TL;DR
- Reading builds a language when the text is comprehensible — mostly understandable, slightly challenging.
- This ~12-week plan pairs 15–20 min of daily reading with 5–10 min of spaced-repetition review.
- Raise the share of target-language words gradually as your comprehension grows.
- Save and review every unknown word so it sticks.
- Timelines vary widely; treat weeks as a framework, not a promise.
Learning a language by reading means acquiring vocabulary and grammar by processing large amounts of text you can mostly understand, then reviewing the new words until they stick. The reading-first plan below turns that principle into a concrete routine you can start today. It will not hand you fluency, but it gives you a defensible path from a shaky A2 toward a functional B1.
Why reading-first actually works
Stephen Krashen's Input Hypothesis (Krashen, 1985) argues that we acquire language when we understand messages slightly beyond our current level, often written as i+1. Reading is the most controllable way to get that input: you set the pace, you can reread, and you can look up what blocks comprehension. Unlike a live conversation, a page waits for you.
Reading pairs naturally with two other well-supported ideas. Paivio's dual coding theory (Paivio, 1971) shows that words tied to images tend to be remembered better than words alone. And spaced repetition — formalized in the SM-2 algorithm developed by Piotr Woźniak for SuperMemo, and later popularized by Anki — counters Ebbinghaus's forgetting curve (Ebbinghaus, 1885) by resurfacing a word right before you would forget it.
The catch that trips up most self-learners: raw native text at A2 is not comprehensible. A newspaper article where you know one word in three is decoding, not reading. The plan below keeps difficulty in the productive zone.
The reading-first framework (12 weeks)
The core loop is the same every day, regardless of week:
- Read for 15–20 minutes on a single text or story.
- Save every word you have to look up.
- Spend 5–10 minutes reviewing saved words in a spaced-repetition game.
What changes across the plan is what you read and how much of it sits in the target language.
| Phase | Weeks | Daily reading | Target-language share | Text type |
|---|---|---|---|---|
| Foundation | 1–3 | 15 min | ~20–25% | Short familiar articles, A1–A2 graded stories |
| Expansion | 4–8 | 20 min | ~35–45% | A2–B1 graded stories, blog posts on your hobbies |
| Consolidation | 9–12 | 20–25 min | ~50–65% | B1 stories, simplified news, short fiction |
The "target-language share" column is where a tool like LingoBlend fits. Its Smart Blend takes text you already care about and swaps a percentage of the words into the language you are learning, controlled by an adjustable slider. You start low so the passage stays readable, then raise the dial as your eye adapts. This is a modern take on the diglot weave method first proposed by Robbins Burling (1968), who wove foreign words into English stories so learners absorbed them in context.
Choosing texts by level
The single biggest lever is text choice. Two rules keep you honest:
- The coverage rule. Paul Nation's vocabulary-coverage research suggests you need to know roughly 95–98% of the words on a page to read comfortably — about 98% for truly independent reading, and around 95% when you have some support (a glossary, pre-translation, or a dictionary at hand). Below that band, the text is too hard for extensive reading and you will burn out. Drop to graded material or raise the amount you have pre-translated.
- Read what you would read anyway. Day and Bamford's work on extensive reading (Day & Bamford, 1998) stresses that easy, self-selected material is what sustains volume — and volume is what moves the needle. A cooking blog you love beats a "classic" you dread.
At A2, lean on graded stories written for your level and short, predictable texts (routines, recipes, product reviews). Approaching B1, mix in opinion posts, fan wikis, and simplified news. If you want a curated ramp, LingoBlend's graded stories run A1 through C2, and there are language-specific starting points for Spanish, French, German, and Japanese.
Pairing reading with spaced repetition
Reading exposes you to words. Review is what converts a fleeting recognition into durable memory. The moment you look a word up, save it, then let a scheduler decide when you see it again.
A standard SM-2 ladder introduces a new word at short intervals and stretches them as you succeed:
| Stage | Interval |
|---|---|
| Learning steps | 10 min → 1 h → 8 h |
| First reviews | 1 day → 6 days |
| Long-term | Each success multiplies the gap by your ease factor (up to ~2.5×), capped at 180 days |
Miss a word and it drops back down the ladder. Get it right and the gap widens. Over 12 weeks this quietly builds a few hundred words you actually retain, rather than a giant list you highlighted once and forgot.
In practice, this is where saving from context matters. LingoBlend lets you tap any blended word for its translation, grammar form, and base form, then save it in one tap; its five games — Flashcards, Matching, Fill in the Blank, Listening, and Word Quiz — run the reviews on the SM-2 schedule above. You can read more about the mechanics on the how it works page or the research behind it on the science page. The underlying principle is provider-agnostic, though: any Anki deck fed by your reading will do the same job.
Tracking progress without fooling yourself
Track inputs, not vibes. Three numbers are enough:
- Days read this week (aim for 5–6 of 7).
- New words saved and reviewed (a rough proxy for exposure).
- Blend percentage you can read comfortably (should creep upward).
A weekly self-check works well: reread a passage from three weeks ago. If it now feels easy, your comprehension is climbing even when daily practice feels flat.
An honest word on timelines
A2 to B1 is a real jump. Commonly cited CEFR guidance suggests it can take on the order of 150–200 hours of focused study, but that figure swings hard based on your native language, the target language's distance from it, and how concentrated your hours are — so treat it as a rough estimate, not a rule. Twelve weeks at ~25 minutes a day, six days a week, is only around 30 hours. That is enough to feel a clear shift and to make native-ish text far less painful; it is usually not enough to certify at B1 on its own. Treat this plan as the reading engine, and expect to add speaking and listening practice on top. Anyone promising a guaranteed level by a fixed date is selling, not teaching.
FAQ
Can you really learn a language just by reading?
Reading alone builds strong vocabulary, grammar intuition, and reading comprehension, but it under-develops speaking and listening. Use reading as your input backbone, then layer in audio and conversation practice for balanced ability.
How many minutes a day should I read?
Fifteen to twenty-five minutes daily is a sustainable target for most learners, plus a short review session. Consistency across five or six days beats occasional long sessions, because spaced exposure is what drives retention.
What percentage of a text should I understand?
Aim to know roughly 95–98% of the words without help — about 98% for comfortable independent reading, and around 95% if you have some support. Below that, switch to graded material or pre-translate part of the text so the passage stays comprehensible.
Is reading better than flashcards for vocabulary?
They do different jobs. Reading teaches words in context and shows how they behave; flashcards on a spaced-repetition schedule lock them into long-term memory. Combining both, as this plan does, tends to beat either alone.
How does blending help a beginner read native text?
Blending replaces a set percentage of words with the target language inside text you already understand, so the surrounding context carries you. As Robbins Burling (1968) noted, weaving new words into familiar sentences lets you absorb them without constant dictionary breaks.
Start your first week
Pick one text you find genuinely interesting, set a modest target-language share, read for 15 minutes, and save every unknown word. Do that tomorrow, and the day after. You can build the routine by hand with a reader and an Anki deck, or run the whole loop in one place. If you want the second path, compare tiers on the pricing page and read a blended article to see whether reading-first fits how you learn.