Nine Meetings. Nobody Asked What You Thought.

Load  ·  The Reading Room, Vol. 30  ·  7 min read  ·  Where this has actually been measured, what the numbers were, and the one thing that works better than speaking up more

Based on the published research of David Neumark, PhD — Chancellor’s Professor of Economics, University of California, Irvine, and Research Associate, National Bureau of Economic Research. NBER profile

Feeling invisible at work is almost impossible to prove from inside a meeting. In one field experiment with more than 40,000 applications, the same pattern was measured cleanly, and it was worse for women.

I said the thing about the reporting cycle at eleven minutes past ten. Nothing. Conversation moved on like a car going round a parked bicycle.

At twenty-six minutes past, a man eleven years younger than me said a version of the same thing, slightly worse, and the head of the department said “that’s a good point, let’s pick that up.”

I want to be honest about what happened next, because it is the actual subject of this piece. I did not think “that is a familiar pattern.” I thought “I must not have said it clearly.” I spent the rest of the meeting drafting a better version of a sentence I had already said perfectly well.

The short answer: feeling invisible at work is very hard to establish from the inside, because the evidence is always a room full of things that did not happen. What can be established is what happens in hiring, where researchers can send matched applications and count the replies. In a field experiment covering more than 40,000 applications, callbacks for older women applying to administrative roles ran at 7.6 per cent against 14.4 per cent for young women with equivalent CVs.¹ The same study found the pattern for men was weaker and less clear.¹ That is not proof about your Tuesday. It is proof that the thing you are describing has been measured, in the one place it can be, and it was there.

Is feeling invisible at work real, or am I imagining it?

Both questions are reasonable and only one of them can be settled, which is the trap.

From inside a room, the evidence for being overlooked is entirely negative. A question that was not asked. A follow-up that went to somebody else. A piece of work you shaped that got summarised upward without your name on it. None of that is falsifiable, and every single item has an innocent explanation available.

So a competent woman does the competent thing, which is to discount the pattern and look for her own error. That is why the meeting ends with you rewriting a sentence rather than noticing a shape.

The field experiment matters because it removes the innocent explanations. Researchers sent applications that were identical except for the age they implied, in enormous numbers, and then simply counted who got called back.¹ No interpretation, no meeting, no room. Just replies.

What the numbers were

The experiment covered more than 13,000 positions in 12 cities across 11 states, with over 40,000 applications sent in total.¹

For administrative jobs, applications from women aged 29 to 31 produced callbacks 14.4 per cent of the time. Applications from women aged 64 to 66, with matched experience, produced callbacks 7.6 per cent of the time.¹ In sales roles the gap was 28.7 per cent against 18.4 per cent.¹

For men, the same design produced a less clear picture, and the researchers said so plainly: robust evidence of age discrimination in hiring against older women, and considerably less evidence of it against men.¹

That asymmetry is the part worth carrying. The thing being described is not simply about being older. It is more visible for women, and it was measured that way.

If the exhausting part is not the meeting but the hours afterwards spent auditing your own delivery, that is the load worth looking at. The Quiet Audit is a private ten-minute pass through what you are carrying home. Start here.

Present is not the same as attributable

This distinction is the one that changes what you do on Monday.

Most advice for this problem lives entirely in the left column: speak up sooner, project more, do not soften it. That advice asks you to win a contest for attention in a room where attention is the scarce thing. The right column does not require the room to cooperate. The same shift matters after a setback, and it is a close cousin of looking fine while nothing lands.

The four minutes the night before

Send one paragraph before the meeting, not after it.

Three or four sentences to the person who will chair, or to the thread everyone is on. Here is the thing I think we should decide. Here is why. Here is what I would recommend. Send it the evening before or first thing that morning.

What this does is put the idea into the record with your name attached before anybody has to notice you saying it. When the man eleven years younger says the worse version at twenty-six minutes past, it is now a restatement of something already written down rather than a fresh contribution, and at least one person in the room knows that without having to be told.

It also removes the part that costs the most, which is having to be heard in real time by eleven people. You are no longer competing for the room. You are supplying a document. Written channels have their own hazards, and this is one of the few places where they work in your favour.

Once, this week, for one meeting that matters. Not a system. One paragraph.

Who else has measured this

Correspondence experiments — matched applications sent, replies counted — are a small and rigorous field. A second landmark study used the same method on a different variable.

Marianne Bertrand, PhD (faculty profile · NBER profile), Chris P. Dialynas Distinguished Service Professor at the University of Chicago Booth School of Business, co-authored the widely cited “Are Emily and Greg More Employable Than Lakisha and Jamal?” resume-audit study, establishing the same correspondence-experiment method Neumark’s team later applied to age.²

Different variable, same design, same conclusion about how hiring discrimination actually gets proven: not from inside the room, but by counting who gets called back.

Conversation moved on like a car going round a parked bicycle.

Questions worth having an answer to

Is this definitely about age rather than about me?
Neither the study nor this article can answer that for your workplace. What the study establishes is that the effect exists and is measurable where measurement is possible, and that it falls harder on women.¹ That is a reason to stop treating your read as automatically wrong. It is not a reason to stop asking for specific feedback.

Should I raise it with HR or my manager?
That depends entirely on your organisation and your evidence, and this is a decision worth taking advice on locally rather than from an article. The written record you build with the pre-meeting paragraph is useful either way, and it is useful for reasons that have nothing to do with a complaint.

Doesn’t sending something in advance look like I am overpreparing?
It looks like a briefing note. Senior people send them constantly and nobody has ever read one as insecurity. Two paragraphs is a briefing note. Two pages is overpreparing.

What if the pattern is real and nothing changes?
Then you have a decision to make with better information than you had, which is a different position from the one where you assume the problem is your delivery. Some of these situations are worth staying in and some are not, and that judgement is yours.

Informational, not medical advice. Concerns about workplace treatment belong with HR, a manager or an employment adviser, not with this article.

On the researcher. David Neumark is a labour economist, not an employment lawyer: his field experiment measures callback rates in hiring, which is a different setting from an existing job, and it cannot diagnose any individual workplace.

References
1. Neumark, D., Burn, I., & Button, P. Is It Harder for Older Workers to Find Jobs? New and Improved Evidence from a Field Experiment. Journal of Political Economy, 127(2), 2019. Callback figures as reported in Neumark & Button, FRBSF Economic Letter 2017-06. Read the summary · Read the working paper

Written by Jo Brennan for Blue Leaf Journal. Updated: 30 August 2026.