A complete Face Rate report, start to finish

The Score is a single, clearly-labelled estimate derived from six on-device geometry signals. It is not a grade, not a forecast, and not a measurement of you.

Read the score as a band, not a point: 90% of photos like this land between 63 and 79.

The framing this report is built on

Your face is a weak, largely static signal; your dating profile is a set of editable decisions. We treat the face as roughly 30% of the picture and the controllable 70% — photos, styling, bio, platform choice and behaviour — as the work. That is not a promise that changing those things produces a particular number of matches. It is a statement about where your effort has leverage.

This chapter is your baseline. CH-27 to CH-33 are the editable 70%, ranked and scheduled.

Generated Nov 3, 2026 · Heuristic pps-v2.0.0 · Disclaimers dis-v1.0 · Order SAMPLE-0000. Estimated with a lightweight on-device geometry heuristic — not a 478-landmark clinical measurement.

Four findings, each with the number behind it. A short section is a clean bill, not a gap.

1. Symmetry at 81/100 — the strongest of your six geometry signals, sitting 80-100.
2. Golden Proportion at 74/100 — a second signal inside the broad reference range.
3. Canthal tilt - right eye measured at 7.60, inside its published reference band of -8–10.
4. Photo presentation 63/100 — fine, one clear fix; the seven properties and your three fixes are printed in full.

Your harmony profile

Each bar is a core dimension of facial harmony, scored 0–100. The dents are your highest-leverage areas. The median column is the published reference midpoint, not a measured cohort — we do not invent a sample size.

Six dimensions, exact scores, and what each one means. The median is a descriptive anchor, never a target.

Everything below is a photography or styling decision. Nothing here costs money, requires a purchase, or changes your face. The day-to-day plan comes before the evidence on purpose: the first week is about what to do, not how the number was calculated.

First 72 hours — start tonight

1. Tonight and tomorrow morning, apply a cold compress or chilled spoon under the eyes and along the jawline for 60 seconds each side.
   Temporary de-puffing only. It clears transient water retention for a few hours and changes nothing structural.
2. Do a 30-second posture reset every time you sit down: ears over shoulders, chin level, shoulders dropped.
   A forward head compresses the visible neck-jaw line in photos. This is a presentation fix, not a structural one.
3. Comb brows into place morning and night with a clean spoolie — no product needed.
   Stray hairs are the only thing competing visually with a measurement that is already yours.
4. Blot lips with a damp warm cloth for 10 seconds before balm, once daily.
   Removing flaky dry skin lets the lip border catch light evenly. A texture read, not a shape change.
5. Re-shoot one photo fixing background quiet only: Move two steps so the background is plain or softly out of focus. You should be the sharpest thing in the photo.
   One variable at a time is the only way to tell which change actually moved the numbers.

Your four-week program

Where you sit: Upper third of the reference band. Your overall harmony sits at the 69th percentile of the reference band, under this photo's conditions.

That is the plain answer to "what level am I?" — and it is a statement about where a number sits in a published band, not about where a person sits among other people. No ranking, no league table, and no claim that you are better than a percentage of other people.

Measured-scan percentiles are omitted here: the anonymous distribution has fewer than the published minimum, and printing a percentile from a handful of scans would be an invented statistic wearing a decimal point.

How this is calculated

There are not yet 50 anonymous scans in the distribution, so this report shows placement inside the published reference band only — no invented cohort percentile, no made-up sample size. Once enough scans exist, a real, labelled distribution appears here automatically. The reference band is drawn from anthropometric literature and rounded outward; it is a range, not an ideal.

Band position is pure arithmetic: the value's place between the lower and upper bound of the published reference range. When enough anonymous scans exist, the same cell also prints the real distribution of scans measured on this site, with its sample size and self-selection bias stated. Buckets under five are hidden. The histogram holds only integer counters — no scanId, no email, no image, no per-person row. Deleting a photo after two hours never deletes a counter, and a counter can never be turned back into a photo.

In short — placement, not grades: each reading shows where your geometry sits in a measured range, and 100 is not the goal. Photo conditions can affect every comparison.

