Methodology & evidence
What the score actually means, why potential is capped, how we measure change over time, and where the cited studies come from. These are the rules the rating follows.
Why is the score an estimate, not a measurement?
The 0–100 score is an AI estimate from a single 2D photograph. It is not a measurement and not objective truth; it is a structured opinion against a fixed rubric of five areas (harmony, eye area, jaw and chin, skin, hair) and 30-plus traits inside them. That is why we never state numeric values for structural traits such as a gonial angle in degrees or a philtrum in millimetres: a phone photo cannot support that number, and pretending otherwise is false precision. Every trait comes back as a directional band instead — more or less, positive or negative, sharper or softer — with a short note on what drives the read. The same face photographed at a different angle, distance or light can land a few points apart, and the report says so rather than hiding it. What the number is good for is a starting point and a trend, not a verdict.
Why is the calibration so strict?
An average, healthy, groomed adult should land around 50–60 on our scale, not 75 and above. A rating that tells everyone they are above average is worthless, and the audience for these apps is the first to notice; most raters drift upward because a flattering result gets shared and an honest one does not. Scores above 80 are reserved for genuinely striking structure, and the two top tiers, Elite and Apex, are rare by design. The eight tiers are fixed bands on this scale, not a ranking against other users: there is no leaderboard and no comparison to anyone else who has scanned. The PSL equivalent shown next to the score is a fixed mapping onto the community's 1–8 ladder (50 sits at 4, 62 at 5, 75 at 6), offered as a translation for people who already think in that scale, never as the primary number.
Why does potential have a hard ceiling?
Potential is what is reachable without surgery, and the server caps it per area according to what genuinely moves: harmony +6 points, eye area +10, jaw and chin +14, skin +25, hair +20. The model cannot exceed that ceiling even if it wants to; the cap is enforced in code after the model answers, not requested in the prompt. Bone geometry does not change, which is why harmony gets the least and skin gets the most: softmaxxing works on skin, hair, composition, grooming and sleep, and those are the areas where published evidence exists. The gap between your score and your potential is therefore a bounded, honest estimate of a few months of consistent work, not a promise. If every area hit its cap at once the total would move by roughly a tier's width, and almost nobody hits every cap.
What is structural and what is workable?
Every weak point is tagged workable or structural. Structural means bone or fixed anatomy: the gonial angle, ramus length, maxilla position, orbital shape, facial thirds and fifths, canthal tilt. For those the plan will never promise a fix; it may only work with how the trait reads through body composition, posture, hair framing, brows and grooming. Workable means the tissue itself changes without a procedure: skin quality, under-eye appearance, hair density and cut, facial fat, and the expression and framing the camera sees. Selling a routine that claims to reshape bone is exactly what turns a rating app into a scam, so anything in that category, from tongue posture to devices, is labelled as having no studies or refused outright.
How does photo quality change what we claim?
Every photo is graded for lighting, angle and sharpness before it is scored. If a photo is not comparable to the last one, we withhold the score comparison rather than report a change that the camera caused. Differences within ±2 points are shown neutrally as measurement noise. The most common false positive is angle: a chin tilted down artificially sharpens the jaw and shifts the entire lower-third score, and a phone held at arm's length distorts the midface and the width-to-height ratio in ways that have nothing to do with the person. Overhead light deepens under-eye shadows; a wide smile changes the philtrum and the eye area. The report tells you what the capture grade was, so a low score on a bad photo reads as a bad photo, not as a bad face.
How is progress compared between scans?
On a re-scan the model sees both photos side by side and may only report visible differences. “Unchanged” and “regressed” are first-class answers; the prompt explicitly forbids inventing progress, because a rater that always finds improvement is selling reassurance, not information. The model may not report a change in bone geometry between scans at all, because there is none: if two photos make the jaw or the eye area look structurally different, the cause is angle, distance or light, and the capture grade usually catches it. The useful cadence is a checkpoint every six to eight weeks under matched conditions, judged as a trend rather than as single readings; the tracking guide describes the setup. PRO keeps the series so the trend is visible across scans.
What do the four evidence levels mean?
Every plan step carries an evidence level: strong (multiple randomized trials or meta-analyses), moderate (a few solid studies), weak (a single small study), no studies (practice without formal research). The level is attached to the step itself, next to the citation, so you can see at a glance whether daily sunscreen and a retinoid (strong) sit in the same plan as brow shaping (no studies). Mewing, jaw exercisers and most grooming advice fall into the last category and are labelled that way; the label is not a judgement that they are useless, only a statement that nobody has measured them properly. Where the evidence is for a different population than you (a trial in men, a different age range), the step says so. The four levels are the reason the plan can be honest about what it does not know.
Where do the citations come from?
