unfinished. Large studies of self-paced online courses put completion in the low single digits (about 3% in Reich & Ruipérez-Valiente's 2019 analysis of millions of learners). The problem isn't the videos; it's doing it alone. Every design choice in a study party answers that, and each one rests on a well-established finding from learning research.
Learn together
A study party puts a batch of up to 12 on one shared clock: everyone watches, answers and sees the reveal at the same moment, with batchmates who notice whether you came. People work harder and keep going longer when others are present and expecting them — one of psychology's oldest results — and feeling that you belong to the group is a strong predictor of sticking with hard study (Walton & Cohen, 2011). It's why study groups and cohort courses finish at rates self-paced courses never reach.
A teacher in the loop
Bloom (1984) found that students taught with tutoring and mastery methods performed about two standard deviations above a conventional class — the famous "2 sigma" result, and exactly the ingredient online courses drop first. A study party keeps the teacher: they can pause everyone, talk, see raised hands and watch how many have answered. The AI helper works for the teacher, not instead of them — instant help the moment a student is stuck, with every question flagged to the teacher's inbox.
Practice that's proven
The check-ins inside every video are retrieval practice: having to recall something beats re-watching it, a result so consistent it has a name — the testing effect (Roediger & Karpicke, 2006). The rotating timetable and Catch up spread each topic out over time, which reliably beats cramming (Cepeda et al., 2006). And the reveal after every check-in and the tally at the end of each party are immediate feedback — among the strongest influences on learning we know of (Hattie & Timperley, 2007).
This is a plain-words summary of published research, not a claim that any of these studies tested Puzzicle itself — we built the school on these findings. The originals: Bloom (1984), Educational Researcher; Roediger & Karpicke (2006), Psychological Science; Cepeda, Pashler, Vul, Wixted & Rohrer (2006), Psychological Bulletin; Walton & Cohen (2011), Science; Reich & Ruipérez-Valiente (2019), Science; Hattie & Timperley (2007), Review of Educational Research.
Ready to start?
Puzzicle is free to run.