Every region read from your photo, with the band each measurement sits in and an explicit "not scored" state where a front frame cannot support the claim. Placement inside a shaded reference band — not a bare grade out of 100.

Overall proportions

Eye area

Nose, mouth and jaw

Explicitly not scored

• Nose projection — Not scored, needs a side photo.
• Bridge profile — Not scored, needs a side photo.
• True gonial angle — Not scored, needs a side photo.
• Chin projection — Not scored, needs a side photo.
• Skin texture, pores and scarring — depends on light and resolution far more than on skin.

Reference bands are photo-calculator ranges drawn from anthropometric literature and rounded outward, not ideals. A value outside a band is a statement about the reference sample, never about a person.

The Photo Presentation Score (PPS) is the part of this report that is measured rather than estimated. Seven properties are computed from your pixels on your device: light evenness, sharpness, framing, background calm, contrast, headroom and colour cast. Nothing is uploaded. Full weights and thresholds are on the methodology page.

Your score is 63/100 — fine, one clear fix.

Your three fixes, in order

1. 1. Background quiet (44/100). Move two steps so the background is plain or softly out of focus. You should be the sharpest thing in the photo.
2. 2. Light evenness (52/100). Face a window at eye level. Overhead light and one-sided sun are the two most common reasons a photo reads flat.
3. 3. Colour cast (61/100). Mixed lighting (window plus tungsten) tints skin. Shoot in one light source or correct white balance afterwards.

Because these numbers come from pixels, they move when you change the photo. Re-shoot, run the free scan again, and compare — that loop is the whole point. Nothing about your face is being re-scored here.

Important: a low PPS is not a statement about your face. It is a statement about one photograph — and a photograph is the most fixable thing in this entire report.

Text alone should not be the difference between a free scan and a paid report. Your report includes 3 AI styling concepts rendered on your own photo — included in the price, with no credits spent.

This is a public sample, so the figures above are illustrative styling concepts generated on a sample portrait. In a delivered report this panel asks for your explicit consent before generating the concepts on your own photo — nothing is sent without it.

Included concepts do not use or reduce your AI credit balance. Extra looks and re-generations are available separately with prepaid credits. Generated concepts are stored with this report and can be deleted at any time from Delete my data.

Six cuts matched to a oval read from your ratios. The top card is a hypothesis, not a verdict. Pick by the maintenance line, not the picture: the cut you will actually style beats the cut that photographs best.

Derived from your measured height-to-width (1.21), jaw-to-cheekbone (0.75) and middle-band share (45%). A photo-based shape estimate for styling, not a classification of your face.

Every card carries the decision data a yes/no needs, including the honest tradeoff. Want different cuts? The Hairstyle Try-On generates six more on your own photo using prepaid credits.

Color season: Light Spring

Season is shorthand for three measured attributes of your colouring — undertone (neutral), value (light) and contrast (medium) — and it decides which colours flatter by default. This is a photo-based estimate from one frame; Light Summer is the closest alternative, and the light in the room can move the verdict.

Wear these near your face

These fight your coloring

A photo-based estimate from undertone, value and contrast in one frame. It is not a draping session, not a diagnosis, and the light in the room can move the verdict.

Each number below is an estimate from a lightweight on-device image statistic — not a 478-landmark clinical measurement. The bars show placement inside a broad reference band, not a grade out of 100.

Not scored: nose projection, bridge profile, true gonial angle and chin depth. A single front-facing photo cannot support those claims, so we say "not scored" instead of guessing. A single front-facing photo cannot support side-profile claims. Where that is true, we say "not scored" instead of guessing.

Scores are placement inside a broad reference band, not grades. 100 is not the goal. This tool measures the photo, not your face directly. Lighting, angle, focal length and expression all move the reading. Aesthetic reference ranges in the literature carry documented bias. We treat them as rough bands, not ideals.

Symmetry carries the heaviest weight in the published formula (22%), which is exactly why it deserves its own chapter: the number is more fragile than it looks.

