How to Learn English by Yourself: A Six-Month Plan That Survives

People who stall alone tend to stall in the same spots, and none of them is really about discipline.

• Starting far too big. Two hours a day in week one, zero minutes a day by week three. The sustainable amount in self-study is whatever you can do on a day you do not feel like it. For most people that is around thirty minutes, and thirty minutes that happens beats ninety that does not.
• Switching materials constantly. A third of a book, then something better appears, then a third of that one. Someone who did a third of six books knows far less than someone who finished two. The quality gap between good resources is much smaller than the gap between finishing and abandoning.
• Doing only the comfortable half. Listening and reading can be done alone, in silence, badly, with nobody watching. Speaking and writing cannot. So six months go into input, and you end up understanding almost everything and producing almost nothing — exactly the state described in understanding but not speaking.
• Never measuring. If you have no way to tell whether you improved, motivation lasts about three weeks. After that the only thing carrying you is habit, and most people quit before the habit forms.

All four are design problems rather than effort problems, which is good news: a design can be fixed in an afternoon.

Spend the first week measuring instead of studying. Without a baseline you will choose your materials, your goal and your timeline by guesswork, and you will have no way to know six months from now whether any of it worked.

Three measurements are enough.

• Listening. Play five minutes of a podcast or an unscripted interview with no subtitles and write down roughly what percentage you followed. Below about 70 per cent and that material is still too hard to study with; it will feel like work and teach you very little.
• Speaking. Record yourself on your phone talking about yesterday for two minutes. Count the stalls, the times you translated from your own language first, and the words you repeated because a better one would not come. It is unpleasant listening. It is also the single most accurate thing you will own in six months, because you will record it again.
• Vocabulary. Any online size test gives a rough number. Under 3,000 words and vocabulary is your bottleneck. Over 5,000 and it almost certainly is not — what you are missing is retrieval speed, not words.

Then write your goal as one sentence. ‘Get better at English’ is not a goal. ‘Say my opinion in a video meeting without freezing, by March’ is, and so is ‘travel for two weeks without a translation app’. A specific goal filters materials on its own: exam workbooks are wasted on someone whose problem is meetings, and shadowing a drama is wasted on someone sitting an exam in eight weeks.

Repeat all three measurements every six weeks. That alone removes the largest hole in self-study, which is not knowing whether anything is moving.

In self-study, resources are a liability rather than an asset. Ten of them means deciding every single day what to do today, and that decision is what actually exhausts people — not the English. Use exactly three for six months.

• One grammar book, thin enough to finish. A thick reference is for looking things up, not for working through. If your goal is conversation you need far less grammar than you think: word order, three tenses, articles and question forms carry almost everything, which is the argument made in English grammar basics.
• One listening source. A podcast, a channel, a show — pick one and stay. Choose something you follow between 70 and 90 per cent of. Easier than that teaches nothing; harder than that is just noise with your language in it.
• One vocabulary app with spaced repetition. One. People running two end up doing neither.

Do not spend more than a day choosing. The time it takes to watch five comparison videos would have finished your first chapter. And once chosen, do not switch for six months. The urge to switch usually arrives at exactly the point where the material starts being difficult, which is the point where it starts working.

Store vocabulary as sentences rather than pairs. Postpone = delay survives in a test and nowhere else; I keep putting it off survives in a conversation, because it arrives with a subject, a tense and a situation attached. Everyday English vocabulary covers which words are worth that effort in the first place.

English has four skills but only two characters. Listening and reading are input; speaking and writing are output. Self-study tilts toward input for a structural reason: you can do unlimited input alone, and you cannot do unlimited output alone.

Realistic splits by stage:

• Beginner (under roughly 3,000 words): 70 input, 30 output. With no raw material there is nothing to retrieve, and at this stage sheer volume of listening and reading is genuinely the biggest lever.
• Intermediate (you understand nearly everything and produce very little): 40 input, 60 output. This is where most self-learners are, and where adding more input does almost nothing. The shortage is not knowledge; it is practice at getting known things out.
• Advanced: 30 input, 70 output, with the input coming from real material in your own field rather than learner material.

Some output genuinely works alone. Reading aloud and shadowing build the mouth and the rhythm, which nothing silent can do. Writing trains sentence-building cheaply and without stage fright — an English diary at three sentences a day is the lowest-friction version. Talking to yourself works too: narrate what you can see on the way to work.

But notice what those three share. You choose the situation. The real difficulty of conversation is that you do not choose it — and that is the ceiling on solo output, no matter how disciplined you are.

Thirty minutes a day, five days a week. Plans bigger than this mostly do not survive week three. Spend the thirty like this:

• 10 minutes — vocabulary. Review in the app; add at most five new words, each with one example sentence about your own life.
• 10 minutes — input. Your chosen listening source. Once without subtitles, once with. Harvest exactly three expressions. Try to capture everything and you will not open that material again next week.
• 10 minutes — output. Build sentences with today’s three expressions and say them out loud, or write three diary sentences. Use your mouth, not only your eyes: a phrase you have only read stays available for listening and unavailable for speaking.

Add one 40-minute block a week for a grammar chapter, and repeat the step-one measurements every six weeks. Re-record the two-minute clip and compare it with the old one; it is a far more honest report card than a progress bar in an app.

Three rules keep the routine alive. Anchor it to something that already happens — before your commute, straight after lunch — rather than to a time of day you have to remember. Miss a day and simply do the next one; a surprising number of people quit out of guilt at breaking a streak rather than out of boredom. And take one full day off a month on purpose. What carries you through six months is recovery, not intensity.

Do everything above and vocabulary, grammar, listening and pronunciation will travel a long way on their own. Two things will not, and not for want of effort — they are impossible alone by definition.

• Feedback. You cannot correct a mistake you do not know you are making. Someone who has said I’m agree for three years says it because for three years nobody told them. Practice alone is perfectly happy to cement the wrong version with exactly the same care as the right one.
• Unpredictability. A real conversation is someone saying a thing that is not in your script, using a word you did not prepare, and not giving you time to assemble a sentence. Shadowing cannot manufacture that, because shadowing runs at your pace and stops when you stop.

So the honest version of self-study is not ‘do everything alone’. It is do everything alone that can be done alone, and buy only the part that cannot. In practice that is about thirty minutes out of every 150 you study. The question is whether those thirty minutes exist at all — without them, a good share of the other 120 accumulates unverified.

The last piece can be small. Once or twice a week, fifteen minutes, is enough to close both gaps above.

• Decide three things before you start. Write down three expressions you intend to use today. Anything you plan to use ‘if it comes up’ never comes up.
• Short and frequent beats long and rare. Four fifteen-minute conversations beat one hour-long lesson, because you rehearse the opening and the ending four times instead of once, and those are the most repeated parts of any conversation.
• Ask exactly one question at the end. ‘Did anything I said sound unnatural?’ Two corrections from that question move you further than the fifty words you memorised that week.

CoffeeTalk connects you with native speakers for short conversations, which fits this use precisely: fifteen minutes is long enough to spend your three prepared expressions and short enough to do on a weekday, and it supplies the one ingredient no app can fake — a question you did not choose. Setting the topic to daily life or travel raises the odds that what you prepared gets its moment.

What changes in six months is not your word count. It is the delay before you start a sentence. Play back that two-minute recording from week one and the difference is audible rather than theoretical. If you want to go deeper on the speaking half alone, how to study English conversation covers just that, and English conversation sentences gives you material to spend in those fifteen minutes.