Shrink the review until it survives a busy Sunday.
The best review is the one that actually happens. Most review systems fail because they are designed for a calm Sunday that never arrives, so the useful version is the one that survives a busy week: five minutes, four questions, no writing beyond three lines.
The four questions are these. Which round do I remember best? Which error type appeared most? Which lane did I not meet? What will I listen for next week? The answers can be one word each. The point is not detail; the point is that the loop closes.
The memory question is the interesting one. The round you remember is usually the round that taught the most, and it is worth one extra minute of reflection. The error-type question is the most actionable. The coverage question protects against narrowness. The final question sets next week's intention.
Three lines, then stop. If the review runs long, it will be skipped next week, and a skipped review breaks the loop. Five minutes is a limit, not a target.
I run the five-minute version on most weeks and the longer version about once a month. The monthly version reads the whole notebook; the weekly version reads only the week. Both use the same questions, which makes the monthly version a summary rather than a new system.
Start the loop this week on the practice room with three rounds and the four questions. It is enough to see how the loop feels, and if it feels easy, it will survive.
read the hints after committing, as always, so the hints teach rather than spoil.
A review habit that fits in five minutes is a habit that will still exist next year. Five minutes is short enough to repeat and long enough to change something. That is the whole design brief, and it beats every ambitious system I have tried.