The AI model is not allowed to “remember” a study; it may cite exclusively from the curated library below. If no library entry supports a recommendation, the step is labelled as having no studies and carries no source. The server checks every citation the model returns against the library by id, so a fabricated reference cannot reach you; a hallucinated paper is the single most common failure of language models asked for sources, and this closes the door on it entirely. Each entry was resolved through PubMed's own lookup and accepted only where PubMed's title matched ours; the few that could not be matched unambiguously link to a title search so you can verify them yourself. The library is small on purpose, covering only the interventions the plan actually recommends, and it grows only when a paper earns its place.
What is the potential image, and what is it not?
The simulation shows a possible direction with consistent work over 6–12 months. It preserves identity, bone geometry and skin tone; it may not narrow a nose, change canthal tilt or redraw a jaw. It changes only what genuinely changes: skin, hair, grooming and moderately facial fat, the same areas the potential caps allow. It is not a prediction or a promise, and it is generated under the same rules as the score, so an image that contradicted the caps would be a bug, not a feature. Treat it as a sketch of the direction the plan points in, and treat any before-and-after you see elsewhere, including ours on the landing page, as an illustration with the same limits: same person, same bone, only workable things moved.
What will BecomeTen never do?
We do not compare you to other users and we run no leaderboard. We do not adjust the score for ethnicity or gender identity; the capture is graded, the person is not adjusted. We do not diagnose, and nothing here replaces a doctor: where a step needs a prescription or a clinician, the plan says so and stops. We never recommend surgery, and never anything that damages tissue; bone smashing and similar forum methods are explicitly forbidden in the prompt and refused if asked. We do not use the demeaning terminology common in this community; the tiers have neutral names on purpose. We do not keep a free scan's photo, and no photo is ever used to train a model. The rest of the looksmaxxing guide and the glossary follow the same rules.
Evidence library
The complete list of publications the plan is allowed to cite. Every entry is a real, hand-picked paper — links search the title on PubMed so you can verify it yourself.
SPF & photoaging
- Sunscreen and prevention of skin aging: a randomized trial(Opens in a new window) — Annals of Internal Medicine (2013)
- Reduced melanoma after regular sunscreen use: randomized trial follow-up(Opens in a new window) — Journal of Clinical Oncology (2011)
- The efficacy and safety of sunscreen use for the prevention of skin cancer(Opens in a new window) — CMAJ (2020)
Retinoids
- Retinoids in the treatment of skin aging: an overview of clinical efficacy and safety(Opens in a new window) — Clinical Interventions in Aging (2006)
- Improvement of naturally aged skin with vitamin A (retinol)(Opens in a new window) — Archives of Dermatology (2007)
- Restoration of collagen formation in photodamaged human skin by tretinoin (retinoic acid)(Opens in a new window) — New England Journal of Medicine (1993)
Vitamin C
- Topical ascorbic acid on photoaged skin. Clinical, topographical and ultrastructural evaluation(Opens in a new window) — Experimental Dermatology (2003)
- Double-blind, half-face study comparing topical vitamin C and vehicle for rejuvenation of photodamage(Opens in a new window) — Dermatologic Surgery (2002)
Niacinamide
- Niacinamide: A B vitamin that improves aging facial skin appearance(Opens in a new window) — Dermatologic Surgery (2005)
- Nicotinamide and the skin(Opens in a new window) — Australasian Journal of Dermatology (2014)
Exfoliation (AHA/BHA)
- Effects of alpha-hydroxy acids on photoaged skin: a pilot clinical, histologic, and ultrastructural study(Opens in a new window) — Journal of the American Academy of Dermatology (1996)
- Salicylic acid as a peeling agent: a comprehensive review(Opens in a new window) — Clinical, Cosmetic and Investigational Dermatology (2015)
Moisturizer & barrier
- The role of moisturizers in addressing various kinds of dermatitis: a review(Opens in a new window) — Clinical Medicine & Research (2017)
- Moisturizers: The Slippery Road(Opens in a new window) — Indian Journal of Dermatology (2016)
Acne
- Guidelines of care for the management of acne vulgaris(Opens in a new window) — Journal of the American Academy of Dermatology (2016)
- A low-glycemic-load diet improves symptoms in acne vulgaris patients: a randomized controlled trial(Opens in a new window) — American Journal of Clinical Nutrition (2007)
- Azelaic acid. A review of its pharmacological properties and therapeutic efficacy in acne and hyperpigmentary skin disorders(Opens in a new window) — Drugs (1991)
Pigmentation
- Melasma: a comprehensive update: part I(Opens in a new window) — Journal of the American Academy of Dermatology (2011)
- Randomized, placebo-controlled, double-blind study of oral tranexamic acid in the treatment of moderate-to-severe melasma(Opens in a new window) — Journal of the American Academy of Dermatology (2018)