Your symmetry estimate is 81/100 — 80-100. It is computed by reflecting every landmark across your facial axis and measuring how far each point lands from its mirror twin.

Three things that move this number without moving your face

• Head tilt. Two degrees of roll shifts the fitted axis and can move the estimate by five points or more.
• Side lighting. A window on one side darkens one cheek; the landmark detector then places the shadowed contour slightly differently.
• Expression. A half-smile pulls one mouth corner up and reads as asymmetry that is not there at rest.

What we will not tell you to do about it

Nothing. Measured asymmetry is bone and soft tissue, and the honest response to a low symmetry number is a straighter photo, not a procedure. This report never recommends surgery, fillers or injectables, and it does not rank ethnic features against a reference ideal. Aesthetic reference ranges in the literature carry documented bias. We treat them as rough bands, not ideals.

Proportions are the most over-claimed measurement in this category, so here is the whole of what we do: three horizontal bands from hairline to brow to nose base to chin, and five vertical bands across the face measured in eye widths.

Your three bands

What the classic "rule of thirds" actually is

It is a drawing convention from portraiture, not a finding about human faces. Real populations vary widely, and the top band is dominated by where your hairline happens to sit rather than by bone. A face with a high hairline will read as a short upper third no matter what the bone underneath is doing.

What is genuinely worth knowing

• The middle band is the stable one. Both ends are bone (brow and nose base), so it survives a change of camera better than the other two.
• The upper band is a hairline measurement. Haircut and forehead coverage move it; nothing else does without intervention we do not recommend.
• The lower band moves with the lens. Camera below eye level lengthens the chin; above eye level shortens it. This is perspective, not anatomy.

Fifths are reported as a consistency check rather than a target: five bands of roughly one eye width each. Frontal camera position matters more than any deviation from the convention.

The eye area carries 14% of the Score and most of the questions. Four measurements are printed below, each between two named landmarks.

Measured, left and right separately

Reading the canthal tilt honestly

Canthal tilt is the angle between the inner and outer corner of the eye. A positive value means the outer corner sits higher. The band is wide (-5° to +8°) because ordinary faces span it — and because the measurement moves with head roll and with the eyelid position in a single frame.

A left/right difference of a few degrees is normal and extremely common. It is not a defect, and this report does not treat it as one.

What is actually movable here

• Brow shape and position — grooming changes the brow-to-eye distance and the perceived width of the upper face. It is the only high-leverage, reversible lever in this chapter.
• Sleep and hydration — lid puffiness changes the measured aperture on the day. Nothing permanent, but it changes what the photo shows.
• Camera height — shooting slightly above eye level narrows the aperture and flattens the tilt. This is a photography fact, not a face fact.

What we do not recommend: surgery, filler, injectables or anything else aimed at changing the shape of your eyes. The report is a measurement, not a referral.

Front photo: width and symmetry only. A front frame supports width and symmetry readings only. Projection and bridge profile need a side view — this report does not score what it cannot see.

Nose and midface

• Nose projection — Not scored, needs a side photo.
• Bridge profile — Not scored, needs a side photo.
• True gonial angle — Not scored, needs a side photo.
• Chin projection — Not scored, needs a side photo.
• Skin texture, pores and scarring — depends on light and resolution far more than on skin.

In short: the central face has one honest lever, and it is photographic — even light. Nothing here recommends a procedure.

Lip balance is about the ratio between lip height and mouth width and how the mouth sits inside the lower face — not absolute size.

Lips and smile

Mouth width is measured at rest. A smile widens it by several points, which is why the relaxed value is the one printed. Hydration and a tidy lip border change how the border catches light in a photo; nothing here changes shape or fullness.

Cosmetic suggestions for information and entertainment — not medical, dermatological or professional advice.

The lower third is where most of the internet's jaw advice makes its money, and where a front-facing photo is least able to support the claims being sold. Here is what we can measure and what we refuse to.

