Ruth Wooderson came on Corey-osity Unleashed with Deb and me in 2025. One of her ideas has stuck with me ever since:
Glass ceilings still exist. Now they're frosted.
The old ceiling was clear glass. You saw the boardroom through it. You saw who sat there, and you knew why you weren't in the room. People said the quiet part out loud. "We need a man for this client." Ugly, yes. Also visible.
The frosted version is harder to fight. You still see shapes moving above you. You know a room exists up there. You don't see the door.
Nobody says the ugly thing anymore. Companies have values posters, women's networks and allyship training. Most people mean well. And the numbers barely move.
I'm a man. I've spent my whole career in tech. Nobody has ever questioned my judgment because of my gender. So I don't trust my own gut on this one. My experience tells me nothing about hers. I need data.
The First Step Is the Broken One
McKinsey and LeanIn.Org have run the Women in the Workplace study every year since 2015. The 2024 edition pulled data from 281 organizations employing over 10 million people, and surveyed more than 15,000 employees.
Here's the line I keep rereading. For every 100 men promoted to manager in 2018, 79 women got the same step. In 2024, 81 women did.
Six years. Two women per hundred.
The report calls this the "broken rung." The blockage isn't the CEO chair. It's the first step up to manager. Miss the first rung and you never reach the ones above it. Men outnumber women at manager level, so the senior pipeline starves years before anyone in the C-suite makes a decision.
And before anyone says women don't want it... the same report found seven in 10 women want a promotion to the next level this year. The same rate as men. Among younger women, more than eight in 10.
The ambition shows up. The promotion doesn't.
Who Cleans Out the Fridge?
Here's a frosted pane you've walked past a hundred times.
Every office has work nobody gets promoted for. Planning the holiday party. Taking notes. Sitting on the committee nobody wants. Cleaning out the break room fridge.
Linda Babcock, Lise Vesterlund, Maria Recalde and Laurie Weingart studied this work in a 2017 paper for the American Economic Review. In an interview with the AEA, they described it as the tasks everyone knows need doing, where nobody pays attention to how well you do them.
Their lab test stripped away every excuse. Groups of three. Two minutes. Anyone had the option to click a button and volunteer. If nobody clicked, everyone got a dollar. If someone clicked, the volunteer got $1.25 and the other two got $2 each. Clicking a button takes no special talent. Nobody is "naturally better" at it.
Women volunteered about 34% of the time. Men volunteered about 23% of the time.
Then the part closest to home. Women were also more likely to be asked. And when asked, they were far more likely to say yes.
The researchers spotted the pattern in their own careers first. They even formed a club to talk it through, named after their inability to say no. Babcock admitted they had become part of the problem. When they needed someone for a thankless job, they asked women first, because they expected a yes.
No policy says "give the grunt work to women." Nobody decides it. The ask happens in a hallway, in a Slack message, in a meeting where somebody says "Who'll take notes?" and the room looks at the same person.
Do this for three years and she has lost hundreds of hours. Hours she would have spent on the work someone notices at promotion time.
"You Had a Great Year"
Performance reviews look neutral on paper. Same form. Same rating scale.
Shelley Correll and Caroline Simard at Stanford read written reviews from three high-tech companies and one professional services firm. Slate covered their findings from the Harvard Business Review piece they wrote.
Praise for women leaned on generic lines like "You had a great year." Developmental feedback for men tied to business outcomes 60% of the time, against 40% for women. The phrase "too aggressive" showed up roughly three times as often in women's reviews.
Both male and female supervisors gave men more specific feedback. So this isn't about a few biased managers.
Picture the calibration meeting. Your manager walks in to argue for your promotion. Your file says "great year." The guy next to you has a file full of shipped projects with numbers attached. Who wins the argument?
Nobody in the room acted with bad intent. The frost went on months earlier, in the writing.
I wrote recently about how your silence is feedback. Vague praise belongs in the same family. It feels kind. It leaves people with nothing to build on and nothing to point to. And the "too aggressive" label ties straight into what I wrote in Bold Is Not Loud. We keep mistaking volume for confidence, and we judge the same behavior differently depending on who does it.
A Thousand Small Cuts
The McKinsey report tracks everyday slights too. Women remain twice as likely as men to be interrupted or spoken over. Women also face far more comments undermining their competence, such as having their judgment questioned in their own area of expertise. The report's verdict on this one: it "has not improved over time."
None of these moments looks like discrimination on its own. Each one is deniable. "He interrupts everyone." "She volunteered." "The review was positive."
Deniability is the frost.
Good Intentions Don't Clear the Glass
Here's my uncomfortable opinion. Progressive companies are often the worst at seeing this, because they believe they've already fixed it.
They ran the training. They launched the network. They put a woman on the leadership team page. Box checked.
Meanwhile the 2024 report found companies have scaled back mentorship, sponsorship and career development programs aimed at women. And relatively few companies track outcomes for the programs they do run.
You don't fix what you don't measure. You don't measure what you think you've already fixed.
What I'd Do on Monday
If you manage people, here's where I'd start. None of this needs a budget.
Count your first promotions
Pull the last three years of promotions into first-line manager roles. Split them by gender. If the ratio isn't close to 100 to 100, ask why before anyone gives you a reason.
Track the fridge
For one month, write down every non-promotable task on your team. Who took notes. Who organized the offsite. Who onboarded the new hire. Who fixed the build nobody owned. Then stop asking for volunteers. Assign the work, and rotate it.
Rewrite your reviews before you submit them
For every person, name one specific business outcome they delivered. If a review has no outcome in it, you haven't looked hard enough. Then search your drafts for words like "aggressive," "abrasive" and "emotional." Ask yourself whether you'd write the same word about a man who did the same thing.
Keep a tally in your next meeting
Count who gets interrupted. Count whose idea gets credited to the person who repeated it. Do it once and you'll never unsee it.
Wipe the Glass
Frost doesn't melt by itself. Somebody has to wipe it, and the people with the cloth are the ones sitting above it.
So here's my question for you this week. Who on your team cleaned out the fridge last month... and who got the stretch project?