Eye area
- Infraorbital dark circles: definition, causes, and treatment options(Opens in a new window) — Dermatologic Surgery (2009)
Procedures
- Microneedling: A Review and Practical Guide(Opens in a new window) — Dermatologic Surgery (2018)
- Acne scarring treatment using skin needling(Opens in a new window) — Clinical and Experimental Dermatology (2009)
- Fractional photothermolysis: a new concept for cutaneous remodeling using microscopic patterns of thermal injury(Opens in a new window) — Lasers in Surgery and Medicine (2004)
- A multicenter, double-blind, randomized, placebo-controlled study of the efficacy and safety of botulinum toxin type A in the treatment of glabellar lines(Opens in a new window) — Journal of the American Academy of Dermatology (2002)
- Hyaluronic acid fillers: a comprehensive review(Opens in a new window) — Facial Plastic Surgery (2009)
- Chemical peels in aesthetic dermatology: an update 2009(Opens in a new window) — Journal of the European Academy of Dermatology and Venereology (2010)
- Assessment of color changes and adverse effects of over-the-counter bleaching protocols: a systematic review and network meta-analysis(Opens in a new window) — Clinical Oral Investigations (2024)
Hair
- A randomized clinical trial of 5% topical minoxidil versus 2% topical minoxidil and placebo in the treatment of androgenetic alopecia in men(Opens in a new window) — Journal of the American Academy of Dermatology (2002)
- Finasteride in the treatment of men with androgenetic alopecia(Opens in a new window) — Journal of the American Academy of Dermatology (1998)
- Interventions for female pattern hair loss(Opens in a new window) — Cochrane Database of Systematic Reviews (2016)
- Ketoconazole shampoo: effect of long-term use in androgenic alopecia(Opens in a new window) — Dermatology (1998)
- Low-level laser (light) therapy (LLLT) for treatment of hair loss(Opens in a new window) — Lasers in Surgery and Medicine (2014)
- A randomized evaluator blinded study of effect of microneedling in androgenetic alopecia: a pilot study(Opens in a new window) — International Journal of Trichology (2013)
- The effect of platelet-rich plasma in hair regrowth: a randomized placebo-controlled trial(Opens in a new window) — Stem Cells Translational Medicine (2015)
- The diagnosis and treatment of iron deficiency and its potential relationship to hair loss(Opens in a new window) — Journal of the American Academy of Dermatology (2006)
- The Role of Vitamins and Minerals in Hair Loss: A Review(Opens in a new window) — Dermatology and Therapy (2019)
- Shampoo and Conditioners: What a Dermatologist Should Know?(Opens in a new window) — Indian Journal of Dermatology (2015)
- The biology of hair follicles(Opens in a new window) — New England Journal of Medicine (1999)
- Telogen effluvium. New concepts and review(Opens in a new window) — Archives of Dermatology (1993)
Lifestyle
- Does poor sleep quality affect skin ageing?(Opens in a new window) — Clinical and Experimental Dermatology (2015)
- Cues of fatigue: effects of sleep deprivation on facial appearance(Opens in a new window) — Sleep (2013)
- Beauty sleep: experimental study on the perceived health and attractiveness of sleep deprived people(Opens in a new window) — BMJ (2010)
- Tobacco smoke causes premature skin aging(Opens in a new window) — Journal of Dermatological Science (2007)
- Cigarette smoking: risk factor for premature facial wrinkling(Opens in a new window) — Annals of Internal Medicine (1991)
- Impact of Smoking and Alcohol Use on Facial Aging in Women: Results of a Large Multinational, Multiracial, Cross-sectional Survey(Opens in a new window) — Journal of Clinical and Aesthetic Dermatology (2019)
- Brain-skin connection: stress, inflammation and skin aging(Opens in a new window) — Inflammation & Allergy Drug Targets (2014)
- Dietary water affects human skin hydration and biomechanics(Opens in a new window) — Clinical, Cosmetic and Investigational Dermatology (2015)
- Discovering the link between nutrition and skin aging(Opens in a new window) — Dermato-Endocrinology (2012)
- Cosmetic and Therapeutic Applications of Fish Oil's Fatty Acids on the Skin(Opens in a new window) — Marine Drugs (2018)
Facial perception
- The evolutionary psychology of facial beauty(Opens in a new window) — Annual Review of Psychology (2006)
- Facial attractiveness: evolutionary based research(Opens in a new window) — Philosophical Transactions of the Royal Society B (2011)
- Human (Homo sapiens) facial attractiveness and sexual selection: the role of symmetry and averageness(Opens in a new window) — Journal of Comparative Psychology (1994)
- Facial adiposity: A cue to health?(Opens in a new window) — Perception (2009)
- Deciphering faces: quantifiable visual cues to weight(Opens in a new window) — Perception (2010)
- Visible skin color distribution plays a role in the perception of age, attractiveness, and health in female faces(Opens in a new window) — Evolution and Human Behavior (2006)
- Color homogeneity and visual perception of age, health, and attractiveness of female facial skin(Opens in a new window) — Journal of the American Academy of Dermatology (2007)
- Influence of facial skin attributes on the perceived age of Caucasian women(Opens in a new window) — Journal of the European Academy of Dermatology and Venereology (2008)