Short answer
Stop measuring fixed traits and work on what changes: clothes that fit, grooming, strength, open posture, sleep, warmth and humor. Each has research behind it, mostly photo ratings and short encounters, so treat it as averages that predict no single woman. Stated preferences predict live attraction poorly. Be presentable, kind and in the room.
Key takeaways
- The changeable inputs with research behind them are clothes that fit, grooming, strength, open posture, sleep, warmth and humor.
- Most attraction studies use photo ratings or four-minute dates, so they describe averages across raters and predict no single woman.
- Looks count for both sexes when people meet, yet couples who knew each other longer before dating were less matched on looks.
- A larger repeat of the power-pose study found no hormone change, and orthodontists say no scientific evidence shows that mewing reshapes a jawline.
- A model fed more than 100 traits and preferences could not predict which pairs would click, so be presentable, be kind and meet more people in person.
Start with the visible basics in the free plan

What women find attractive: strong, mixed and myth
No single list fits every woman. Across large samples, three findings hold up best: people rate kindness, health and intelligence above looks and money when asked what they want, looks count about equally for women and men once they meet in person, and several things you can change shift how you are rated. The myths are the formulas.
| Claim | Evidence | Verdict |
|---|---|---|
| Kindness, health and intelligence rate above looks and money in what people say they want | 14,399 people in 45 countries, replicating a 37-culture study [1] [2] | Supported, as stated preferences |
| Looks count in person, for women about as much as for men | Meta-analysis of 97 samples [3] | Supported |
| Clothes that fit change first impressions | 274 raters, five-second photos [4] | Supported, for confidence and status |
| Looking strong raises ratings of a man's body | Photos of 192 men, faces hidden [5] | Supported, for photos |
| Open posture helps in short encounters | 144 speed dates and a dating-app test [6] | Supported, in one paper |
| Looking rested helps | Photos before and after sleep loss [7] [8] | Supported, small effect |
| Being funny helps | 51 pairs of strangers in conversation [9] | Mixed |
| A power pose raises testosterone | A repeat test with 200 people found no change [10] | No evidence |
| Mewing reshapes an adult jawline | Orthodontists' association statement [11] | No evidence |
| A look or a line that works on any woman | A model trained on 350 speed daters [12] | No evidence |
Read the verdicts narrowly. "Supported" means the effect showed up in the study cited, usually as an average across raters who looked at photos or met someone for four minutes. It says nothing about the size of the effect in your life, and nothing about what one particular woman will think of you.
The stated-preference finding deserves a closer look, because it runs against what the forums say. In the 45-country survey, people rated an ideal long-term partner on five traits, each on a scale from 1 to 7. Women's average for kindness was 6.23 and men's was 6.12. Looks came in at 5.56 for women and 5.85 for men. Both sexes put kindness, intelligence and health above good financial prospects and above physical attractiveness [1]. The 2020 team reports that the original 37-culture study, with 10,047 participants, also had both sexes ranking kindness and intelligence highest [1] [2]. These are questionnaire answers about an ideal, though, and the section on stated preferences below explains why that matters.
The sections below sort the changeable inputs by how long they take. For the wider rebuild, see the Rebuild yourself guides.
What you cannot change, and why to stop measuring it
Height, bone structure and age are fixed, so every hour spent measuring them is an hour taken from the things that move. The research gives two more reasons to let them go: they are not hard filters, and they count for less once people know you.
Take height first, since it is the one men ask about most. In a study of 5,782 speed daters making 128,104 choices, women were most likely to pick a man about 25 cm (10 inches) taller than themselves, and men a woman about 7 cm (3 inches) shorter. Mutual matches, though, were most common at a difference of 19 cm (about 7.5 inches), which was neither sex's preference [13]. An average preference is real. It is also something people trade against everything else in the room.
Then time. Coders rated the physical attractiveness of both partners in 167 couples. Couples who got together soon after meeting were more closely matched on looks. Couples who had known each other longer, or had been friends first, were less matched [14]. It is one correlational study of people already in relationships, so it cannot prove cause. The pattern is still useful: first-glance rankings count most between strangers, and less between people who have shared a room for months.