Measured lower-face proportions

Three limits, stated plainly

• Gonial angle needs a profile. The angle at the back of the jaw is invisible from the front. We report jaw width as a proxy and label it as one.
• Chin projection needs a profile. Whether the chin sits forward of the lips cannot be read front-on. Not scored, anywhere in this report.
• Jaw definition is partly body composition. Body fat affects how the jaw reads; that is a health-and-habit question, not a bone question, and this report does not prescribe either.

What genuinely moves the lower third in a photo

• Camera height. A phone held at chest level foreshortens the chin and widens the jaw. At eye level, the proportions change without your face changing.
• Jaw posture. A relaxed jaw with the tongue resting on the palate reads differently from a clenched one. It is free, reversible, and it is the only jaw advice in this report.
• Beard line. If you wear one, the line sets where the jaw appears to end. That is a grooming decision with a visible effect and no risk.

Mouth width is measured at rest. A smile widens it by several points, which is why we print the relaxed value and say so.

Skin is the one part of this report where the honest answer is mostly "we measured a colour, not a condition". The tone read below comes from the average of the cheek and forehead regions, converted to an Individual Typology Angle (ITA) — a colour-space coordinate, not a health or beauty grade.

What this is not

• Not a dermatology read. No diagnosis, no severity score, no recommendation to treat anything.
• Not a skin-age estimate. The wide perceived-age band on the first screen is an explicitly-labelled entertainment read from photographic cues in one frame — not your chronological age, not a dermatological estimate, and not a diagnosis.
• Not a reason to lighten anything. This report never recommends skin lightening, bleaching or tone modification. Aesthetic reference ranges in the literature carry documented bias. We treat them as rough bands, not ideals.

What is reasonable, and boring

Broad-spectrum SPF every morning, sleep, water, and a cleanser that does not sting. That is the whole of the evidence-backed advice, and it is the same advice regardless of what the ITA number says. If a skin concern is bothering you, that is a conversation with a clinician, not with a face-measurement report.

Observed in this photo, not clinical

Three photographic observations of this frame in this light — useful for tracking change, never a diagnosis. They are computed from the same pixel properties as your Photo Presentation Score, so a different light moves them without your skin changing.

Perceived-age band, at a glance: 27-37, typical guess 32. A wide, entertainment-only band from photographic cues in this one frame — light, texture and shadow. It is not your chronological age, not a skin-age diagnosis, not a health read, and not a prediction. Re-shoot in different light and it moves. This is a wide entertainment read from photographic cues, not a skin-age estimate, not a diagnosis, and not your chronological age.

Ordered by visible payoff for the effort. Each move names the measurement it targets, its upkeep, and how confident the analysis is — none promises a score.

In short — everything else in the diagnostic half is evidence; these three moves are its output. The dating-domain action plan is a separate chapter in the Dating Report.

The day-to-day side of the report. General, boring, evidence-backed, and identical regardless of what any number in this report says.

Daily routine, in order

1. AM: a gentle cleanser that does not sting
2. AM: broad-spectrum SPF, every morning regardless of the ITA reading
3. PM: remove the day with a second cleanse
4. PM: a plain moisturiser that supports the skin barrier
5. Retest in matched light three to four weeks after any change — progress judged in matched photos is real, progress judged in the mirror is mood

Key ingredients

• Vitamin C for tone evenness
• Niacinamide for texture
• Ceramides for the barrier

Training and habits

• Gentle neck side-stretches to release tension
• A relaxed jaw rather than a clenched one
• A short walk outside — the least cosmetic item here, and the one with the best evidence
• Consistent high-factor SPF
• Regular sleep
• Enough water
• Phone at eye level instead of a sustained downward chin tuck

Grooming

• Clear brow gel to keep the frame tidy
• A plain lip balm with a satin finish
• Flyaway hairs tidied around the hairline

Frames for your face shape

For a oval read: maintain the natural balance — almost any frame width works.

The safest rule is a frame as wide as the broadest part of your face. Cosmetic suggestions for information and entertainment, not medical or professional advice. For any skin or health concern, see a qualified professional.

Your face measures inside reference bands on most of the structural ratios tracked. The strongest signal in this scan is symmetry at 81/100. That is a description of one photograph, not a verdict.

