One hard book abandoned on page 12 teaches you less Italian than five easy books finished. Reading research has been making this case for decades, and it still loses arguments to pride. Learners pick up Elena Ferrante at A2 because reading a “real” book feels like progress, grind through two chapters at dictionary speed, and quietly stop reading in Italian at all.
Extensive reading is the alternative with the evidence behind it: read a large amount of material that is easy for you, chosen by you, for the story rather than for study. Day and Bamford’s classic principles put it plainly: the reading material should be well within the reader’s competence, and the reward is the reading itself. Paul Nation’s vocabulary research supplies the engine underneath: words are mostly learned incrementally, across many meetings in context, and only high volume at high comprehension delivers enough meetings.
The numbers that make it work
The key figure comes from coverage research (Hu and Nation, 2000): comfortable comprehension needs you to already know around 98% of the words on the page, roughly one unknown word in fifty. Below that, guessing from context breaks down, and with it the learning. This is why “just read native books, you’ll get used to it” fails as advice: at 85% coverage you are not reading, you are excavating. More on the vocabulary side in how many words you actually need.
So the practical rule from the comprehensible input article applies doubled: for extensive reading, err easy. A page with 2–5 unknown words is perfect. A book you finish in a week beats a book you respect for a year.
Graded readers are the technology
A graded reader is a real book rewritten to a level: the vocabulary controlled, the sentences shortened, the story intact. It is the only reliable way to hit 98% coverage before B2, which is why I keep recommending them and why I wrote a whole guide to which Italian classic works at which level and another on why they beat dual-language editions, where the translation on the facing page quietly does your reading for you.
This is also the heart of 
amo Italian: 25 Italian classics (Deledda, Pirandello, Svevo, De Marchi) rewritten to fixed levels, with the vocabulary of every book classified against CEFR word lists so the level is engineered rather than guessed. Each book carries a hundred footnotes and a glossary, so the occasional unknown word costs a tap instead of a dictionary trip, and three questions per chapter quietly verify that comprehension stayed where extensive reading needs it.
How to actually do it
- Volume target, not page target: aim for a book a month at minimum; a book a fortnight is better. At graded-reader length that is 15–20 minutes a day.
- Choose by interest, quit freely: motivation is a mechanism here, not a nicety. A boring book at the right level still fails, because you stop.
- Do not stop for every word: look up only what blocks the story or what you have now met three times. Everything else, let context do its slow work.
- Reread old books fast: yesterday’s finished reader is perfect fluency material.
What reading will not do
Extensive reading grows vocabulary, grammar intuition and reading speed, and the evidence there is strong. It will not, by itself, make you speak (that is the output problem), and its listening cousin needs its own time. It is the input strand done properly: one quarter of the job, done with the best tool the field has produced.
References and attributions: Day & Bamford, Extensive Reading in the Second Language Classroom, 1998; Hu & Nation, Reading in a Foreign Language, 2000. The extensive reading principles summarised are Day & Bamford’s and Paul Nation’s; paraphrased with attribution, no original text reproduced. Part of the Science of Learning Italian series.
