The Glow-Up Was Never About Your Face

Vogue Beauty posted it this month. Reese Witherspoon on a sun lounger in a pink feather coat, the Legally Blonde still that has been doing this job for twenty-five years, and across it a headline: "Yes, the Post-Breakup Glow-Up Is a Real Thing." The caption made the argument plainly. "No, you're not imagining that post-breakup glow… It's not just a coincidence, and the explanation is rooted in emotion, biology, and more."

Eighty-five comments, two hundred and ten shares. Nobody in the replies was arguing.

And the thing is, the glow-up is real. I have watched friends walk out of six-year relationships and look, within a season, like a different and better-lit person. The part I want to argue with is the second half of that sentence — rooted in biology — because that is the weakest available explanation, the evidence for it is much thinner than the confidence with which it gets repeated, and there is a far more interesting explanation sitting right underneath it that almost nobody writes about.

Start with the word, because the word tells on itself

"Glow-up" did not come from beauty. It came from Chicago drill. Chief Keef released "Gotta Glo Up One Day" on 15 August 2013, and in it, glowing up means money. Status. Getting out. His label was Glo Gang, reformed from Glory Gangz, and Merriam-Webster's own etymology traces the term to a play on grow up crossed with glow or glory.

For its first four years the word had nothing to do with anyone's face. It meant escaping the circumstances you were born into.

Then, around 2017, it got picked up online and welded to the before-and-after photograph. Merriam-Webster now defines it as "a striking transformation in a person's appearance; a makeover." A word that meant I got out now means I got hot. That drift is not a footnote. It is the whole argument in miniature: a story about changing your life, quietly rewritten into a story about changing your appearance, and then sold back to the person it happened to.

The cortisol story does not survive contact with the research

The standard explanation goes: bad relationship, high cortisol, bad skin. Relationship ends, cortisol falls, skin clears. It is tidy, it sounds like science, and every joint in it is weaker than it looks.

I went looking for the study — a prospective one, measuring the same people's cortisol before and after a breakup. I could not find that it exists. What exists is cross-sectional work comparing people by marital status, and it points the wrong way. Chin and colleagues (2017, Psychoneuroendocrinology, n=572) found that previously married adults had higher cortisol than married ones, with flatter diurnal slopes. Weisman and colleagues (2014) found people in new relationships had lower daily cortisol than singles. If cortisol is the engine, it appears to be running in reverse.

The oil claim fares worse. Yosipovitch and colleagues (2007) put a sebumeter on 94 adolescents during exams and again on holiday and reported it flatly: "Sebum measurements did not differ significantly between the high stress and low stress conditions." Stress did track with acne — but not through oil. The "stress hormones drive oil production" line that circulates in beauty writing traces back to a 2002 PNAS paper on cultured human sebocytes. Skin cells in a dish, presented as if it described a face.

What is real is narrower and less glamorous. Garg et al. (2001, Archives of Dermatology) found that acute stress measurably slows skin-barrier repair — around 30% recovery at three hours during exams versus 45% on holiday. Twenty-seven students. Replicated. Genuine psychodermatology. But the honest framing is measurable on a lab instrument, not visible across a room.

The genre's founding document: Revenge Body with Khloé Kardashian, which ran on E! for three seasons from January 2017 — the same year "glow-up" stopped meaning money and started meaning appearance.

Some things genuinely do change

I do not want to overcorrect into denial, because the best-evidenced part of this is solid and it is worth having.

Josefsson and colleagues (2018) tracked 81,925 people across four waves over twelve years, using within-individual fixed effects — the strong design. After a separation, BMI fell (by up to 0.55 kg/m²) and physical activity rose. That is the glow-up, showing up in national data.

It also found the odds of smoking rose, up to 2.22 times. That is the price of citing the study honestly, and you will not see it in the trend pieces.

Sleep is real too, and more specific than people realise. Sundelin and colleagues (2013) measured what sleep deprivation actually does to a face: hanging eyelids, droopy mouth corners, swollen eyes, dark circles — and it reads to observers as sadness as much as tiredness. But the effect on rated attractiveness in Axelsson et al. (2010, BMJ) was 38 versus 40 on a hundred-point scale. Two points. Real, and small.

The thing that actually changes is not your face

Here is the research the magazines skip, and it is much better material.

Erica Slotter, Wendi Gardner and Eli Finkel published "Who Am I Without You? The Influence of Romantic Breakup on the Self-Concept" in 2010. What they found is that a breakup does not primarily damage your mood — it damages your self-concept clarity. Your sense of knowing who you are. And critically: it was the loss of clarity, not the amount of change, that predicted distress. People did not suffer because they changed. They suffered because they no longer knew what they were.

Appearance turns out to be one of the five domains people report changing. Not the point of the transformation — a symptom of it.

Then Larson and Sbarra (2015) ran something quietly remarkable. Two hundred and ten people, randomly assigned either to a normal before-and-after study or to one where they were repeatedly asked to describe themselves. The group who kept being asked who are you recovered better — and the mechanism was improved self-concept clarity. Being asked to articulate yourself, over and over, helped.

So the glow-up is real. It is just that the visible part is downstream. People who have rebuilt a sense of who they are stand differently, dress differently, and take up more room. We see that and call it skin.

And now the part that should give everyone pause

If the glow-up narrative were merely imprecise, fine. But there is a finding here with teeth.

Frazier and colleagues (2009, Psychological Science) compared how much growth people reported after adversity with how much they had actually demonstrated, measured prospectively. The two were largely unrelated. Worse: perceived growth correlated with increased distress, while actual growth correlated with decreased distress.