Distinctive features

• A oval read with symmetry at 81/100 — the strongest signal in this scan.
• Golden Proportion at 74/100, giving the region a consistent frame with no styling work required.
• A measured 14 ratios inside published, deliberately wide reference bands — the "improvement areas" in this report are refinements, not corrections.
• An explicit "not scored" list for nose projection, bridge profile, true gonial angle and chin depth, because one front photo cannot support them.

Reference silhouettes

Your proportions sit near the balanced-oval family in the reference atlas. We deliberately do not name real people: a celebrity face is not a reference population, and a lookalike list is not a measurement.

Confidence notes

• Regions that need nothing get told nothing — a short section is a clean bill, not a gap.
• Scores here are placement, not grades: 100 is not the goal, and strong faces sit all over a wide band.
• Nothing in this report predicts matches, dates, replies or your worth.
• Reference silhouettes: your proportions sit near the balanced-oval family in the reference atlas. We deliberately do not name real people — a celebrity face is not a reference population.

Most face-rating products stop at a number and leave you to work out what it implies. This chapter does the opposite: it sorts every input in this report into three buckets, so you can see exactly where effort pays and where it does not.

Where we draw the line. This report never recommends surgery, fillers, injectables, skin lightening, or any procedure. It also does not rank ethnic features against a reference ideal. Aesthetic reference ranges in the literature carry documented bias. We treat them as rough bands, not ideals. There is no leaderboard here. We don't rank you against other people.

Every number in this report comes from one of three layers, and the layer is always stated. Mixing them silently is how face-rating products end up sounding more certain than the evidence allows.

The measurement pipeline, in order

1. Decode locally. Your photo is drawn into a canvas in this browser. Re-drawing strips EXIF, including GPS. For the free scan your photo is processed on your device and is never uploaded, stored, sold, or used to train AI. The optional AI styling concepts included in a paid report are the only exception, and need a separate, explicit consent on the report page.
2. Map 478 landmarks. MediaPipe's face landmarker runs on the device (WASM, no network call) and returns x/y coordinates plus a coarse pose estimate.
3. Measure, then score. Distances and angles are computed between fixed landmark pairs; the six signals and the Score are computed from those with the published weights.
4. Store numbers only. The scan record holds values, never pixels. Only numeric scores are kept for 2 hours, then deleted.

Engine pps-v2.0.0 · landmark model MediaPipe Face Landmarker · measurement code is deterministic, so the same photo and the same version produce the same numbers.

This is the most important chapter in the report. Read it before you act on any number here.

1. It cannot predict matches, dates, replies, or your worth. The Score is a weak geometric signal. It does not predict matches, dates, replies, or your worth.
2. It does not measure you as a person. Warmth, humour, reliability, curiosity and kindness do not appear in any of the six signals.
3. It is not a scientific measurement. Estimated with a lightweight on-device geometry heuristic — not a 478-landmark clinical measurement. The weights are a disclosed heuristic calibration, not a universal formula.
4. One photo cannot show everything. A single front-facing photo cannot support side-profile claims. Where that is true, we say "not scored" instead of guessing.
5. The photo moves the number. This tool measures the photo, not your face directly. Lighting, angle, focal length and expression all move the reading. A different lens, light or expression would produce a different estimate.
6. Reference ranges carry bias. Aesthetic reference ranges in the literature carry documented bias. We treat them as rough bands, not ideals. We use them as rough bands, not ideals.
7. There is no leaderboard here. There is no leaderboard here. We don't rank you against other people. We never rank you against an ethnicity-based ideal or against another person's face. If a low score is making you feel bad, close the tab. This is a measurement, not feedback that should affect how you see yourself.

This report is cosmetic and non-surgical guidance for information and entertainment. Here is what it will never do, in any section:

• Never Recommend surgery, fillers, injectables, skin lightening or any procedure — anywhere in this report.
• Never Rank you against an ethnicity-based ideal, or against another person's face.
• Never Show a "potential you" morph, or promise that a score will go up.
• Never Diagnose a skin condition, estimate your chronological age, or predict dating outcomes.

