How to Plan an ESL Lesson in 20 Minutes (and Why Most Plans Fail)

Two lesson plans for the same hour. One is eight pages with forty-seven boxes filled in, took ninety minutes to write, and will be abandoned twelve minutes into the class. The other is five lines on a page, took eighteen minutes, and will survive contact with actual students.

The difference is not effort. It is that the second plan is a set of decisions and the first is a script. This post is about which decisions actually matter, in the order you make them.

The frameworks below are the ones taught on most CELTA and TEFL courses and in our own lesson planning course. They are conventions rather than laws, and the useful part of learning them is knowing when each one stops working.

The Plan Is a Decision Record, Not a Script

A lesson plan exists to answer four questions before you are standing in front of people: what will students be able to do at the end, how will you get them there, how long does each part take, and what will you do when a part fails.

Everything else on a plan template is decoration. If your plan takes longer than twenty minutes to write, you are over-planning, and the extra time is not buying you a better lesson. Experienced teachers plan fast because they work from a template they trust and a bank of activities they have already used.

The plan should also be short enough to hold in your head. A plan you read from during class becomes a script, and reading from a script in front of students looks less prepared, not more.

PPP: The Default Framework, and What It Assumes

PPP stands for Presentation, Practice, Production. It is the most widely taught lesson shape in TEFL training, and it is popular because it is simple and easy to plan around.

Presentation. You introduce the target language. Students listen and observe.

Practice. Students use it in controlled activities. You monitor and correct.

Production. Students use it freely with minimal intervention from you.

The underlying idea is a gradual release of control from teacher to student. It is a genuinely good default, especially for teaching a discrete language point to a class that has not met it before, and especially early in your career when you need the structure.

What it assumes is that language learning runs input first, output later. That assumption is where it breaks.

Where PPP Falls Apart

PPP tends to fail in four situations, and they are common enough that you should know them by sight:

- Students already know the target language passively but cannot produce it. Presenting it again teaches nothing and signals that you have not diagnosed them.

- The lesson goal is fluency rather than accuracy. PPP optimizes for correctness.

- The class is higher level and gets visibly bored during the presentation stage.

- The topic rewards discovery and problem-solving rather than memorization.

If you recognize your class in that list, the fix is a different framework rather than a better presentation.

Task-Based Learning: The Other Default

Task-Based Learning inverts the order. Instead of presenting language and then practicing it, you give students a real task and let them discover what language they need to finish it.

It runs in three phases:

1. Pre-task. You introduce the topic and the task. You might give useful vocabulary. You do not pre-teach the grammar.

2. Task cycle. Students do the task, then plan and report on what they did.

3. Language focus. You analyze the language that came up during the task and address the problems.

The mechanism worth understanding is the gap. Students attempt the task with whatever language they have, run into the difference between what they wanted to say and what they could say, and then arrive at the language focus stage actually wanting the answer. That motivation is the whole point, and it is not available in PPP, where the answer arrives before the question.

TBL works best with intermediate and advanced learners who have a lot of passive knowledge they cannot yet activate in real time. It is harder with true beginners, who may not have enough language to attempt the task at all, and it demands more of you: your job shifts from presenting to designing tasks and diagnosing live.

Test-Teach-Test sits between the two. Give a short task, find out what they cannot do, teach exactly that, then repeat the task. It is a good compromise when you do not know a class yet.

Objectives: SWBAT and the Verbs That Ruin It

Most lesson objectives are useless. "Students will learn the present perfect" is a topic, not an objective. Learn it how? To what standard? How would you know?

The formula that fixes it is SWBAT: Students Will Be Able To, plus an action verb, plus the language or skill, plus a context. "Students will be able to describe three past experiences using the present perfect in a short conversation with a partner."

The verb is doing the work. Avoid "understand", "know", and "learn", because you cannot observe any of them. Use verbs you can watch: produce, identify, compare, explain, correct, order, describe.

Keep it to one or two objectives per lesson. Five objectives is not an ambitious lesson, it is an unfocused one. The test of a real objective is that at the end of the hour you can say honestly whether they can do it or not.

The Lead-In: Three Minutes That Decide the Hour

Walk in and say "open your books to page 47" and you have already lost the energy for the rest of the lesson.

A warmer gets students talking, thinking, or laughing. A lead-in connects to the topic without giving away the language point. Both should run two to four minutes and require nothing from students except attention.

Five lead-ins that need no preparation:

1. Picture prompt. Show an image, ask questions about it.

2. Personal question. Connect the topic to their own lives.

3. Prediction. Give a headline or title and ask what the text will be about.

4. Short anecdote. Tell a brief true story related to the topic.

5. Quick poll. Hands up if you have ever, or a vote between two options.

All five take under three minutes and none require anything found the night before.

