Extensive vs Intensive Reading: How to Split Your Time
In short
- Extensive reading means reading a lot of easy material for meaning. Intensive reading means studying a short, difficult text closely.
- They are not competing methods. Extensive reading builds reading speed and consolidates words you half-know; intensive reading is how you learn to handle grammar and argument structure you cannot yet parse.
- Hu and Nation found that most learners reached adequate unassisted comprehension at around 98% known-word coverage, and only a minority did at 95%. Ninety-eight per cent is roughly one unknown word every two to three lines.
- A workable default for an independent adult learner is about four parts extensive reading to one part intensive: three or four easy sessions a week, one hard one.
- The most common mistake is doing only intensive reading, because it feels productive. It is the slower of the two routes to actually reading fluently.
Do both, and lean towards the easy one. Most adult learners get this backwards: they spend almost all their reading time grinding through difficult articles with a dictionary open, and almost none reading things they find easy. That is the wrong ratio, and it is the reason a lot of people with a large vocabulary still read slowly and tire quickly.
Here is what each method is actually for, and how to divide your week.
The two methods, defined properly
Intensive reading is what school taught you. You take a short text that is harder than your current level, and you take it apart. You look up unknown words, work out why a sentence is structured the way it is, answer comprehension questions, maybe translate a paragraph. The text is a specimen. The point is not to finish it; the point is to understand it completely.
Extensive reading is the opposite in almost every respect. You take a long text that is easy for you, and you read it the way you read in your first language: quickly, for meaning, without stopping. No dictionary. No note-taking. If you do not understand a sentence, you keep going. The point is not to understand every word; the point is volume and momentum.
The modern framework most teachers use comes from Richard Day and Julian Bamford, whose ten principles for teaching extensive reading were published in Reading in a Foreign Language in 2002, four years after their book Extensive Reading in the Second Language Classroom. The principles are blunt: reading material should be easy; a variety should be available; learners choose what they read; they read as much as possible; reading is for pleasure, information and general understanding rather than 100% comprehension; reading speed should increase; reading is its own reward; it is individual and silent; teachers orient and guide; and the teacher is a role model of a reader.
| Extensive reading | Intensive reading | |
|---|---|---|
| Text difficulty | Below your level | Above your level |
| Text length | Long: chapters, whole books | Short: a page, an article |
| Speed | Fast, uninterrupted | Slow, with stops |
| Dictionary | Almost never | Constantly |
| Goal | Fluency, automaticity, consolidation | Precision, new grammar, new words |
| Success looks like | Finishing the book | Understanding the paragraph |
| Failure mode | Text is secretly too hard | Never reading anything else |
The number that decides which one you are doing
The dividing line between "extensive" and "intensive" is not a feeling. It is a measurable property of the text: what proportion of the running words you already know.
The reference point most teachers use comes from Marcella Hu and Paul Nation's 2000 study on unknown-word density. They gave learners the same fiction text with vocabulary manipulated so that known-word coverage sat at 80%, 90%, 95% or 100%, and measured comprehension. The finding that stuck: at around 98% coverage most learners reached adequate unassisted comprehension. At 95% only a minority did. At 80% almost nobody did.
Worth knowing before you treat 98% as a constant: the regression behind it was run on 66 university students in New Zealand, and the figure has since been revisited in a replication published in Language Learning. It is a well-supported working threshold, not a physical law.
98% sounds close to perfect until you convert it. Ninety-eight per cent coverage means two unknown words in every hundred — in an ordinary paperback, roughly one unknown word every two to three lines. That is the ceiling for extensive reading. Anything denser than that and you are, whether you meant to or not, doing intensive reading.
This is why so much "extensive reading" fails. A learner picks a real novel, hits six or seven unknown words per page, decides that is close enough, and ends up reading forty pages a month at a crawl. The method was never given a chance, because the text was never easy enough.
A quick way to check a specific text: paste a page of it into the readability checker and look at the CEFR band it comes back with. If it lands a full level above where you are, it is intensive material. If it lands at or just below your level, it is extensive material.
What each method actually gives you
They are not two roads to the same place. They produce different things.
Intensive reading gives you new knowledge. A grammatical structure you cannot yet parse — a reduced relative clause, a fronted concessive, an academic hedge — will not decode itself no matter how much easy text you read past it. You have to stop and take it apart at least once. The same is true of a technical vocabulary you have never met. Intensive reading is how new territory gets mapped.
Extensive reading gives you speed and retention. Two things happen when you read a lot of easy text:
The first is automaticity. Recognising a word is not binary. There is a large gap between "I can work out what this means" and "I read it without noticing I read it". Closing that gap takes repeated exposure at speed, and nothing but volume delivers it. This is where reading rate comes from, and it is the single biggest reason people fail timed exams despite knowing the words.
