Is Memorising Vocabulary First Actually the Right Path? What Really Happens

The short answer: not entirely right, and not entirely wrong. Memorising vocabulary is necessary. Its place in the sequence, as most parents imagine that sequence, is upside down. The common assumption is that a child accumulates a stock of words, and then one day those words assemble themselves into sentences. That day does not arrive on its own.

This is worth pausing on because it decides how you spend your child’s time and your money. A family that buys three vocabulary apps and postpones speaking until “he has enough words” is behaving reasonably, but the outcome after a year is usually a child who knows hundreds of words and still cannot ask for a glass of water in a complete sentence.

What follows is the honest version, including the part that does not favour paid lessons or apps.

Where this belief comes from

The belief is not stupid, which is exactly why it survives. It has three reasonable roots.

1. Parents’ own school experience. Many of us learned English through word lists and weekly spelling tests. We repeat what we know.
2. Memorising is measurable. Fifty words this month is a clean number that reassures. “He is braver about speaking” does not fit in a tracking table.
3. Memorising is comfortable for a shy child. An isolated word cannot embarrass anyone. A broken sentence can. So both sides drift toward the safe zone.

Acknowledging these roots matters, because “stop memorising” is also bad advice. The problem is sequence and proportion, not memorisation itself.

Where the belief actually breaks

Words memorised in isolation are stored differently from words that have been used. The practical signs you notice at home:

• A child who knows what hungry means when you ask him, but the word never occurs to him when he is actually hungry.
• A child who translates from Arabic word by word, producing a sentence with recognisable vocabulary and a broken structure.
• A child who knows a word by sight but does not recognise it by ear, because he only ever met it in written form.

The mechanism is simple: a word is retrieved through the channel it entered by. A word that arrived on a flashcard is summoned by the sight of a flashcard. A word that arrived inside a used sentence in a real situation is summoned by a similar situation. If that sounds familiar, how to Tell if a Teacher Is Right for Your Child in the First Lesson goes further.

This explains something that confuses many parents: a child scores excellently on a vocabulary quiz and cannot answer a simple spoken question. There is no contradiction. The two tests measure different things.

The real threshold: how many words before the first sentence?

Far fewer than most parents assume.

With a very small stock you can build dozens of functional sentences, provided the selection is smart:

Word type	Examples	What it unlocks
High frequency verbs	want, have, go, like, see	The core of almost any sentence
Pronouns and connectors	I, you, it, this, and	Subjects and joining
Function words	a, the, to, in, not	Structural correctness
Nouns from the child’s world	water, book, ball, mama	Immediate meaning and motivation

With those four groups a child produces sentences he can use the same day: I want water. This is my book. I don’t like it.

The important observation: that list is not the prettiest vocabulary, nor the vocabulary that dominates picture books. Colouring books teach animals and colours, which are charming words that build no sentence on their own. A child who knows twenty animals and zero verbs can say nothing. A child who knows five verbs and five nouns can say plenty. A related situation is discussed in my Child Refuses to Speak to the Teacher.

The uncomfortable conclusion: where apps genuinely beat lessons

Here is the part that people selling lessons rarely say, and it is true.

Apps are genuinely better at:

• Spaced repetition that fixes a word in long term memory. This is a mechanical job, and humans are worse at it.
• Daily consistency in very short doses, with no scheduling and no setup.
• Recognising the written shape of a word and matching it to an image, which is useful groundwork before reading.

Apps such as Duolingo or Duolingo ABC are useful for exactly these jobs, and Lingokids, as a play learning app, is good at holding younger children’s attention and works as a supplement. The clear limit of this category of tool is the absence of live human correction during free speech. We cover this in more depth in the First Three Months at a British School.

What no app substitutes for:

• Hearing your own error inside a sentence you have never built before, and having it corrected in that moment.
• Being obliged to answer an unexpected question under real time pressure.
• The recast a human gives when they understand your intent despite your broken phrasing.

The practical and slightly counterintuitive conclusion: an app is an excellent tool for the vocabulary layer and a poor tool on its own. A live lesson is an excellent tool for use, and an expensive, inefficient one if its minutes are spent memorising words the child could have fixed for free in five minutes a day. Combining the two is not a luxury. It is a sensible division of labour. For more on this, see trial Lesson Observation Checklist.

