Grey became the default wall colour for long enough that a great many rooms are now painted in one, and a fair proportion of those rooms feel slightly wrong to their owners without it being obvious why. The colour is not too dark or too light. It is the wrong temperature for the room it is in.

Every grey is a neutral built on top of a colour, and that underlying colour is the undertone. Warm greys are built over brown, yellow or occasionally pink, and they behave like a very pale earth colour: they sit back, they flatter skin, and they go soft rather than grubby as the light drops. Cool greys are built over blue, green or violet, and they behave like a very pale cold colour: they stay crisp, they hold their edges, and in weak light they can tip toward looking like a grey day indoors.

The undertone is almost invisible on a paint chip and unmistakable on a wall, which is why so many people are surprised. A small sample is seen surrounded by other colours and by whatever is behind it; a wall is seen surrounded by itself, and at that size the undertone stops being a hint and becomes the colour of the room. Doubling the area of a colour roughly doubles the amount of undertone you perceive.

The reliable way to see it is by comparison. Put the grey you are considering next to a sheet of plain white paper and the undertone appears immediately: against white, a warm grey looks faintly brown and a cool grey looks faintly blue. Better still, put two or three greys beside each other, because the comparison that matters is between candidates and not between a candidate and your imagination.

Then match the undertone to the light the room actually receives. A room whose windows face away from the sun receives cool, even, indirect light all day, and a cool grey in that room compounds the coolness until the space feels like a basement. A warm grey in the same room pushes back against the light and reads as neutral. A room that gets direct warm light for part of the day can carry a cool grey happily, because the light is supplying the warmth the paint is not.

The same logic applies to everything else in the room, and this is the part most often missed. Grey walls sit against wooden floors, and wooden floors are almost always warm, so a cool grey wall above a yellow-toned oak floor puts a cold colour directly against a warm one. That contrast can be handled deliberately, but it has to be handled. The easier road is to let the wall lean the same way the floor does.

Artificial light shifts the whole calculation again after dark. Warmer bulbs push everything toward yellow, which flatters warm greys and can turn cool greys unpleasantly green. Cooler bulbs do the reverse. Since the evening is when most people sit in the room, it is worth choosing the bulb temperature before choosing the paint rather than afterwards.

Test properly. A patch the size of a postcard tells you nothing. Paint two coats onto a sheet of paper at least A3 in size, or straight onto the wall in a square roughly half a metre across, and put a patch on two different walls, because the wall facing the window and the wall beside it will never look the same. Then leave it for three days and look at it at breakfast, in the middle of the afternoon and under the lamps. If it looks like three different colours, that is not a fault in the paint; that is the information you were testing for.