Room Colour Coordinator

How the recommendations are calculated

There is no stored palette and no model. Every colour in the catalogue is scored against what you entered, and the scoring is made of five named terms, each of which is a real claim about how colour behaves rather than a preference.

Effect fit.

Each desired effect is expressed as target ranges in lightness and saturation rather than as a list of colours — “cosy” is not a colour, it is warmth plus restrained lightness plus moderate contrast, and many colours satisfy it.

Visual scale.

Every element has a chroma ceiling and floor set by how much of the view it occupies. A dominant surface is held well below the saturation an accent needs; an accent below its floor stops reading as a decision and starts reading as an accident.

Near-miss avoidance.

Anything close to an existing colour without matching it is penalised. Two elements nearly but not quite the same is the commonest way a real room goes wrong, and it reads worse than either matching or clearly differing.

Lighting resilience.

A surface can only reflect what the light contains. Under a warm source there is little short-wavelength energy for a blue to return, so a barely-tinted cool colour goes grey. The penalty falls on colours with too little saturation to survive the light they will be seen in.

Material bias.

Each material carries a cast strength and a temperature, weighted by how much light it bounces. Those add to a single figure, and the tool uses it to correct a strongly biased room rather than reinforce it — unless the effect you asked for wants the reinforcement.

The first four are properties of colour and light. The fifth is closer to editorial judgement, informed by the physics but not derivable from it — worth knowing when you decide how much weight to give the result.

Common questions

How is this different from a palette generator?

A palette generator answers 'give me some colours'. This answers 'given what is already in my room, what will actually work'. The difference shows up in the output: every recommendation is calculated against your anchor colour, your materials and your light, and each one comes with the reasoning, the way it could fail, and substitutes if you do not want it. It also decides the elements in sequence rather than independently — the rug is chosen knowing what the sofa is.

Why does it ask which way the room faces?

Because a colour is not a property of a surface; it is what that surface does with the light available. In the northern hemisphere a north-facing room never receives direct sun, so it is lit by sky — steady, soft and distinctly blue. The same paint in a south-facing room is lit by direct sun for much of the day and shifts markedly from morning to evening. Colours that look identical on a chart behave quite differently in those two rooms.

Why does the wood tone change the recommendations so much?

A large timber surface behaves as a secondary light source, not just as an object — a good deal of the light reaching everything else has bounced off it first and carries its colour. Oak throws a yellow cast, pine throws a stronger one, concrete pulls the other way. The tool adds those up into a single bias figure and states it, because it is the mechanism behind most 'it looked different in the shop' complaints.

It suggested a colour I do not like. What now?

Use the alternatives. Each one names the property it preserves — lightness, warmth, saturation — so you can pick the substitution that keeps whatever the original was doing for the scheme. That is the useful question, and it is usually not 'what looks similar': stone substitutes for cream when the job was lightness and warmth, and light grey substitutes when it was lightness and neutrality. Those are different answers to different problems.

Should every element in a room be a different colour?

No, and the tool will not suggest that. What it avoids is the near-miss: two elements close enough to look like an attempted match but far enough apart to fail it. Repeating a colour exactly is a legitimate and often good choice — controlled repetition is one of the main ways a room reads as cohesive. It is the almost-match that looks like a mistake.

Does it know about my room's size or proportions?

No. It knows the visual scale of each element — that a wall is a dominant surface and a cushion is an accent — and it uses that heavily, because how much of the view a colour occupies matters more than most advice admits. But it does not know your ceiling height, your window placement or what you already own, and all three can override what it says.