And looks do count. Across 97 samples, physical attractiveness predicted romantic interest with a moderate-to-strong correlation of about .40 for both sexes, against roughly .10 for earning prospects [3]. A correlation of that size means appearance matters and still leaves most of the variation unexplained. The part of appearance you control is covered in the next two sections.
If thoughts about your face or body take up hours of your day, or keep you from going out, that is more than a style problem. Talk to a doctor. If it ever tips into feeling hopeless, call your local crisis line.
This week: fit, grooming, smell
The fastest changes sit on the surface: clothes that fit, a maintained haircut and beard line, and smelling clean. All three can be handled in seven days, mostly with what you already own.
Fit
In the suit study, 274 people saw the same man, face pixelated, for a maximum of five seconds in a made-to-measure suit and in a similar off-the-rack one. The tailored version was rated higher on confidence and expected salary [4]. It used one model and one pair of suits, and it measured impressions of success, not attraction. It is still the best evidence we found that cut registers fast. The 10-minute mirror test in how clothes should fit a man shows you what passes.
Grooming
Research on grooming itself is patchy. Studies of facial hair disagree with each other, and we found none comparing the same man groomed and ungroomed. So this part rests on convention and common sense: a haircut every few weeks, a clear beard line or a clean shave, clean nails and teeth. The routine, the barber script and the beard studies are covered in men's grooming and style basics.
Smell
In a double-blind test, men given a deodorant spray with active ingredients reported more self-confidence over several days than men given an inactive spray. Women who then watched video clips of them, with no smell involved, rated the active-spray group as more attractive. Still photos showed no difference [15]. It is a single study, the abstract gives no sample size, and two of the six authors worked for Unilever, which makes deodorants. The modest takeaway is that being clean changes how you carry yourself.
Put together, the week looks like this:
- Run the mirror test on the outfit you wear most, and alter or replace what fails.
- Book a haircut, and set a beard line or shave.
- Lock the daily routine: shower, deodorant, teeth, fresh shirt.
This quarter: strength, sleep, posture
Over three months, three slower inputs shift: how strong you look, how rested you look and how you hold yourself. Each has rating studies behind it, set out below with their limits.
Strength
What the research says
Sell, Lukaszewski and Townsley (2017) used photos of two groups of male students, 61 shirtless and 131 in tank tops, with faces obscured. Student raters, most of them women, estimated either how strong each man was or how attractive his body was. Rated strength accounted for 61 to 73 percent of the differences in body attractiveness, depending on the photo set, and adding height and weight brought that to about 80 percent. None of the 160 women who rated attractiveness showed a significant preference for weaker men [5]. The limits: bodies in photos, student samples, no faces and no conversation, and the authors note that bodily attractiveness is only one small part of a man's overall appeal. Verdict: Supported, for bodies in photos.
That is not a case for chasing size. An earlier set of six studies found moderately muscular men rated most attractive. Women in those studies rated muscular men as sexier than non-muscular men, and also as more dominant, more volatile and less committed to their partners [16]. The 2017 photo study saw no such ceiling [5]. Both agree on the part that matters here: a visibly trained build rates better than an untrained one.
The practical target is ordinary. US public-health guidance for adults is muscle-strengthening activity on two days a week, plus at least 150 minutes of moderate activity [17]. That is a health guideline, not an attraction prescription, and it is enough to start. The workout plan after a breakup lays out the sessions. If you have a health condition or have not trained in years, check with a doctor first, and skip anything sold as a shortcut.
Sleep
In a Stockholm lab study, 23 adults were photographed after a normal night and again after 31 hours awake following a short night. Sixty-five observers rated the tired faces as less healthy and less attractive: 38 against 40 on a 100-point scale for attractiveness [7]. A later study photographed 25 people after two nights of about four hours in bed. The 122 raters found them less attractive and less healthy, and were less willing to socialize with them [8]. The effects are small. The second team put theirs at 0.09 to 0.26 points on a seven-point scale [8], and both are photo studies.
