You're reading in English constantly. You're in English meetings all day. You listen to English podcasts on your commute. You write reports, emails, and client communications in English without effort. By any reasonable measure, you're as immersed in English as it's possible to be while also working a full-time job.
And yet your English doesn't feel like it's getting better. The words you reach for in meetings are roughly the same words you were reaching for two years ago. The gap you noticed hasn't closed. At some point you start to wonder whether this is just where you plateau — whether there's a ceiling that's specific to where you are, and whether improvement past it is actually possible.
This is one of the most common and least-discussed experiences among advanced non-native professionals. And the frustration is almost always accompanied by a private interpretation: that something is wrong with you, your effort, or your aptitude. That interpretation is wrong. What's happening is a structural feature of advanced language acquisition that has nothing to do with effort and everything to do with method.
Why progress felt different before
At B1 to B2, progress is visible and fast. New grammar structures click. Vocabulary clusters open up. Conversations that were impossible become possible. You can measure the gains in concrete before-and-after terms — texts you couldn't follow, now you can. Speech at speeds that were too fast, now they're manageable. The feedback loop is immediate, satisfying, and self-reinforcing.
At C1 to C2, the nature of progress changes completely — and most people don't realize the measurement system needs to change with it. You're no longer learning new structures. You're refining precision within structures you already control. You're not adding new vocabulary clusters. You're activating specific words within a vocabulary you already have. The gains are narrower, slower, and harder to see. They don't feel like what progress felt like before because they're categorically different from what progress was before.
This isn't a plateau in capability. It's a plateau in the visibility of progress — and that visibility problem is what makes it feel like you've stopped moving when you haven't.
The measurement problem
At intermediate level, the milestones are clear. You can do something you couldn't do before. At advanced level, most of those milestones are already cleared. What remains is specific and subtle: a word that doesn't arrive fast enough, a register that's slightly off, a phrase that a native speaker would reach for and you don't. These gaps are real. They're professionally consequential, as the research on passive versus active vocabulary makes clear. But they don't show up on any standard proficiency measure. They show up in the room — in the moments where the precise word would have changed the impression and the general word showed up instead.
The problem isn't that you've stopped making progress. It's that progress at this level doesn't register on the measurement systems you've been using. The gap between "can hold a professional conversation in English" and "reaches for the right word automatically in every professional context" is real and meaningful. It just doesn't look like the gap between B1 and B2 looked.
Why passive immersion stops working
At intermediate level, passive exposure — reading, listening, being in English environments — produces visible gains because there's still a large volume of vocabulary and structure to acquire. Every encounter with new material potentially adds something. The acquisition layer is still full.
At advanced level, the acquisition layer is mostly complete. The remaining gaps are in the activation layer: words you know but can't yet deploy automatically under the specific conditions of professional pressure. Passive exposure adds vocabulary at the margins, but it doesn't move the words you already know from passive to active. The mechanism that builds active retrieval is production under realistic conditions — generating the word yourself, in context — not encounter. More immersion at this stage produces diminishing returns precisely because the problem has shifted from acquisition to activation, and immersion is an acquisition tool.
This is why the plateau feels so stubborn. You're putting in genuine effort — the immersion is real, the time is real, the intention is real. But the effort is mismatched to the problem that's left.
What progress actually looks like at this level
Progress at advanced level doesn't look like new capabilities. It looks like specific words becoming automatic.
It looks like reaching for "mitigate" in a meeting and it being there, without the half-second of uncertainty about whether it fits. It looks like an email that sounds like you — precise, authoritative, the register you intended — rather than a slightly flattened version of you. It looks like the gap between how you think and how you sound in live professional contexts getting narrower, one word at a time.
These are small moments individually. Over time they're the difference between a professional whose language reflects their thinking and one whose language consistently underrepresents it. As the evidence on how long activation takes makes clear, reaching these moments requires a specific kind of practice — deliberate production in realistic conditions, tracked at the word level, with feedback on whether it landed correctly.
Progress at this level is measured in activated words, not in new capabilities. That requires a different measurement system and a different practice method than what worked before.
What the plateau actually is
Not progressing doesn't mean you've hit a ceiling. It means the method that got you here has reached the end of its usefulness for the problem that remains. Passive immersion worked at earlier stages because acquisition was the problem. It works less well now because activation is the problem. The plateau is real — but it's a method plateau, not a capability plateau.
The gap that remains is specific: vocabulary that exists in passive form but hasn't been activated for reliable production under professional pressure. It's technical rather than existential. It's closable. It just requires different tools than the ones that worked before.
Lyra Practice is built for exactly this stage — where the problem has shifted from acquisition to activation, and passive immersion has stopped producing visible results. There's a free tier if you want to see what practice looks like at this level.