---
title: "The feedback gap: learning English vs using it at work"
description: "Why you sound fluent in lessons and stuck in meetings: a tutor gives feedback with no stakes, a meeting gives stakes with no feedback. How to get both at once."
canonical: https://speaklens.si/blog/learning-english-vs-using-it-at-work
published: 2026-10-29
author: Ibrahim Rahhal
tags: [fluency, meetings, feedback]
---

# The feedback gap: learning English vs using it at work

Your English is better with a tutor than in meetings because the two rooms reward different things. A tutor catches every mistake in a conversation that does not matter, and a meeting lets every mistake pass in a conversation that does. You learn in one room and perform in the other, and the feedback from the first never reaches the second. That missing link is the feedback gap.

## Why does my English feel better with a tutor?

Because a lesson is built to make you look good. The topic is one you prepared or chose, and the tutor waits while you find the word. You can restart a sentence without anyone losing interest, and you use your best vocabulary because you have time to reach for it.

Robin Hogarth (2001) called this a kind learning environment: feedback arrives fast, it is accurate, and it is linked to the thing you just did. In a kind environment you improve quickly and can feel it, which is why the Tuesday lesson feels like progress.

A work meeting is the opposite. The agenda sets the topic, and someone will take the floor if you pause for more than a second. Your attention is on the numbers being discussed. The feedback, when there is any, is a nod that means "I understood enough". Hogarth's name for that is a wicked environment, where feedback is missing, delayed, or misleading. You can spend years in one without getting better, and many people who work in English do.

## Why is talking at work so different from talking to your teacher?

Because in a meeting your brain is doing two jobs, and in a lesson it is doing one. With a teacher, the English is the job: the topic is easy, and all of your attention can go to how you say things. At work the English is only a tool. Most of your attention goes to the work itself, and the language gets whatever is left.

![Two attention budgets of twenty cells each. With your teacher, three cells go to the topic and seventeen to the English. In a work meeting, fifteen cells go to the work (the argument, the numbers, the people, the time) and only five are left for the English.](/blog/learning-english-vs-using-it-at-work/attention-budget.svg "The same attention spent two different ways. The block counts are illustrative.")

Speaking happens in three stages: deciding what to say, turning it into words and grammar, and saying it out loud (Levelt, 1989). In your first language the middle stage runs almost on autopilot, so nearly all of your attention is free for the message. In a second language the middle stage needs attention too (Kormos, 2006), and attention is a fixed budget. Skehan (1998) describes fluency, accuracy and complexity as competing for that budget: when the task gets harder, one of them gives.

A lesson makes the first stage almost free: you are describing your weekend, and the teacher waits. A meeting makes it expensive. You are following the argument, checking the numbers, reading the people on the call and deciding when to jump in, all before you have chosen a single English word. The grammar you used easily on Tuesday is still in your head on Thursday, but it has no attention left to run on.

> [!IMPORTANT] Two different places
> A lesson and a meeting give your brain two different jobs, so progress in one room does not carry over to the other on its own.

## What happens to a mistake in a real meeting?

Nothing: nobody corrects it, and you do not notice it yourself.

Colleagues hold back because correcting an adult in front of other adults threatens their face, and people avoid face-threatening acts unless the stakes force them to (Brown and Levinson, 1987). In a meeting the stakes point the other way. The group needs the discussion to move, and as long as your meaning came through, correcting the grammar would cost more than it gains, so they let it go.

You cannot fix the mistake yourself either, because you did not notice it. Schmidt (1990) argued that learners only change a form after they notice the gap between what they said and the target. In a meeting you are busy with content, so the gap goes unnoticed. The sentence sounded fine to you, and it will sound fine again on Thursday.

Repeat that for a year and the error becomes part of how you speak, which Selinker (1972) called fossilization. Most people who have worked in English for a long time carry a handful of fossilized mistakes that a tutor would catch in the first ten minutes. They have just never said those sentences in front of one.

![A loop of four steps. You make a small mistake mid-meeting, focused on the content. Nobody corrects it, because it would threaten your face in public. You don't notice it either, because the gap never reaches your attention. It sounds fine to you, so you say it again in the next meeting. After a year of this loop the mistake is fossilized (Selinker, 1972).](/blog/learning-english-vs-using-it-at-work/how-mistakes-fossilize.svg "Each step is polite or reasonable on its own. Together they keep the same mistake alive for years.")