Owenz and Fowers (2019) then tested exactly this on breakups, with the right design — 599 people measured before any breakup happened, 100 of whom subsequently broke up, against a control group. Actual growth was unrelated to perceived growth, unrelated to distress, and unrelated to whether the breakup had occurred at all. Optimism predicted perceived growth. It did not predict real growth.

Read that as a sentence about the internet and it becomes uncomfortable. The glow-up, as a genre, is the performance of perceived growth. And the best available evidence says perceived growth is the version associated with feeling worse.

One more. Marshall (2012), with 464 participants, found that the most natural venue for performing all this — keeping an eye on an ex online — was associated with less personal growth, more distress and more longing.

The dietitian Abbey Sharp takes apart the methods behind the format that named the whole thing.

Diana did it once, and never explained

Every article about this eventually arrives at the same photograph, and it is worth arriving at properly.

29 June 1994. A Vanity Fair fundraising dinner at the Serpentine Gallery. A short black off-the-shoulder dress by Christina Stambolian. Diana wore it the same evening Charles's documentary aired — the one in which he admitted the affair, and said the marriage had irretrievably broken down.

What makes it the founding text is not the dress. It is the economy of it. One appearance. One photograph. No caption, no explanation, no follow-up, no series. She never said a word about it, and thirty-two years later everyone still knows exactly what it meant.

Compare that with the current form, where the transformation is documented in instalments — the haircut, the gym, the first night out, the six-month update — and the audience is not the ex but everyone. Fast Company put the "divorce glow-up" trend at over 28 million views across platforms this summer.

Diana's version had a recipient. The current version has a viewership. That is not a small difference.

The asymmetry nobody mentions

While researching this I noticed something that is itself the story. The pro-glow-up material is enormous and easy to find. The pushback is genuinely hard to locate. The Daily Dot's piece documenting the divorce-glow-up trend — the fullest account of it I found — does not contain a single critical voice.

The most useful thing anyone said came from a creator inside the trend rather than outside it. @mariah_lynn_clark, quoted by Fast Company:

Imagine thinking a woman glows up after divorce because she stopped trying, instead of because she finally stopped surviving.

That is sharper than most of the commentary about her.

The clearest sceptical voice belongs to Renee Engeln, the psychology professor who wrote Beauty Sick. Speaking to i-D about breakup haircuts, she made the point that the whole genre depends on:

how much we look to women's appearance to determine who they are or what's going on in their lives — Renee Engeln

And, flatly: "Changing our appearance is unlikely to lead to meaningful life changes."

There is a detail in a 2019 Vice photo essay that has stayed with me. The photographer Bex Wade shot people who had been through it and asked them to describe their own glow-up. Rosamund, 19: "I'm much happier — like a thousand times happier." Flo, 26: "I think my glow-up is that I'm just more fun." Mercedes, 18: "I think my glow-up is the way I act now."

Not one of them described it as physical. The subjects of the genre do not talk about it the way the genre talks about them.

One of the few widely-watched arguments against the genre, published in May 2026.

What I would actually say to someone in it

The evidence, taken together, supports something quite different from the trend piece.

Most people recover, and faster than they expect. Eastwick and colleagues (2008) found the forecasting error is in the predicted peak of the pain, not in how long it lasts — people are wrong about how bad it will get, not about how long it takes. Acolin et al. (2023) found symptoms back at pre-breakup levels within three months for most.

Though not for everyone, and it would be dishonest to leave it there. Lucas (2005), following more than 30,000 people for eighteen years, found that after divorce "the return to baseline is not complete." Some of this leaves a mark. A narrative in which everyone emerges glossier within the quarter has no room for that, and the people it has no room for are the ones who most need not to feel like they are failing at it.

The strongest single predictor of recovery in this whole literature is not a haircut. Sbarra, Smith and Mehl (2012) found that judge-rated self-compassion predicted less emotional intrusion nine months on, and it survived against twelve competing predictors.

Self-compassion. Not self-improvement. The two are frequently confused and they are close to opposites: one asks less of you, the other asks more.

So, is it real?

Yes — and the honest version is better than the sold one.

Something does happen to people after a relationship ends. They sleep differently, move more, and rebuild a sense of who they are, and that rebuilding is visible in a way that is difficult to name and easy to mistake for skincare. Vogue is right that it is not your imagination. It is just that the mechanism is not cortisol; it is the slow, unphotogenic work of becoming legible to yourself again.

The part I would resist is the obligation. The glow-up has quietly become a duty — evidence you must produce, on a schedule, in public, to demonstrate that you are winning. The research says that particular version, the performed one, is associated with feeling worse rather than better.

Buy the dress if you want the dress. Get the fringe. It is fun, and fun is not nothing at a time when very little is.

But you do not owe anybody an after photograph. Diana never posted a follow-up.

The counter-argument, made at length.

A note on the sourcing

Every study named here was checked against its primary citation. Where the popular claim outruns the evidence, we have said so rather than smoothed it over — in particular, we could find no prospective study measuring cortisol across a breakup, which is the load-bearing gap beneath the most repeated explanation in this entire genre.

Two limits worth stating. The skin-stress studies are small and mostly run on students under exam pressure, not people going through separations. And the Vogue post that prompted this piece was read from the post itself; we have not reproduced the article behind it.

If a relationship ending has left you struggling in a way that is not lifting, that is worth talking to your GP or a therapist about — and it is a far more useful response than a haircut. Videos are shown via YouTube's official embed and remain the property of their creators.