Search "how to improve your vocabulary in English" and the advice is remarkably consistent: read more, keep a word list, use a vocabulary app, learn a word a day. All of it works, in a narrow sense. None of it is built for the actual problem most advanced non-native professionals have.
That problem isn't that you don't know enough words. If you're reading this in English, working in English, and holding your own in meetings conducted in English, your vocabulary is already large. The problem is that a large share of it is recognition vocabulary — words you understand instantly when you read or hear them — rather than production vocabulary — words that come out of your own mouth, correctly, under time pressure, without a pause to translate or search.
That gap between recognizing "mitigate" and reaching for it instead of "reduce" in a live client call is the actual thing worth improving. Most standard advice doesn't touch it.
Why the standard advice mostly builds recognition, not production
Reading more puts new words in front of you. It's one of the best ways to expand recognition vocabulary, and it's genuinely undervalued. But reading is input-only. It doesn't require you to produce the word, under time pressure, in a sentence of your own — which is the specific skill that fails in a live meeting. The research on passive versus active vocabulary development is consistent on this: a word typically needs 10–20 meaningful encounters to become reliably recognized, and recognition alone doesn't predict production.
Word-a-day apps and flashcards are good at exactly one thing: getting a definition to stick. They're not designed to simulate the moment where you need the word — mid-sentence, in front of your VP, with no time to think. Spaced repetition of a definition and retrieval under pressure are different skills, and most apps only train the first one.
General AI language apps are broad by design — built for tourists, students, and beginners as much as for advanced professionals. As the gap in AI English apps for advanced professionals lays out, that breadth means the practice you get is rarely calibrated to the register you actually need: precise, senior, workplace-specific vocabulary, not everyday conversational phrases you already have.
None of this means the standard advice is wrong. It means it answers a different question than the one most working professionals are actually asking.
New to Lyra Practice?
Lyra Practice helps advanced non-native professionals learn the nuance of high-value workplace expressions and practice using them in realistic scenarios until it feels natural, so their English sounds precise and senior at work.
Start learning for free →What actually moves a word from recognition to production
The mechanism that makes a word automatic isn't more exposure. It's repeated, correct production in varied, realistic contexts, with feedback on whether it landed. Four things matter in practice:
1. Target words you'll actually need, not words you happen to encounter. A word-of-the-day app doesn't know whether you negotiate contracts, run engineering standups, or present to a board — so it can't prioritize accordingly. The words worth activating first are the ones your actual role calls for: negotiation language if you negotiate, risk language if you manage risk, meeting language if you run meetings.
2. Practice producing the word, not just recalling its meaning. Recognizing that "mitigate" means "reduce the severity of" is a different skill from reaching for "mitigate" instead of "reduce" mid-sentence, under time pressure, in front of someone whose opinion matters. Practice has to require the second thing, not just test the first.
3. Practice in a context close to where you'll need it. A flashcard with an isolated definition doesn't resemble a client call. Practicing the word inside a realistic workplace scenario — explaining a delay, pushing back on a timeline, presenting a trade-off — builds the retrieval pathway you'll actually use.
4. Get feedback on fit, not just correctness. Using "mitigate" grammatically correctly in the wrong register (too formal for a casual Slack message, too casual for a board memo) is still a miss. The feedback that closes the gap has to judge whether the word fit the moment, not just whether the sentence was grammatical.
What this looks like without a system
You can build a version of this yourself: pick five words relevant to your role, write scenario sentences for each, say them out loud, and ask a colleague or a language model whether they landed. The honest answer about doing this alone is that it works for a week or two — about as long as self-directed habit formation survives against a full professional schedule with no external structure tracking which words still need work and which have actually stuck.
That's not a discipline problem. It's that the loop — target the right word, produce it in context, get calibrated feedback, repeat until automatic — is tedious to run manually and easy to let drift back into passive reading "eventually."
The honest summary
If your goal is a bigger vocabulary in the general sense — more words you'd recognize in an article or a podcast — reading widely and consistently is still one of the best tools available, and no app replaces it.
If your goal is a specific, narrower thing — the precise, senior-sounding word arriving automatically in the meeting where it matters — that requires targeted words, active production, realistic context, and feedback on fit. That's a different kind of practice than most vocabulary advice describes, and it's the one advanced professionals actually need.
Lyra Practice helps advanced non-native English professionals learn the nuance of high-value workplace expressions and practice using them in realistic scenarios, so their English sounds natural, precise, and senior at work. Try Lyra Practice.