## How does a tutor session compare with a work meeting?

| What matters                   | Tutor session                    | Work meeting                                 |
| ------------------------------ | -------------------------------- | -------------------------------------------- |
| Who sets the topic             | You, or the tutor                | The agenda                                   |
| Time to find a word            | As long as you need              | Under a second                               |
| Vocabulary you reach for       | Your best                        | Your safest                                  |
| Who notices a mistake          | The tutor, every time            | Usually nobody                               |
| What happens after the mistake | Correction and an explanation    | The conversation moves on                    |
| What is at stake               | Nothing                          | Your point, sometimes your credibility       |
| What you take away             | A rule you may not use next time | A feeling that it went badly, and no details |

> The lesson gives you a rule detached from the moment you needed it. The meeting gives you the moment, detached from any rule.

Learning happens when the two are in the same place, and in a normal week they never are.

![A two-by-two grid. The vertical axis is feedback on what you said, the horizontal axis is what is at stake. Tutor sessions and practice apps give feedback with no stakes: a kind environment. Work meetings have stakes and no feedback: a wicked environment. The top-right square, both at once, is feedback from the real call, where you improve. Arrows from the tutor and the meeting squares point into it.](/blog/learning-english-vs-using-it-at-work/feedback-quadrant.svg "Lessons and meetings each give you half of what you need. The goal is the square where you get both.")

## Why don't lessons transfer to the meeting?

First, the conditions are different. Baldwin and Ford (1988) reviewed decades of training research and found that skills transfer best when the practice conditions match the conditions of use. A calm one-to-one conversation about your weekend shares almost nothing with a nine-person call about a slipping deadline. The grammar is the same in both, but you have to retrieve it under very different conditions.

Second, correction in lessons is weaker than it looks. Lyster and Ranta (1997) recorded teachers correcting learners and found that more than half of all corrections were recasts, where the teacher simply repeats the sentence correctly. Recasts produced the lowest rate of learner uptake of any correction type, and many learners did not even register that they had been corrected. So even in the kind environment, a lot of the feedback slides past.

Third, pressure shrinks what is left. In a meeting [the work already takes most of your attention](#why-is-talking-at-work-so-different-from-talking-to-your-teacher), and nerves take a share of the rest. MacIntyre and Gardner (1994) showed that anxious learners perform worse at every stage of language processing: taking in input, processing it, and producing output. The rule you can recite to a tutor is still in your head, but there is no capacity left to apply it, so you fall back on the sentence shapes you can produce without thinking. That is why you "know better words" and still say the simple ones.

## Why don't practice apps close the gap either?

Role-play apps and AI conversation partners fix one half of the problem: they give you a place to talk with instant feedback. But the conversation is invented, so the stakes are zero, and you are back in the kind environment practising for a wicked one. These apps also have no memory of what you said at work on Tuesday, so they cannot tell you which of your mistakes keep coming back.

> [!IMPORTANT] The test for any practice method
> Does the feedback come from the conversation that mattered, or from a substitute for it?

## How do you close the feedback gap?

> [!NOTE] Feedback that closes the gap
>
> 1. It is about a real meeting.
> 2. It arrives soon enough that you still remember the moment.
> 3. It points at a specific sentence you said.

There are four ways to get it, from cheapest to most complete.

| Method                              | What it costs                       | What it catches                                                       |
| ----------------------------------- | ----------------------------------- | --------------------------------------------------------------------- |
| Record and review yourself          | Five minutes of listening per call  | Fillers and repetition well, grammar poorly                           |
| Bring meeting moments to your tutor | A note within ten minutes of a call | The sentences that went wrong, with the rule attached                 |
| Ask one colleague a narrow question | One question                        | An overused phrase, or whether your point came through                |
| Get a review of the real call       | A SpeakLens subscription            | Repeated grammar patterns, fillers, unclear sentences, a CEFR estimate |

### 1. Record and review yourself

If your call tool records, listen to five minutes of your own speaking time after the call. It costs nothing and it is better than no feedback. It has two limits: it takes real time every day, and you cannot hear the mistakes you do not know are mistakes, because the noticing problem Schmidt described applies to recordings too. This method catches fillers and repetition well, and grammar poorly.

### 2. Bring meeting moments to your tutor

Lessons work far better when the material comes from your work. Write down the sentence that went wrong within ten minutes of the call, while you still have it word for word; by the evening you will remember only that it went badly. Give your tutor three of these a week and the lesson becomes a review of your real English, with the rule attached to the moment you needed it.

