A bank is organised by chapter, a weakness is not
The thing you are weak at is "splitting the middle term when the leading coefficient is not 1". No question bank has a folder for that. Generation can aim at it directly.
Practice on NYGTLYF starts from a named sub skill and produces questions against it. That is a different operation from opening a chapter exercise, and it is the reason a fifteen minute set can move a rung that three hours of chapter revision did not.
Note that the rungs here are sub skills inside one chapter, not chapters.
None of these are about volume. A large bank and a generator solve genuinely different problems.
The thing you are weak at is "splitting the middle term when the leading coefficient is not 1". No question bank has a folder for that. Generation can aim at it directly.
A fixed bank gets learned as a bank. Students start recognising question 47 rather than solving it. Fresh generation removes recognition as a strategy.
Adjusting level inside a session needs a question that does not exist yet at the level you just reached. Pre written sets cannot do that without huge redundancy.
This set was produced after the doubt on splitting the middle term. Each question exists to catch a specific error, and they are ordered so that the set itself acts as a diagnosis rather than a drill.
Factorise: 2x² + 7x + 3
Positive pair, small numbers. Confirms the a x c step is understood at all.
Factorise: 3x² - 10x + 8
Both factors negative. Catches students who only try positive pairs.
Factorise: 6x² - 5x - 6
Composite leading coefficient with a mixed sign pair. This is where most sign errors surface.
The area of a rectangle is 2x² + x - 15. Find expressions for its sides.
Same move, wrapped in a word problem, so the skill has to be recognised rather than prompted.
Marking only the final value is how a ladder fills up with rungs that are mint on paper and coral in the exam hall.
Where you show steps, the steps are checked. A correct root reached by trial and error is recorded as trial and error, and it does not count towards mastery of factorisation.
A sign error, a concept error and an arithmetic slip are three different signals. Only the concept error reopens a rung. The other two change what the next question looks like.
Getting one wrong opens the same step by step treatment the tutor gives a doubt, built around the specific error you made rather than the model solution.
Depth is not uniform and pretending otherwise would waste your term. This is the current state, and it is the page we update when that changes.
Full ladder, full generated practice, adaptive difficulty across all four levels.
Full ladder and generated numericals. Derivation heavy topics lean more on explanation.
Full ladder. Physical chemistry generation is stronger than organic mechanism generation today.
Full ladder, generated practice weighted towards diagrams, definitions and process ordering.
Ladder available as prerequisite support, mainly reached from above when a gap is found.
Explanation is strong. The ladder is partial and generated practice is limited to grammar.
Explanation and doubt solving only for now. No ladder and no generated practice yet.
CBSE and ICSE are matched most closely. Major state boards are supported with occasional ordering differences.
Generated questions are checked against the syllabus scope for your class and board before they are shown, so practice does not drift into material you have not been taught.
The difference is not effort or discipline. It is whether the questions in front of you are the ones you cannot yet do. Start with one subject on the free tier and see what a targeted set feels like.
Free tier stays free. One subject, a daily doubt limit, no card.