The target is simple: adults aged 18 to 60 need seven or more hours a night [18]. If you have barely slept for weeks, see a doctor.
Posture
Open posture is the body-language finding with the most direct evidence we found for first meetings, and it rests on a single paper. Researchers coded video of 144 four-minute speed dates among 24 people. Each step up on their scale of postural expansiveness (one standard deviation, meaning arms and torso more open and taking up more space) went with 76 percent higher odds of a yes from the other person. The paper's own summary rounds that to "nearly doubling". They then ran a test on a dating app in the San Francisco Bay Area, using six people who each had an open-posture profile and a closed-posture one. Across about 3,000 responses, the open profiles were 27 percent more likely to get a yes. The three men drew only 30 yeses from 1,500 views, and 26 of those went to the open photos [6].
The speed-dating study is small and correlational, so people may have opened up because the date was already going well. The app test used three men and three women, all white, in one metro area, and the authors call the difference between men and women tentative [6]. We did not find an independent replication.
In practice: shoulders back and down, arms uncrossed, hands out of your pockets, head level. In photos, face the camera with your arms away from your body. Open is not the same as sprawling or crowding her. Where that line runs is covered in how to flirt without being creepy.
Ongoing: warmth, humor, social ease
The trait people rate highest in a partner, kindness, is a behavior you practice. It only shows in conversation, so the ongoing work is getting into more conversations and being good company in them.
Warmth
Kindness led the ratings in the 45-country survey [1]. In a first conversation it looks like small things: you remember what she said two minutes ago, you ask a follow-up, you are decent to the staff, you take a no without sulking. None of that is a technique. Better conversations breaks down the listening part.
Humor
In one study, 51 pairs of single students who had never met talked for about 10 minutes. The more often the man attempted humor, and the more of those attempts the woman laughed at, the more interested she was in dating him. Laughing together went with interest for both of them [9]. The effects were small. The share of his attempts that got a laugh made no difference, and the design cannot say whether the humor created the interest or the interest produced the laughter. So skip the rehearsed material. Notice what is funny in the room and say it.
Social ease
Ease comes from repetitions, and repetitions come from being in a room with the same people every week: a class, a club, a team. That is also where the couples study points, since longer acquaintance loosened the grip of looks [14]. Where to meet women lists settings that work, and how to get a girlfriend puts the steps in order.
Does looksmaxxing work? What to keep and what to skip
The basic half works, because it is ordinary self-care under a new name. The extreme half has no good evidence behind it, and parts of it can hurt you. No study has tested "looksmaxxing" as a program, so judge each piece on its own.
| Keep | Skip |
|---|---|
| Clothes that fit | Mewing to reshape your jaw |
| Haircut, beard line, skin and teeth routine | Hitting or pressing on your face ("bone smashing") |
| Strength training and enough sleep | Drugs or hormones taken for looks |
| Open posture | Surgery chosen because a forum rated your face |
| Time in rooms with other people | Rating scales and measuring your features |
On mewing, the American Association of Orthodontists said in 2024 that no scientific evidence supports the claim that tongue posture reshapes the jawline, and warned that the practice can loosen teeth, change your bite and affect speech [11]. That is a professional body's statement and not a trial, but it is the clearest expert assessment available. Verdict: No evidence.
We do not give advice on surgery, medication or hormones. If you are considering any of them, that conversation belongs with a qualified doctor, starting with your doctor.
The rating culture is the part to drop first. It trains you to see a number in the mirror. In the speed-dating data, nothing people reported about themselves could predict who clicked with whom [12], and among couples, matching on looks was weaker the longer two people had known each other first [14].
Does power posing work?
Not in the way it was sold. A larger repeat of the original study found no change in hormones from holding a "power pose". What holds up is smaller: avoiding a slumped posture goes with feeling better, and open posture reads well to other people.