### 3. Ask one colleague a narrow question

A broad "please correct my English" puts a colleague in an awkward position, and the answer will be "your English is great". A narrow question works better:

> [!TIP] Questions that get an honest answer
>
> - "Is there one phrase I overuse in meetings?"
> - "Did the point about the budget come through?"

It is easier to answer honestly, and it gets past the politeness that hides most feedback at work.

### 4. Get a review of the real call

[SpeakLens](/#features) is built for this. It runs on your laptop and follows only your voice during Zoom, Meet or Teams calls, so no bot joins and nobody else is reviewed. About a minute after you hang up you get a written review of what you said: the grammar patterns you repeated, the fillers and their rate, and for each unclear sentence a reconstruction of what you were trying to say, with the natural phrasing. Every meeting also gets a CEFR estimate, so you can watch the level of the English you use at work change over time.

In Hogarth's terms, that is kind-environment feedback on a wicked-environment conversation. The mistake from Tuesday's stand-up gets named on Tuesday, in the sentence it appeared in, and if it shows up again on Thursday the review says so.

## What changes when the feedback comes from the meeting

Three things usually change within the first month.

- Your repeated mistakes get names. Most people have five or six. Once you can see "dropped third-person s, nine times this week" the fix takes days, because you finally notice it while speaking.
- Your safe vocabulary gets challenged. The review shows the senior phrase you could have used in the sentence you said, so the better word is attached to a real moment and it sticks.
- Your level gets measured in your meetings. A quiz shows what you know, and your meetings show what you say. Those can be a full CEFR band apart, and only the second one decides how you come across at work.

If fillers are the part of the gap that bothers you most, the companion post on [how to stop saying "um" in meetings](/blog/how-to-stop-saying-um-in-meetings) gives a four-week routine built on the same idea: count it in the real call, then change it.

Your next meeting is already on the calendar. To get a review of it, [start with SpeakLens](/#start); pricing is on the [home page](/#pricing).

## Sources

- Baldwin, T. T. and Ford, J. K. (1988). Transfer of training: a review and directions for future research. _Personnel Psychology_, 41(1), 63–105.
- Brown, P. and Levinson, S. C. (1987). _Politeness: Some Universals in Language Usage_. Cambridge University Press.
- Hogarth, R. M. (2001). _Educating Intuition_. University of Chicago Press.
- Kormos, J. (2006). _Speech Production and Second Language Acquisition_. Lawrence Erlbaum.
- Levelt, W. J. M. (1989). _Speaking: From Intention to Articulation_. MIT Press.
- Lyster, R. and Ranta, L. (1997). Corrective feedback and learner uptake. _Studies in Second Language Acquisition_, 19(1), 37–66.
- MacIntyre, P. D. and Gardner, R. C. (1994). The subtle effects of language anxiety on cognitive processing in the second language. _Language Learning_, 44(2), 283–305.
- Schmidt, R. (1990). The role of consciousness in second language learning. _Applied Linguistics_, 11(2), 129–158.
- Selinker, L. (1972). Interlanguage. _IRAL_, 10(3), 209–231.
- Skehan, P. (1998). _A Cognitive Approach to Language Learning_. Oxford University Press.

## Frequently asked questions

### Why is my English better with a tutor than in meetings?

A tutor session removes the three things that make meetings hard: time pressure, an agenda you do not control, and something at stake. You also pick your own words. In a meeting you use your safest words at speed while tracking the content, so your level drops by about a band.

### Why do I forget English words in meetings that I know in lessons?

Because the meeting spends most of your attention on the work: the argument, the numbers, the people on the call. Finding a word in a second language needs attention too, and there is little left. The word has not gone anywhere; it just takes longer to reach than the conversation allows.

### Do colleagues notice my English mistakes in meetings?

Mostly yes, and they almost never say so. Correcting someone in front of others is a face-threatening act, so people let it pass as long as they understood you. That politeness is why the same mistake can survive for years at work.

### Should I stop taking English lessons if they do not transfer to work?

No. Lessons are still the best place to learn a rule or a phrase. Feed them with material from your real meetings, so the tutor corrects the sentences you said at work instead of the ones you produced in class.

### How can I get feedback on my English at work?

Record yourself and review, write down the sentence that went wrong within ten minutes of the call, ask one trusted colleague a narrow question, or use software that reviews your own voice from the real call. Whichever you choose, the feedback has to point at a specific sentence you said.