Presenting Language: Meaning, Form, Pronunciation

When you introduce new language, students need three things, and a presentation that misses one is incomplete.

Meaning. What does it mean? Use context, examples, timelines, visuals, concept-checking questions. Not a dictionary definition read aloud.

Form. What does it look like? What is the structure, the word order, the part of speech?

Pronunciation. How is it said? Where is the stress? Which sounds are difficult? Drill it.

The order matters. Start with meaning. A student who does not know what something means will not care about its form, and drilling pronunciation of an item nobody understands is theatre.

Pronunciation is the stage most often dropped when time runs short, and dropping it is how a class ends up able to write a structure it cannot say. Our pronunciation drills post covers the classroom techniques.

Concept-Checking Questions

Never ask "do you understand?". Students say yes, because saying no is socially expensive and because the question is unanswerable in a group.

Concept-checking questions are short questions that test whether the meaning landed. Rules for writing good ones:

- Do not use the target language in the question.

- Keep the question at a lower level than the language you are checking.

- Ask two or three per language point, each checking a different aspect.

- Accept yes, no, or a short phrase as the answer.

For "I used to smoke": Did I smoke in the past? Do I smoke now? Once or many times? Three questions, no target language in any of them, and you now actually know.

Controlled Practice Without the Boredom

Controlled practice is any activity where the language students produce is constrained. Accuracy is the point, and repetition is the mechanism: new language has to be used several times to move out of short-term memory.

The standard formats are gap-fills, substitution drills, sentence transformation, matching, question-and-answer chains, and partially controlled information gaps.

The complaint against them is that they are dull, and they are when the delivery is dull. Five ways to fix the delivery without loosening the language control:

1. Add a game element. Turn the drill into a race or a competition.

2. Use visuals. Flashcards or pictures instead of a list of words.

3. Make it physical. Students stand, move, or change places.

4. Add an information gap. Student A has what Student B needs.

5. Use a chain. One student answers and then asks the next.

The language stays controlled. Only the packaging changes.

Free Practice, and Tasks Students Cannot Fake

Free practice is where students use the language without scaffolding. No frames, no word banks, no gap-fills. This is the stage teachers most often rush, skip when time runs out, or ruin by correcting.

It is also the stage that does the most work. Controlled practice builds accuracy. Free practice builds the ability to use language when nobody is helping, which is the actual goal.

The failure mode is avoidance. Students find workarounds, use simpler structures they are already confident with, or drift into their first language. Design against it:

- Add a reporting element. "You will present your partner's answers to the class."

- Make the target language genuinely necessary to complete the task.

- Build in an information gap.

- Add time pressure so there is no room to overthink.

- Give a clear outcome: a decision, a ranking, a product.

Your job during free practice is to monitor, take notes, and stay quiet. Correction comes after.

Correction: When to Interrupt and When to Wait

The rule is simple and it is tied to the stage, not to the error.

During controlled practice, correct immediately. Accuracy is the goal and letting a wrong form go unchallenged wastes the stage.

During free practice and any fluency activity, note the error and address it afterwards. Interrupting fluency work kills flow and confidence, and the student learns that speaking freely is a trap.

On the feedback itself: the positive-negative-positive sandwich is a fine starting framework and students see straight through it. Specific and honest works better. "Several of you used the third conditional naturally in that discussion, which is hard. I also heard problems with the pronunciation of 'would have', so let me put a few sentences on the board." Direct, no padding, and it gives the class something to do next.

Implicit correction, where you guide a student to find the fix themselves, beats telling them the answer, because self-corrected errors are the ones that stick. It is slower, which is the trade-off.

Cool-Downs and Exit Tickets

Most teachers plan a strong opening and let the ending dissolve. The bell rings, students pack up, and the lesson stops.

A cool-down does three things: it reviews, it gives students a sense of having finished something, and it tells you whether the objective was met. That third one is why it is worth five minutes.

Reliable formats, all two to five minutes:

1. Exit ticket. Two or three sentences or questions aimed directly at the objective.

2. One new thing. Students tell a partner something they learned.

3. Quick quiz. Three to five rapid questions.

4. Traffic light. Students rate their own confidence: green, yellow, red.

5. Summary sentence. One sentence summarizing the lesson.

Pick one and use it consistently. The predictability is a feature. Students start preparing for it, which means they start paying attention to whether they can do the thing.

Skills Lessons Follow Their Own Shapes

Language lessons and skills lessons are not the same planning problem.

