How do I know if I am warm or cool toned?

Hold something gold and something silver under your chin in daylight, one at a time, and watch your skin rather than the metal. Under the right one your skin looks even and rested. Under the wrong one it looks slightly grey or slightly yellow. If both look fine you are probably neutral, which is common and useful, because it means depth and clarity matter more for you than temperature does.

What if I get a different season on every website?

That usually means one of the four qualities is genuinely borderline for you, and different sites weight the four differently. Work out which of the four you were least sure about, because that is the one moving your answer. A result you get twice on two different days is worth acting on; one that moves each time was never a measurement.

Can I wear black if it is not in my palette?

Yes. Palettes describe what sits well directly against your face, and black is only a problem there if you are light and low contrast, where it can overwhelm. Wear it below the waist, or keep a scarf or collar in one of your own colours between it and your skin, and the objection disappears.

Does my colour season change?

The underlying qualities are stable, but two things shift the reading. Sun exposure changes surface tone enough to move a borderline result between neighbouring seasons, and hair colour changes contrast directly. Going several shades lighter or darker genuinely does change which colours work on you, which is the one case where a re-test is worth doing.

Is a paid colour analysis worth it?

It removes the camera and the lighting, which are the two largest sources of error, and adds a trained eye. It also costs money and remains a judgement rather than a measurement. The honest comparison is that in-person analysis is more reliable and less repeatable, and a good tool is less reliable and perfectly repeatable.

What colours suit me? Four questions decide it, and only one of them is about temperature

Most colour quizzes ask whether you are warm or cool and stop. That is one axis out of four, which is why the result so often feels close but wrong.

Last reviewed 3 August 20268 min readSravya

Editorial portrait showing a warm colour drape beside the face

The figures

4
Qualities that decide it: 12
Seasons they resolve to: £0
Cost of testing it properly

The short version

Stand by a window in daylight with no makeup and your hair back. Ask four things. Does gold or silver look better against your skin. Do you read light, mid or deep overall when you squint at yourself. Is there a big gap between your hair and your skin, or do they blend. And do clear bright colours or dusty muted ones make you look well rather than tired. Those four answers, in that order, narrow twelve seasons to one. Anything that asks only the first has answered a quarter of the question.

01Do it in daylight or do not bother

Every one of these tests is a comparison of colour, and indoor bulbs shift colour more than most people expect. A warm domestic bulb pushes skin towards gold, which makes almost everyone test warm. Daylight near a window, indoor lights off, is not a nicety here. It is the difference between a reading and a guess.

02The four questions, in order

What each question is actually measuring

Quality	The test	What it decides
Temperature	Hold a gold item and a silver item under your chin in turn. Look at your skin, not the metal.	Whether you sit in the warm families or the cool ones.
Depth	Squint at yourself in the mirror until detail drops out. Do you read light, mid or deep as a whole.	Whether you belong at the light end or the deep end of your family.
Contrast	Look at a photo of yourself. How far apart are your hair and your skin.	How strong a colour you can carry without it wearing you.
Clarity	Hold a clean bright colour and a dusty muted one of the same shade under your chin.	Whether your colours should be clear or softened.

Answer them in that order and do not go back. The first answer that felt obvious usually is, and the one you keep changing your mind about is the one that matters, because it is the axis your result turns on.

03Why four seasons was never enough

The original system had four and sorted almost entirely on warm against cool. It fails on a common case. Two people can both be cool, and one of them light and low contrast while the other is deep and high contrast. Put them in the same palette and one of them is wearing colours that flatten them.

Twelve seasons split each of the four along depth and clarity, the two qualities the original ignored. That is why a modern reading scores four things and why a result that only says warm or cool has not finished.

Every season, every colour

12 seasons · 72 shades

Light Spring

Warm Spring

Clear Spring

Light Summer

Cool Summer

Soft Summer

Soft Autumn

Warm Autumn

Deep Autumn

Deep Winter

Cool Winter

Clear Winter

These are the actual palettes the colour tool works from, not a sample. Point at one to hold it. Twelve seasons rather than the old four, because four cannot separate someone light and cool from someone deep and cool.

04Test the palette, not the label

A season name is a filing system, not an outcome. What you actually want is the ten or so colours you will buy again, and you find those by holding candidates near your face in daylight and looking at your skin rather than at the fabric.

1Take two garments you already own that people compliment you in. Look for what they have in common on all four qualities, not just hue.
2Take one you never reach for despite liking the colour. That is usually a clarity or a depth mismatch rather than a temperature one.
3Screenshot your palette and hold the phone up at the rail. Matching a garment to a swatch beats remembering a season name in a shop.
4Ignore any rule that says you cannot wear a colour. The palette tells you which version of that colour is yours, not that it is off limits.

Common questions

Sources

Every figure quoted above is listed here with what it was used for and when the source was last checked. Anything not attributed to a source is this site’s own reading and is labelled as such where it appears.

Society for Imaging Science and Technology
Shades of Gray and Colour Constancy (Finlayson and Trezzi, 2004)
The white balance correction applied before any skin colour on this site is measured.
Last checked 2 August 2026

Check it against your own measurements

The tool asks the same four questions this page describes, scores them separately, and tells you which of the four your result actually turned on. Change that one answer and watch the season move.

Work through the four