How the color analysis quiz works, signal by signal

The three axes: undertone, depth, clarity

Every season system, whatever it calls itself, reduces to three measurements. Undertone (hue): is your coloring warm or cool underneath? Depth (value): does it run light or dark overall? Clarity (chroma): does it read soft and muted, or bright and clear? Each of the 12 seasons occupies a fixed position on those three axes, so a person's answers can be scored as a point in the same space and matched to the nearest season. That is the entire trick; everything else is deciding how much to trust each answer.

The 8 signals and their weights

Each answer casts a weighted vote on one or more axes. The weights are not equal, because the signals are not equally reliable:

• Vein color(undertone, heaviest weight): blue or purple veins vote cool, green or olive vote warm. A genuine can't-tell mix votes neutral at reduced weight.
• Gold versus silver (undertone, heaviest weight): which metal makes your skin look clearer in daylight. Ties vote neutral.
• Bright white versus cream (undertone, medium-high weight): the same test stylists run with drapes, using two things you already own.
• The black test (clarity plus depth): whether pure black looks sharp on you or drains you is one of the cleanest soft-versus-bright separators available without fabric drapes.
• Natural hair color in your 20s (depth, heaviest weight): the anchor for how light or dark your palette runs. If you have been gray for years, there is a dedicated answer for that; it deliberately withholds the undertone vote a remembered color would cast, which usually caps the result at moderate confidence. That cap is honesty, not a malfunction.
• Eye pattern (undertone and clarity): gold-flecked warm eyes, a clear ring around the iris, or soft blended flecks are each a different clarity signal.
• Skin-to-hair contrast (clarity, heaviest weight): high natural contrast points bright, low contrast points soft.
• Sun reaction (undertone, lightest weight on purpose): burn-versus-tan correlates with undertone but weakly, so it can nudge a result and never decide one. The quiz says so on the question itself.

Every question also offers "not sure." A not-sure answer casts no vote at all, which matters for confidence below.

Scoring: a weighted average, then the nearest season

The votes on each axis are combined as a weighted average, so one strong signal cannot be outvoted by three weak ones of the same total count. The resulting point is compared against all 12 seasons' positions, with each axis normalized so undertone's wider scale cannot dominate depth and clarity, and the nearest season wins. The runner-up matters as much as the winner: the gap between first and second place is the raw material for the confidence label.

How confidence is decided, and when the quiz refuses to be sure

A result is labeled high confidence only when two things are true at once: the winning season leads the runner-up by a clear margin, and enough questions were actually answered that the margin is built on real evidence. Not-sure answers reduce the evidence total, and below a minimum threshold the quiz caps itself at low confidence no matter how clean the margin looks.

The consequence, measured across every possible way to answer the quiz, is that only 10.5% of the 143,360 possible answer combinations reach the high-confidence bar, and 89.5% return a named alternate season with a daylight drape test chosen for that exact season pair (36 boundary tests are hand-written into the season data). The full breakdown, including the season-by-season distribution, is on the statistics page.

What the result contains

Your season, the confidence label, the alternate season when one exists, and the drape test to settle it. The palette behind the result is the 340-color dataset: each season's named colors with hex values, capsule neutrals, metals, and avoid-colors with the mechanism stated, all on the season pages. If you go on to the capsule wardrobe builder, the result carries over and colors every neutral and accent slot with your actual palette.

Frequently asked questions

Why doesn't the quiz ask for a photo?

Because the photo is the least reliable input available. Lighting, white balance, and camera processing shift an undertone reading before any analysis starts, and there is no way to tell from the photo alone whether that happened. The signals the quiz asks about instead, veins, metals, sun reaction, contrast, are checkable by you in daylight, which a pixel is not.

Which quiz questions matter most?

The vein check and the gold-versus-silver jewelry test carry the most weight on undertone, natural hair color carries the most on depth, and the skin-to-hair contrast check carries the most on clarity. Sun reaction is deliberately the lightest-weighted question, because burn-versus-tan is the weakest of the signals on its own.

What happens if I answer 'not sure' a lot?

The quiz withholds those votes instead of guessing, and if too little evidence accumulates it caps its own confidence at low regardless of how clean the remaining answers look. A clean-looking margin computed from almost no evidence is a coincidence, not a signal, so the result arrives labeled as a best guess with a second season and a drape test.

Can the quiz be wrong?

Yes, and it is designed to tell you when that is most likely. Near a boundary between two seasons, a different tool or analyst can honestly type you differently. That is why most results include an alternate season and a daylight drape test: fabric against your face in daylight outranks any questionnaire, ours included.