Colour, Contrast & Coding

The first job colour does on a sign is to make the text legible, and the crucial insight is that legibility comes not from the colours themselves but from the contrast between the text and its background - specifically the difference in lightness (how light or dark), not merely the difference in hue. A dark word on a light panel or a light word on a dark panel reads strongly because the lightness gap is large; two colours that are both bright and vivid but similarly light - a saturated red on a saturated green, say - can look striking and still read terribly, because there is little lightness difference for the eye to grab. This is why you must judge a sign's legibility by squinting or imagining it in greyscale: if the text nearly disappears in greyscale, it has a contrast problem no matter how colourful it looks.

This matters enormously because the people most dependent on signage often have the least contrast sensitivity - older eyes and low-vision users lose the ability to distinguish subtle lightness differences first - so a sign that is just legible enough for a young designer on a bright screen can be illegible to the very person who needs it. Strong contrast is therefore not an aesthetic preference but an accessibility essential, and it is one of the highest-leverage things you can get right on a sign.

Because contrast is so important and so measurable, it is exactly the kind of thing standards specify and you defer to them. Accessibility standards set minimum contrast levels between text and background (and the principles are shared with digital accessibility guidance like the Web Content Accessibility Guidelines, which quantify contrast ratios). Treat any ratio you have in your head as illustrative, to verify against the current accessibility standard (in India, NBC 2016 and the Harmonised Guidelines) and an accessibility consultant - and remember the standard sets a floor, not a target; more contrast is usually better for the stressed, older, low-vision reader. Practical moves that help: choose genuinely light-versus-dark combinations rather than two mid-tones; avoid putting text over busy images or patterns; watch out for glare and reflection on glossy surfaces, which destroys effective contrast in the real world even when the numbers look fine on paper; and, once more, verify the sign in a real mock-up in the real light, because measured contrast and perceived contrast can differ.

The second, quite different job colour does is coding - using colour systematically to group, identify and connect parts of a place so people can orient and remember by colour. The familiar examples are everywhere: the coloured lines of a metro (the Red Line, the Blue Line), the colour-coded zones or wings of a hospital, the level colours in a multi-storey car park, the department clusters in a large campus or office. Done well, coding is a genuine aid: a consistent colour gives a zone an identity, lets a map and the signs reinforce each other ('follow the green route to Maternity'), helps people build a mental map, and offers a quick, pre-verbal sense of 'am I still in the right area?'

For coding to work, a few disciplines are essential. The colours must be genuinely distinct from each other - not just in hue but in lightness and in how they read under the building's actual lighting - and the palette must stay small; people can hold only a handful of colour-codes in mind, so a system with fifteen colour-coded departments asks more of memory than it delivers (beyond roughly half a dozen, colours start to blur together and lose their mnemonic power). The coding must be applied consistently and completely across maps, signs, and ideally the environment itself (a coloured floor band, a wall, a supergraphic), because a code that appears on the map but not reliably in the corridor breaks trust. And the colour should be reinforced by the architecture where possible - colour-coding works best when it rides on a spatial logic (each wing really is a distinct place), not when it is asked to substitute for one.

There is an important limit even when coding is working: colour codes lean heavily on memory and on a key. A colour only means something if you have learned or can look up what it stands for - red means nothing until a map or sign tells you 'red = Cardiology'. So coding supplements naming and directing; it does not replace them. The strongest systems pair the colour with a name, number or symbol at every point (the 'Red Line' is also labelled 'Red' and often numbered), so the code accelerates recognition for those who have learned it while the label carries the meaning for everyone else. That pairing is not only good for memory - it is the bridge to the next section, because it is also what makes colour-coding survive the people who cannot see the colour at all.

Colour-coding is powerful, but it is fragile in specific, predictable ways, and a responsible designer treats those limits as first-order, not footnotes. The most important is colour vision deficiency. Around one in twelve men (and a smaller share of women) has some form of colour blindness, most commonly difficulty distinguishing reds from greens - and red and green are, not coincidentally, among the most popular coding colours. To such a user, a red route and a green route may look identical, a 'follow the red line' instruction is meaningless, and a system that relies on telling those colours apart simply does not work. This is not a rare edge case; in any large public place, many hundreds of people a day experience your colour code differently from how you designed it, and in a country of India's scale that is an enormous number of real users.

The second limit is memory. A colour code is only useful once learned or looked up, and stressed, first-time, hurried visitors - the true client of wayfinding - are exactly the people least able to memorise a legend and most likely to forget which colour was theirs halfway down a corridor. A regular learns the code and loves it; the anxious stranger it was partly meant to help may never hold it. The third limit is environment and consistency: colours shift under different lighting (the confident palette on screen looks muddy under warm corridor light or sodium lamps), surfaces fade and get repainted, print and paint and screen never match exactly, and any drift between the map's colours and the building's colours quietly destroys the code. A fourth is culture: colours carry meanings that vary between cultures (associations around red, white, green and so on differ), so a coding or warning colour that reads one way for one group may read differently for another - worth weighing in a diverse, international or multi-faith setting.

None of this means abandon colour-coding - used well it genuinely helps the majority and adds identity and orientation. It means design it knowing it will fail for some people and in some conditions, keep the palette small and the colours distinct in lightness as well as hue, test it (including with colour-blindness simulation and, ideally, colour-blind users), and - above all - never let it be the only thing carrying a message. Which is the rule the whole lesson has been building toward.

Everything in this lesson converges on one rule, simple enough to carry forever and strict enough to design by: never rely on colour alone to carry meaning. Colour is a superb reinforcement - it speeds recognition, groups zones, adds identity and helps the mental map - but it must always be redundant, backing up a cue that stands on its own without it. The test is blunt and useful: would this sign or system still work in greyscale, for someone who cannot perceive the colours at all? If the answer is no, it is not finished.

In practice, redundancy means pairing colour with something colour-independent at every point it carries meaning. A colour-coded line or zone is also given a name, a number, and ideally a symbol or a pattern - the 'Red Line' is labelled 'Red' and numbered and placed consistently, so a colour-blind rider navigates by the word and number while a colour-sighted rider enjoys the speed of the colour. A warning conveyed by red is also conveyed by a word, a symbol and often a shape (which is precisely why safety signs standardise shape and symbol, not just colour). A status shown by colour (available/occupied, go/stop) is also shown by a label, an icon or a position. This redundancy is not clutter or compromise; it is simply inclusive design, and it has the happy property of making the system clearer and more robust for *everyone*, not only those who cannot see the colour - the hallmark of good universal design.

And the companion rule, for contrast: defer the binding numbers to the standard. The minimum contrast between text and background is a safety-and-accessibility matter that the current accessibility standard specifies, with shared principles in guidance like the Web Content Accessibility Guidelines; treat any ratio you carry as illustrative as of 2026, verify it against the current standard (NBC 2016, the Harmonised Guidelines) and an accessibility consultant, meet the minimum as a floor and exceed it where you can, and confirm it in a real mock-up under the real lighting. Put the two rules together and colour becomes exactly what it should be on a sign: a powerful, delightful, orienting reinforcement layered on top of a system that is already complete and legible without it - contrast you have verified against the standard carrying the legibility, and a name, number or symbol always standing behind every colour. That is how you get colour's genuine benefits while keeping faith with the person who cannot use it - which is, in the end, what this whole module and this whole course are about.