The sequence that works: a word inside a frame

Instead of memorise then speak, the version that works in practice is interleaved: every new word enters inside a ready sentence frame from the first moment.

1. Pick one frame for the week, such as “I want ___”, “This is my ___”, or “I can ___”.
2. Put three new words into that frame, not twenty.
3. Attach every use to a real moment in the day: mealtime, play, bedtime.
4. At the end of the week, ask your child to use the frame with a word he chooses himself. That is the actual test.
5. Next week: a new frame, with the old words recycled inside it. Rotation matters more than addition.

Three words a week inside a used frame produce more speaking ability than twenty isolated words, even though the second number looks better on paper. That gap is the whole illusion.

What this means when choosing a learning format

What 51Talk is

51Talk is an online English learning platform founded in 2011, listed on NYSE American under the ticker COE and operated by HelloWorld Online Education PTE. LTD. It provides live one to one lessons with teachers from countries where English is an official language, TESOL certified, across a network of more than 20,000 teachers. Lessons typically run 25 minutes, and details are best confirmed on official channels.

Why this format addresses the sequencing problem

The problem this article describes is not a shortage of vocabulary. It is vocabulary that never converts into use. A live one to one lesson puts the child in a position where he has to use language for most of the session, which solitary memorising never does. The curriculum sits on the CEFR framework and is aligned with Cambridge YLE and Cambridge English Qualifications across levels LS and L0 to L9, so vocabulary arrives attached to a language function at each level rather than as a list. In the early levels, TPR is used at L0 and phonics builds the link between sound and written shape, which is directly relevant to the child who knows a word by sight but not by ear. See the level and curriculum structure and the teacher standards.

Honest limits

A live lesson is not the ideal place to memorise vocabulary, and it makes little sense to spend its limited minutes on something a child can fix alone. The better pattern is for the child to meet the words during pre lesson preview and spend lesson time using them. There is also no guarantee that a given number of lessons produces spontaneous sentences, because the larger difference is made by what happens between lessons. Company background is on the about page.

Bonus tips: auditing your child’s word bank yourself

1. Test recall, not recognition. Do not ask “what does hungry mean”. Say “tell me in English that you are hungry”. The gap between those two answers is the real gap.
2. Count the verbs in his stock, not the nouns. A low verb ratio explains most cases of knows a lot, says nothing.
3. Test by ear as well. Say the word without showing it written. If he does not recognise it, it is stored visually only.
4. Watch response time. A word that takes more than about two seconds will not yet appear in real conversation.
5. Measure by how many sentences your child produced unprompted during the week. That is the only metric tied directly to speaking ability.

Frequently asked questions

How do 51Talk lessons handle vocabulary compared with solo memorising? In a CEFR based curriculum, vocabulary is introduced tied to a language function and a level, and the learning cycle includes pre lesson preview where the child meets the words, a live lesson where he uses them, then practice and unit diagnostic reports. That addresses the use problem rather than the memorising problem, and retention still needs repetition between lessons. Details should be confirmed on official channels.

How many words does my child really need before speaking? Far fewer than most people assume, provided they are chosen functionally: high frequency verbs, pronouns, function words and nouns from his daily surroundings. Type matters more than count.

Should I drop vocabulary apps then? No. Apps are effective at spaced repetition and daily consistency, which humans are not better at. The error is using them as the only tool and expecting speech to emerge from them.

My child scores highly on vocabulary tests but does not speak. Is that normal? Yes, and it is very common. A vocabulary test measures recognition, while speaking requires recall under time pressure. Improving the second needs speaking practice, not longer lists.

Are songs and picture books enough to build vocabulary? They help with concrete words and with rhythm familiarity, but they skew toward nouns and adjectives and are thin on verbs and function words. Pair them with sentence frames the child uses himself.

When is slow speaking a reason to seek a professional opinion? When a similar delay or difficulty also appears in Arabic. A child who is quiet in English while speaking normally in his first language is showing an expected second language pattern. If the first language is affected as well, the right reference point is a pediatrician or a licensed speech language pathologist.

If you want to see the gap between what your child knows and what he can use, book a free one to one trial lesson and count how many words from his bank actually surfaced in his own speech during the session.