What the research says
Carney, Cuddy and Yap (2010) reported that two one-minute expansive poses raised testosterone, lowered cortisol and increased risk tolerance in 42 people [19]. Ranehill and colleagues (2015) repeated the test with 200 people and found no effect on testosterone, cortisol or risk-taking, though the posers did report feeling more powerful [10]. A 2022 meta-analysis led by Körner pooled 128 experiments with almost 10,000 participants. It found effects of body position on how people saw themselves, weaker effects on behavior and none that held up for hormones or other physiology [20]. A second 2022 meta-analysis, of 73 studies, found that the difference came from contracted, slumped positions (g = 0.45 against neutral) and not from expansive ones (g = 0.06 against neutral) [21]. The limits: most experiments lacked a neutral control group and measured short-term effects in a lab. Verdict: No evidence for the hormone claim. Supported for the smaller claim that slumping goes with feeling and acting worse.
So skip the two minutes in front of the bathroom mirror. Fix your default instead: how you sit at your desk, how you stand in a line, how you walk into a room. That default is what other people see, and in the speed-dating data it was the posture people held during the date that went with a yes [6].
Why stated preferences predict so little
What people say they want and whom they go for in person are only loosely connected. That is good news for you, because nobody can rule you out on paper and nobody can sell you a formula.
Two studies make the case. In the first, 163 undergraduates, 81 of them women, each went on 9 to 13 four-minute speed dates. Beforehand, the men said looks mattered more to them and the women said earning prospects mattered more. In the room, those differences disappeared, and what a person had said they wanted did not predict whom they wanted [22].
In the second, 350 undergraduates in two samples completed more than 100 self-report measures of their traits and preferences before their speed dates. A machine-learning model could predict part of who was liked by most people (7 to 27 percent of that variance) and part of who liked most people (4 to 18 percent). It could not predict any of the attraction that was specific to a pair [12].
Both studies used students on four-minute dates, so they describe first impressions and say little about how attraction grows over months [22] [12]. Within that limit, they point to three working rules:
- General appeal is partly predictable. That is the case for the basics above.
- The click between two people is not. The only way to find it is to meet more people in person.
- "Women want X" is weak evidence. There are averages, and then there is the woman in front of you, who has not read the study.
Do this today
Pick your first visible upgrade. Put on what you would wear to a first date, stand in front of a full-length mirror and take two photos, front and side. Mark four things pass or fail: fit, hair and beard line, posture, and whether everything is clean. The first fail is your upgrade for this week. Ten minutes.
Then let the plan set the order. The free four-week plan's meet-someone-new path starts with "Fix the fit", "Get the cut" and "Find your max" in Week 1: a mirror check on what you own, the barber booked today and one all-out set of push-ups that sets your daily number. It moves on to conversations and dates after that. One step at a time, no account, progress saved on your device. Start with the visible basics in the free plan.
Questions men ask
Can an average-looking guy become attractive?
Yes, in the ways that are open to change. Fit, strength, posture and sleep all shift how people are rated in studies, and none of them needs a different face. Time works for you too. In a study of 167 couples, partners who had known each other longer before dating were less matched on physical attractiveness. People who get to know you judge more than a first glance. Put yourself where that can happen.
Do looks matter more than personality?
Looks matter, and about equally for men and women once people meet in person. A meta-analysis of 97 samples found physical attractiveness predicted romantic interest with a moderate-to-strong effect for both sexes. That still leaves most of the variation unexplained, and in a survey across 45 countries both sexes rated kindness and intelligence above looks in an ideal partner. Handle the visible basics, then let warmth and conversation do their share.
Do women care about height?
On average, yes. In a study of 5,782 speed daters, women were most likely to choose a man about 25 cm (10 inches) taller than themselves. The same study found that actual matches happened at a smaller height gap than women's preference, so it is not a hard filter. Height is fixed. Stand straight, wear clothes that fit and spend your effort on what moves.
Does going to the gym make you more attractive?