Cosmetic suggestions for information and entertainment — not medical, dermatological, psychological or professional advice. For any skin or health concern, see a qualified professional.

The published formula

Score = 0.22·D_sym + 0.20·D_gold + 0.18·D_thirds + 0.14·D_fifths + 0.14·D_eyes + 0.12·D_jaw

Each D is a 0-100 estimate from an on-device image statistic (mirror difference, band luminance/edge distribution, region projection). Estimated with a lightweight on-device geometry heuristic — not a 478-landmark clinical measurement. The weights are our disclosed heuristic calibration, not a scientifically universal formula. Full method: How the Score is calculated.

Sources behind the framing (background, not validation of the score)

Background reading: Langlois & Roggman (1990); aggregated swipe-data analyses. Directional context, not a validation of this tool.

• Langlois & Roggman (1990), Journal of Personality and Social Psychology — averaged faces and perceived attractiveness.
• Rhodes (2006), Annual Review of Psychology — symmetry, averageness and their limits.
• Little, Jones & DeBruine (2011/2014) — facial cues and individual variation.
• Farkas (1994) — anthropometric facial norms; we cite it to explain why reference ranges are population-specific and biased, not as an ideal.
• Lugaresi et al. (2019), MediaPipe — the landmark model behind the measured ratios in CH-06.
• Publicly reported aggregated swipe data (e.g. SwipeStats) — we cite the general observation that photo quality affects swipe decisions, without inventing precise percentages.

Your rights

• 7-day refund: 7-day refund, no questions asked — just reply to your delivery email. We do not withdraw the report after a refund.
• Your data: free scans leave no image behind; only numeric scores are stored briefly. Paid reports are private and kept up to 12 months unless you delete them.
• Your photo stays in your own browser (IndexedDB) for the free scan. Delete it from the report header at any time; clearing site data does the same thing. The included AI styling concepts are optional and are generated only after a separate, explicit agreement: the source photo is held for that request only and never stored, while the generated concepts are kept with this report and are deletable at any time.
• Anonymous percentiles use integer counters only — score buckets and dimension buckets. They hold no scanId, email, image or per-person row, and nothing below five scans is shown. Deleting your photo after two hours never deletes a counter, and a counter can never reconstruct a photo.
• Delete any time: Delete my data. Export: Email preferences / export.
• For adults 18+. Lost your report? Enter your email and we will resend the links. Find my report with the email you bought with.
• Formula: See exactly how the score is calculated.
• Privacy: Privacy Policy · Terms · Cookies.

Entertainment & self-insight only · Not medical advice · Score is a weak geometric signal and does not predict matches · facerate.co

Photos are the highest-leverage editable input, and they are the only part of this report measured directly from your pixels rather than estimated. This chapter is about the photograph, not your face.

Current photo type (self-reported): A group shot. Main friction: Matches but no replies.

Your photo, measured

Presentation score 63/100 — fine, one clear fix. Seven properties, measured on your device, with the published weights from the methodology page.

Your three fixes, in order

1. 1. Background quiet (44/100). Move two steps so the background is plain or softly out of focus. You should be the sharpest thing in the photo.
2. 2. Light evenness (52/100). Face a window at eye level. Overhead light and one-sided sun are the two most common reasons a photo reads flat.
3. 3. Colour cast (61/100). Mixed lighting (window plus tungsten) tints skin. Shoot in one light source or correct white balance afterwards.

Because these numbers come from pixels, they move when you change the photo. Re-shoot, run the free scan again, and compare — that loop is the whole point. Nothing about your face is being re-scored here.

Your three-photo set

1. Photo 1 — the clear solo. Recent, whole face, soft front light, plain background, chest-up crop. No sunglasses, no heavy filter, no other people.
2. Photo 2 — the context shot. You doing something real, at a readable distance, still recognisably you.
3. Photo 3 — the texture shot. A different setting or angle that adds information rather than repeating photo 1.

