A recommendation is a small act of teaching.
Most engines optimise for what you'll consume next. BookSeeking optimises for what would help you grow — and that distinction comes straight from the research on how curiosity and learning actually work.
Curiosity is the feeling of a gap between what you know and what you want to know — and it sharpens as you near the edge of your own knowledge.
Every concept BookSeeking extracts is sized to be a recognisable gap: specific enough to notice when you've filled it, general enough to recur across books. Recommendations aim at the gaps nearest your sky.
We learn most from material that's mostly comprehensible with a little that's new — your current level, plus one.
We rank books by how much of their content you already hold. The sweet spot is roughly 50–80% familiar: enough footing to follow, enough new to grow.
Absorption happens when challenge meets skill — and as skill grows, the challenge has to rise to match it.
As your library grows, BookSeeking quietly tightens its band: a new reader gets variety, a seasoned one gets depth. The challenge tracks your skill.
More options can mean fewer decisions. In the famous jam study, a smaller display drove far more choices than a larger one.
We show eight recommendations by default, not eighty — enough variety to find a match, few enough to actually decide.
What readers love isn't only a book's subject but how it reads — its pacing, characterisation, story line, frame and style.
BookSeeking profiles the texture you enjoy and weights matches toward it, so a contemplative reader isn't handed a brisk explainer of the very same idea.
Engines tuned purely for engagement narrow what you read. Tuned for development, they widen it. We chose the second on purpose — and every number in the ranking serves it.