It moves one well-studied input. When raters judged photos of men's bodies with the faces hidden, how strong the men looked accounted for 61 to 73 percent of the differences in body attractiveness ratings. That is bodies in photos, rated by students, and the authors call it one small part of a man's overall appeal. Train for strength on two or more days a week and give it a few months.
Is confidence really attractive?
The visible part seems to be. In one paper, speed daters who sat in open, expansive postures were more likely to get a yes, and on a dating app, open-posture photos drew more matches than closed ones. What does not hold up is faking it with a power pose to change your hormones. Build confidence the slow way: competence, kept promises and practice talking to people.
Sources
Every source below was opened during drafting and review. The note under each one says who was studied and what it cannot show.
Walter, Conroy-Beam, Buss, et al. (2020). Sex Differences in Mate Preferences Across 45 Countries: A Large-Scale Replication. Psychological Science, 31(4), 408-423.
14,399 adults in 45 countries (median age 25, student and community samples) rated an ideal long-term partner on five traits; these are stated preferences on a questionnaire, not real choices.
Buss (1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12(1), 1-14.
10,047 people in 37 samples from 33 countries ranked and rated traits in a potential mate; questionnaire answers from the 1980s, and we confirmed the kindness and intelligence ranking through the 2020 replication's summary of it.
Eastwick, Luchies, Finkel, Hunt (2014). The predictive validity of ideal partner preferences: A review and meta-analysis. Psychological Bulletin, 140(3), 623-665.
Meta-analysis of 97 samples from attraction and relationship studies; correlations pooled across very different designs, describing averages and not any one person.
Howlett, Pine, Orakçıoğlu, Fletcher (2013). The influence of clothing on first impressions: Rapid and positive responses to minor changes in male attire. Journal of Fashion Marketing and Management, 17(1), 38-48.
274 online participants, most of them women, rated one faceless male model in two similar suits for up to five seconds; one model, small effects, and it measured impressions of confidence and success, not attraction.
Sell, Lukaszewski, Townsley (2017). Cues of upper body strength account for most of the variance in men's bodily attractiveness. Proceedings of the Royal Society B, 284(1869), 20171819.
Photos of 61 and 131 male US students with faces obscured, rated by student volunteers in Australia and the US (160 women among the attractiveness raters); bodies in photos only, with no faces, voices or conversation.
Vacharkulksemsuk, Reit, Khambatta, Eastwick, Finkel, Carney (2016). Dominant, open nonverbal displays are attractive at zero-acquaintance. Proceedings of the National Academy of Sciences, 113(15), 4009-4014.
144 video-coded speed dates among 24 people, then a dating-app test with six profile owners and about 3,000 responses; the first study is correlational and the second used only three men, in one US metro area.
Axelsson, Sundelin, Ingre, Van Someren, Olsson, Lekander (2010). Beauty sleep: experimental study on the perceived health and attractiveness of sleep deprived people. BMJ, 341, c6614.
23 adults photographed after a normal night and after sleep deprivation, rated by 65 observers; a small lab study with a heavy dose of sleep loss and a small difference in attractiveness ratings.
Sundelin, Lekander, Sorjonen, Axelsson (2017). Negative effects of restricted sleep on facial appearance and social appeal. Royal Society Open Science, 4(5), 160918.
25 adults photographed after two nights of about four hours in bed and after normal sleep, rated by 122 people; a small sample, small effects, and photo ratings by strangers.
Hall (2015). Sexual Selection and Humor in Courtship: A Case for Warmth and Extroversion. Evolutionary Psychology, 13(3).
Study 3 observed 51 pairs of single US students, strangers who talked for 10 to 12 minutes; correlational, so it cannot show that humor caused the interest.
Ranehill, Dreber, Johannesson, Leiberg, Sul, Weber (2015). Assessing the Robustness of Power Posing: No Effect on Hormones and Risk Tolerance in a Large Sample of Men and Women. Psychological Science, 26(5), 653-656.