The ten-minute checklist

• Light: face a window; avoid overhead light and direct midday sun.
• Background: tidy and low-contrast; you should be the sharpest thing in frame.
• Angle: camera at eye level, lens at arm's length or longer to avoid perspective distortion.
• Expression: relaxed and closed-mouth or a small genuine smile; take ten and pick the one that looks like you on a good ordinary day.
• Recency: within the last six months. If someone would not recognise you from it, replace it.

Come back with a new photo any time — the free Dating Photo Analyzer reads lighting, background, angle and expression without rating your face.

One repeatable look beats a wardrobe of experiments when you only have three photos and limited time.

The baseline look

• One well-fitting top in a solid colour that contrasts with your background and sits near your face.
• One visual anchor — a jacket, a colour, a frame, a texture — that appears in at least two photos so the set feels coherent.
• Grooming kept simple and consistent; nothing that needs a special occasion to make sense.

Budget bands

We may earn a commission on partner links at no extra cost to you; see the Affiliate Disclosure. Comparisons are not paid. We do not recommend procedures, surgery, fillers or injectables anywhere in this report.

Three complete, copyable bios and five openers. Use them as starting points — the best line always references something real in the other person's profile.

Openers

1. Okay, that line about your weekend is a genuinely good answer. What got you into it?
2. I was going to play it cool, but your prompt deserves an actual question — how did that start?
3. Two options: we debate your take, or we skip the small talk and compare worst first dates. Your call.
4. That detail made me stop scrolling. What is the story behind it?
5. Small question first: is that still true this week, or has it escalated?

The coaching note: a contextual question beats a clever generic line. Ask about one detail, then make a concrete, low-pressure suggestion. If a conversation stalls, let it stall — volume is not the goal.

Stated goal: Long-term relationship. Current main app: Hinge. Time available: 1-3 hours.

A face score is the same number everywhere. A profile is not: the same photos and bio have different friction in different dating markets. This chapter ranks four markets for your profile, using the rule published below — not a vibe, and not a prediction of matches.

Fit = 0.34 goal alignment + 0.28 photo readiness + 0.20 effort fit + 0.18 tool fit. Fit is a positioning score, not a match forecast, and it does not change who you are.

What each market asks for

Affiliate disclosure: some links are partner links. We may earn a commission at no extra cost to you. Tinder, Bumble and Hinge don't run affiliate programs — our comparisons aren't paid. Tinder, Bumble and Hinge do not run affiliate programs, so our comparisons are not influenced by them. We only link partners we have actually joined — there are no placeholder partner links on this site.

Everything else in this report is evidence; these three moves are the output. Each one targets an editable input, with cost, effort and an honest time-to-first-signal.

In short — your highest-leverage area is Background quiet, which is entirely under your control. Scores are placement inside a broad reference band, not grades. 100 is not the goal.

A schedule, not a subscription. There is no auto-renewal, no streak penalty and nothing to cancel. Each week has one theme, one action and one thing you can point to as done.

Week 8 is the point: compare against your week-1 baseline, keep what worked, drop one thing that did not, and stop measuring your face. The controllable 70% is where the leverage was all along.

This report does not rank you into a "tier" of people, and it does not promise a match rate. Dating outcomes depend on timing, location, other people's preferences and chance — none of which a photo score can model.

Calibration

• Low activity for a few weeks is normal and is not evidence about your face.
• A quiet week is a reason to review photos and openers, not to retake a face test.
• Do not buy anything to chase a number. Nothing in CH-31 or CH-32 requires a purchase.

Red flags and breadcrumbing

• Pressure to move off-platform immediately, requests for money, or inconsistent identity details.
• Endless low-effort replies that never lead to a plan ("breadcrumbing") — it is fine to stop replying.
• Anyone who makes you feel bad about your appearance is giving you information about themselves, not about you.

Meeting safely

• First meeting: public place, your own transport, a friend who knows where you are.
• Keep personal details private until trust is established; you owe nobody an explanation for leaving.

If a low score is making you feel bad, close the tab. This is a measurement, not feedback that should affect how you see yourself.