How to Teach ESL Vocabulary So It Actually Sticks (Not Just Gets Recognised)

Why "knowing" a word isn't the same as being able to use it

A student gets full marks on Friday's vocabulary quiz, matching every word to its definition without hesitation. On Monday, telling you about their weekend, they stall mid-sentence searching for the word "borrowed" — a word they matched correctly three days earlier — and settle for "took" instead.

This gap is normal, and it has a name: the difference between passive and active vocabulary. Passive vocabulary is what a student can recognise when they see or hear it. Active vocabulary is what they can retrieve and produce on their own, under the pressure of a real conversation or a blank page. Most classroom vocabulary work — matching exercises, definition quizzes, flashcard drilling — builds passive vocabulary very efficiently and active vocabulary barely at all, because recognising a word and generating it from memory are different mental tasks.

That's not a reason to drop flashcards. It's a reason to add techniques that specifically target production, alongside the recognition work you're already doing. The four below are ordered roughly by how much classroom time they need, from a five-minute addition to an existing lesson through to a genuine restructuring of how you introduce new words.

Sort words into categories instead of memorising lists

Vocabulary taught as an alphabetical or unit-based list gives a student's memory very little to hook onto — each word sits next to whatever happened to come before it in the textbook, with no real connection. Vocabulary taught in categories gives the brain a structure to file it under, and structured memory is recalled far more reliably than a flat list, because retrieving one word in the category primes the others.

The activity is simple: give students a mixed set of word cards — some animals, some emotions, some food items, whatever fits your current unit — and ask them to sort the cards into groups before you've told them what the groups are. Let them argue about borderline cases; a student who insists "shark" could go under both "animals" and "scary things" is doing exactly the kind of semantic processing that makes a word stick. Only after the sorting is done do you introduce or confirm the category labels.

This works particularly well as a quick centre or warm-up activity, since it needs no explanation beyond "put these into groups that make sense to you" — a genuinely no-prep task once the cards exist. For a mixed-level class, pair a stronger student with a developing one and have them negotiate the categories together out loud; the negotiation is where the vocabulary gets rehearsed, not just the sorting itself.

Make students produce the word, not just recognise it

Recognition tasks ask "which of these is correct?" Production tasks ask a student to generate the word themselves, with nothing to choose from — a much harder and much more useful skill, and one that most vocabulary programmes under-practise.

A few production techniques that fit into a normal lesson without extra prep:

• Sentence generation. Instead of asking students to match a word to its definition, ask them to write one original sentence using it correctly. A sentence that's slightly wrong ("I borrowed my friend a pencil") tells you far more about what a student actually understands than a correct multiple-choice answer does, and it's easy to correct on the spot.
• Describe it without saying it. In pairs, one student describes a target word using only other words they already know, while their partner guesses. This forces the describer to actively search their own vocabulary for a workaround, which is close to what happens in real conversation when a word won't come.
• Word chains. Say a category — "things you find in a kitchen" — and go round the group, each student adding a new word without repeating one already said. The pressure of not repeating forces genuine retrieval rather than recognition.

These take two or three minutes each and can slot into the last part of a lesson as a closing activity, which also does double duty as a quick check on what actually landed that day.

Build in spaced, low-stakes retrieval

A word introduced once and tested once is a word most students will have lost within a few weeks, regardless of how well they seemed to know it on the day. Spaced retrieval — coming back to the same words at increasing intervals — is one of the most reliably effective techniques in vocabulary retention research, and it costs almost nothing in class time if you build it into your routine rather than treating it as extra work.

A simple version: keep a running "word wall" of everything taught so far, and spend two minutes at the start of each lesson pulling five words at random — a mix of last week's and last month's — for a quick oral or written retrieval task. If a word that was "mastered" a month ago gets missed, that's useful information rather than a setback; it goes back into rotation rather than being marked as a failure.

The mistake to avoid is reviewing only the current unit's words. A student who can recall this week's vocabulary perfectly but has lost half of last month's list hasn't actually built lasting vocabulary — they've built short-term recall that looks like progress on a weekly test and evaporates over the term.

Adapting these techniques across different settings

In a classroom, category sorting and word chains work best as group activities, since the discussion and disagreement between students is doing much of the retention work. In a mixed-level group, adjust the target word count rather than the technique itself — a stronger student sorting fifteen words and a developing student sorting six are doing the same kind of thinking at different scales.

For self-study learners working without a class around them, the "describe it without saying it" technique still works alone: write the target word on one side of a card and a mini-definition, in your own words, on the other, then test yourself by describing the word before flipping the card. Keeping a vocabulary journal organised by category rather than by date — a page for "emotions," a page for "weather," a page for "food" — gives the same retrieval structure that group sorting gives a class.

Parents supporting a child's English at home don't need any teaching background to help here. A simple "I spy a category" game — naming three things that belong together and asking the child to guess the category, then swapping roles — practises exactly the same semantic grouping skill as a classroom sorting activity, with no materials beyond whatever's in the room.

None of this replaces the recognition-based work you're already doing — matching, flashcards and definition checks still have a place, because a student needs to recognise a word reliably before they can be expected to produce it. What changes is what happens after that first recognition stage: the difference between a student who can define "borrowed" on a quiz and a student who reaches for it naturally on a Monday morning is almost always the amount of production practice they've had, not the number of times they've seen the word.