200 men and women assigned at random to high-power or low-power poses; one lab's replication, measuring hormones, risk-taking and self-reported feelings of power.
American Association of Orthodontists (2024). American Association of Orthodontists Warns Against Popular 'Mewing' Trend. American Association of Orthodontists (press release, January 22, 2024).
A professional association's public statement, not a study; it says the jawline claims lack scientific evidence and lists dental and speech risks.
Joel, Eastwick, Finkel (2017). Is Romantic Desire Predictable? Machine Learning Applied to Initial Romantic Attraction. Psychological Science, 28(10), 1478-1489.
Two speed-dating samples of 350 undergraduates (163 and 187) on four-minute dates, each with more than 100 self-report measures; first impressions among students only.
Stulp, Buunk, Kurzban, Verhulst (2013). The height of choosiness: mutual mate choice for stature results in suboptimal pair formation for both sexes. Animal Behaviour, 86(1), 37-46.
5,782 speed daters making 128,104 choices; brief speed-dating meetings, so it shows who gets picked at a first meeting and not who ends up together.
Hunt, Eastwick, Finkel (2015). Leveling the Playing Field: Longer Acquaintance Predicts Reduced Assortative Mating on Attractiveness. Psychological Science, 26(7), 1046-1053.
167 couples whose physical attractiveness was rated by coders; a correlational study of people already in relationships, so it cannot prove that time causes the pattern.
Roberts, Little, Lyndon, Roberts, Havlicek, Wright (2009). Manipulation of body odour alters men's self-confidence and judgements of their visual attractiveness by women. International Journal of Cosmetic Science, 31(1), 47-54.
Men given an active or an inactive deodorant spray in a double-blind design, then rated by a panel of women; the abstract gives no sample size and two of the six authors were Unilever employees.
Frederick, Haselton (2007). Why is muscularity sexy? Tests of the fitness indicator hypothesis. Personality and Social Psychology Bulletin, 33(8), 1167-1183.
Six studies built on ratings and self-reports (the abstract gives no participant totals); it found moderately muscular men rated most attractive, which conflicts with the no-ceiling result in the 2017 photo study.
Centers for Disease Control and Prevention (2023). Adult Activity: An Overview. CDC, Physical Activity Basics (public guidance page dated December 20, 2023, accessed November 2026).
Public-health guidance for adults in general, not an attraction study; it sets broad activity targets and does not replace advice from a doctor.
Centers for Disease Control and Prevention (2024). About Sleep. CDC.
Public-health guidance on recommended sleep by age group, not an attraction study; a general target that does not replace advice from a doctor.
Carney, Cuddy, Yap (2010). Power posing: Brief nonverbal displays affect neuroendocrine levels and risk tolerance. Psychological Science, 21(10), 1363-1368.
42 participants who held two one-minute high-power or low-power poses; a small sample whose hormone and risk findings were not reproduced in a larger study.
Körner, Röseler, Schütz, Bushman (2022). Dominance and prestige: Meta-analytic review of experimentally induced body position effects on behavioral, self-report, and physiological dependent variables. Psychological Bulletin, 148(1-2), 67-85.
Preregistered meta-analysis of 128 body-position experiments with almost 10,000 participants (figures and findings taken from the University of Bamberg's summary, because the journal's abstract page would not open for us); most studies lacked a neutral control group and came from Western samples.
Elkjær, Mikkelsen, Michalak, Mennin, O'Toole (2022). Expansive and Contractive Postures and Movement: A Systematic Review and Meta-Analysis of the Effect of Motor Displays on Affective and Behavioral Responses. Perspectives on Psychological Science, 17(1), 276-304.
Systematic review and meta-analysis of 73 studies on expansive and contractive postures; few studies used neutral control groups, and the outcomes are short-term mood and behavior.
Eastwick, Finkel (2008). Sex differences in mate preferences revisited: Do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology, 94(2), 245-264.
163 undergraduates (81 women) at one US university who each had 9 to 13 four-minute speed dates; a single student sample of people willing to try speed dating.