The second is consolidation of half-known words. Vocabulary is not learned in one meeting. Stuart Webb tested this directly, giving learners unknown words 1, 3, 7 or 10 times in context and measuring ten different aspects of word knowledge. Every increase in repetition improved at least one aspect; ten encounters produced sizeable gains, and he concluded that developing full knowledge of a word needs more than ten. You will not get ten encounters with a mid-frequency word from flashcards you stopped reviewing. You get them from reading two hundred thousand words a year.
The honest limit of extensive reading is that incidental pickup is inefficient per encounter. Rob Waring and Misako Takaki tracked what fifteen Japanese learners retained from one 400-headword graded reader, using invented words substituted into the text and testing immediately, after a week and after three months. About 40% of the words were remembered on the immediate multiple-choice recognition test. Three months later, even words that had appeared more than eighteen times had only a 10-15% chance of the meaning being remembered.
One book does not teach vocabulary. A hundred books do — and they teach it in a form flashcards cannot: with collocation, register and grammatical behaviour attached.
How to split the week
There is no research-backed golden ratio, so treat this as a working default rather than a law:
| Sessions per week | What you do | Roughly how long |
|---|---|---|
| 3-4 | Extensive: an easy book or article, no dictionary | 20-30 minutes each |
| 1 | Intensive: one hard text, taken apart properly | 30-45 minutes |
| 1 | Review: the words you flagged during the intensive session | 10-15 minutes |
Four parts easy to one part hard. If you are preparing for IELTS or TOEFL, shift towards intensive for the last six weeks before the exam, because exam texts sit above most candidates' comfortable level by design — but do not drop extensive reading, because the exams are timed and timing is a fluency problem. There is more on that in IELTS reading time management.
If you are at A2 or low B1, weight it even further towards extensive. At that level almost any authentic text is intensive material, and the risk is spending a year doing nothing but decoding.
Three failure modes worth naming
Reading too hard, forever. The most common one. Difficult text feels like real study, so people keep choosing it. The tell is that your reading speed has not changed in a year.
Reading too easy, forever. Rarer, but real. If you never meet anything above your level, you stop acquiring new structures. The tell is that your reading is fast and comfortable but your writing has not improved.
Turning extensive into intensive by accident. You start a novel, look up the first few unknown words because it seems responsible, and by chapter two you are looking up everything. The fix is mechanical: put the dictionary in another room, or use a text where you genuinely do not need one.
What to do this week
- Find out where you actually are. A free English level test takes a few minutes and gives you a CEFR band to aim below.
- Pick one easy thing and one hard thing. The easy thing should be long — a graded reader, a middle-grade novel, a familiar series. The hard thing should be short — one article you actually care about.
- Read the easy one four times this week without a dictionary, and the hard one once, slowly, with everything.
- At the end of the week, note how far you got in the easy text. That number, tracked over a month, is your fluency measurement. It is more useful than any vocabulary count.
If finding text at exactly the right difficulty is the part that stops you, that is the specific problem LingoRead exists to solve: it generates reading passages at a CEFR level you choose, with comprehension questions and a vocabulary list built from the passage itself, so the "is this too hard" question stops being a guess. But the method above works with library books and a bit of discipline, and the method is the part that matters.
Sources
- Day, R. and Bamford, J. (2002). Top ten principles for teaching extensive reading. Reading in a Foreign Language, 14(2). Source for the ten principles quoted above.
- Hu, M. and Nation, P. (2000). Unknown vocabulary density and reading comprehension. Reading in a Foreign Language, 13(1), 403-430. Source for the 80/90/95/100% coverage conditions and the 98% threshold.
- Kremmel, B. et al. (2023). Unknown vocabulary density and reading comprehension: replicating Hu and Nation (2000). Language Learning. The replication and the caveat about the original sample.
- Webb, S. (2007). The effects of repetition on vocabulary knowledge. Applied Linguistics, 28(1), 46-65. Source for the 1/3/7/10 encounter design and the conclusion about ten encounters.
- Waring, R. and Takaki, M. (2003). At what rate do learners learn and retain new vocabulary from reading a graded reader? Reading in a Foreign Language, 15(2), 130-163. Source for the 40% immediate recognition figure and the three-month decay.
- Millett, S. Speed reading and listening fluency and graded readers, Te Herenga Waka - Victoria University of Wellington. Free out-of-copyright graded readers at 4,000, 6,000 and 8,000 word levels.