Reading runs pre, while, and post. Pre-reading activates background knowledge and pre-teaches three to five critical words, not every unknown word. While-reading is two passes: a fast gist read with one simple question, then a detailed read with four to six questions. Post-reading is where students respond. Two passes matter because that is how reading works outside a classroom. You scan an article to see if it is worth your time, then read it properly if it is.

Listening mirrors it: pre, gist listen, detail listen, post. The pre stage matters more because students cannot re-read. The recurring mistakes are playing the audio only once, giving all the questions before the first listen, not letting students compare answers before you check them, and using the recording as a test rather than as teaching material.

Speaking needs structure, not just talk time. "Discuss this topic" is not a speaking lesson. A real one has a sub-skill focus such as turn-taking or hedging, input before output, preparation time, a genuine reason to speak, and feedback on the speaking rather than only on the content. Information gaps, ranking tasks, problem-solving, and role-play all produce more language than open discussion because each one has an outcome.

Writing is not "write 200 words about this". That is a writing test. A writing lesson teaches the process: analyze a model, focus on the language, plan, draft, peer review, edit. The model text is the part people skip and the part that does the most work.

Integrated lessons join two skills in a natural sequence: read then debate, listen then summarize in writing. The language focus stage between the two halves is the bridge. Remove it and the lesson feels like two unrelated activities.

The 20-Minute Template

Five lines. This works for almost any lesson type.

1. Objective. One sentence. What will students be able to do?

2. Lead-in. One or two sentences. How will you hook them?

3. Main stages. Three or four stages with timings.

4. Materials. What do you need in the room?

5. Anticipated problems. What might go wrong, and what is your backup?

Line five is the one new teachers leave out and the one that saves lessons. Plans fail routinely. The activity you spent thirty minutes on falls flat, the audio does not play, or students finish in five minutes what you planned for twenty. That is not a crisis, it is a Tuesday.

Keep a back-pocket set that needs no materials: a discussion question on the board with pair work, a vocabulary review game, five wrong sentences for pairs to correct, or dictation. Any of them fills ten minutes at any level.

Timings Are the Discipline

Every stage needs a time. Not because you will follow it exactly, but because writing timings forces you to be honest about what fits in an hour.

A workable sixty-minute shape: lead-in 5, presentation 10, controlled practice 10, free practice 15, feedback 10, cool-down 5, buffer 5.

The buffer is not optional. Transitions eat time, instructions eat time, and distributing materials eats time. If your stages total exactly sixty minutes with nothing spare, something will be cut in the room rather than at the desk.

Recycling: Six Lessons From One Text

New teachers build everything from scratch. Experienced teachers adapt. That is efficiency, not laziness, and it is the difference between planning in twenty minutes and planning in ninety.

Take one reading text. You can get at least six lessons out of it:

1. Reading for gist and detail, the standard reading lesson.

2. Vocabulary focus: pull ten useful words and teach them properly.

3. Grammar analysis: students find examples of a target structure in the text.

4. Discussion springboard: use the topic for a speaking lesson.

5. Writing model: students write a similar text using the same structure.

6. Dictation: read sections aloud for listening practice.

You do not need more materials. You need more ideas about what to do with the ones you have.

The other half of this is storage. Two hundred files named "worksheet3_final_FINAL_v2" is not a library. A library needs organization by level or skill, a naming convention you actually follow, and the ability to find a specific activity in under a minute. Keep what worked and what is flexible. Throw away what bombed for reasons you cannot identify and what is too specific to reuse.

Using AI Without Producing Generic Lessons

AI tools compress the mechanical part of planning. They are good at generating gap-fills and matching exercises, brainstorming discussion questions, producing graded reading texts at a stated level, adapting material to a different level, and writing model texts.

They are bad at the things that make a lesson land: knowing your specific students, designing a lead-in that is genuinely interesting, and remembering what worked in your room last month.

If you use them, the prompt determines everything. Include the level, the age and context of the students, the target language or skill, the activity format, the topic, and the number of items. Then edit. Delete anything that reads artificially and add your own examples. The output is bones. You supply the rest.

Bottom Line

The plan is not the lesson. Its job is to force four decisions before you are in the room: what students will be able to do, how you will get there, how long each part takes, and what happens when a part fails.

Pick a framework that matches the goal rather than the one you were trained on. Write objectives with verbs you can observe. Correct immediately in accuracy work and afterwards in fluency work. Finish with something that tells you whether it worked. Then reuse the material instead of building it again.

Five lines, twenty minutes, and a buffer.

Next Steps

Work through the full frameworks in the Lesson Planning Masterclass, pair it with Classroom Management for ESL Teachers, read the companion post on mixed-level classes, browse ready-made curriculum packs, or submit your resume to be found by schools hiring now.