Short answer
By keeping the product colourless for as long as possible and adding the colour at the last step. Decorative paint is the clearest case: a few factory-made tint bases and a set of concentrated tinters let a shop mix any shade to order, and a paint maker's technical director put the range formally offered across New Zealand's paint industry at about 12,000 shades. Where colour has to go in early, ranges stay short: the EU reference document says a car model's portfolio rarely exceeds 40 colours and five tones take more than 90% of the volume.
The starting point is that buyers do not spread themselves evenly across a range. The EU reference document for vehicle painting states that for passenger cars, generally five different tones cover more than 90% of the product volume, which it calls high runners, and that a maker's colour portfolio for a given model rarely exceeds 40 colours. Put together, those two statements mean that most of the colours on a list are made for a small minority of buyers. Each of them still needs a paint supply, a place in the sequence and a stream of colour-matched parts. A range is therefore mostly tail. The economic question is what it costs to keep the tail available, and the answer depends on how early in the process a product has to commit to its colour.
Paint makers solved the problem by moving the colour to the shop counter. A technical memorandum by Resene's technical director describes the idea, developed in the United States in the 1950s: a small range of tint bases, each containing a different level of factory-dispersed pigment, to which a known recipe of concentrated tinters is added at the point of sale to produce the whole range. He calls it the salesperson's and accountant's dream, because it offers an effectively unlimited number of colours that are never out of stock without carrying a huge number of stock units. The stock that remains is the bases and the tinters. The same memorandum puts the number of shades formally offered by the New Zealand paint industry at about 12,000, with any other shade available on request.
The memorandum is unusually frank about the price, calling the system the technologist's nightmare. Tinters contain water, glycols and surfactants that weaken the paint they are added to, and the persistent weakness of tinting systems is deep, bright colour with adequate covering power. Resene's answer was more bases, including coloured ones for strong yellows, and stronger tinters. That improves the paint and puts back some of the stock the system was meant to remove. The author notes that makers selling through other people's shops face the opposite pressure: competition for shelf space pushes them towards fewer bases, often as few as 2, with a significant compromise in quality. So the number of bases is the dial. Fewer bases means less stock and worse strong colours; more bases means better paint and more tins on the shelf. The account is one maker's, and it argues for that maker's choice.
Operations researchers call this delayed product differentiation, or postponement. A much-cited paper by Lee and Tang models it: products and processes are redesigned so that the stages common to all variants are prolonged and the point at which variants become distinct is deferred, and the cost of the redesign is weighed against the benefit of holding stock in a generic form. The abstract does not mention colour, and applying it here is Colourwise's reading, but colour is among the easiest attributes to postpone when it sits in a surface layer. It is hard to postpone when it runs through the material or through many matched parts. The one large study of car plants found that more exterior colours meant more labour hours per vehicle and that the paint department was not where the hours went, which its authors put down to colour-dependent parts. A car's colour is applied late, but its bumpers, mirrors and trim have to be made, stocked and sequenced in every colour beforehand.
Set side by side, the documented cases suggest that a colour range is as wide as the cheapest place to hold undifferentiated stock allows. A paint shop holds bases and can sell twelve thousand shades. A truck maker painting fleet liveries to order holds almost no finished stock and runs more than 600 colours. A car maker selling from dealers' stock offers a few dozen. That comparison is an interpretation, and one step beyond it is a hypothesis Colourwise has not seen tested: that short ranges are driven mainly by the risk of holding finished goods in a colour few people choose. Testing it would need records of what was offered, stocked and sold by colour, and how long each waited, which no public dataset contains. Published figures on write-downs of slow colours, or on what a colour adds to inventory, were not found for any industry.
| What it concerns | Colour or finish factor | Published figure | Where and when | Kind of evidence |
|---|---|---|---|---|
| Base coat colours in passenger-car production | How demand is spread across the colour range | more than 90 % of volume in five tones | European Union; Information exchange to 2018; published 2020 | Documented industry practice |
| Colour portfolio for one car model | Number of colours offered | up to 40 colours per model, rarely exceeded | European Union; Information exchange to 2018; published 2020 | Documented industry practice |
| Car assembly plants | Number of exterior colours built | No figure published | 16 countries; 1989–1990 | Measured cost |
| Decorative paint sold through shops | A very large colour range | about 12,000 shades formally offered by the national paint industry | New Zealand; 2004 | Documented industry practice |
| Decorative paint tinting systems in multi-brand shops | Deep and bright colours | 2 bases in the most constrained systems | Asia-Pacific; 2004 | Documented industry practice |
| Any product made in many variants | The stage at which a product takes its colour | No figure published | Not specific; 1997 | Documented industry practice |
| Goods sold from finished stock | A colour few buyers choose | No figure published | Not specific; Proposed 2026 | Colourwise interpretation |
| Product | Where it takes its colour | Stock held before that point | Documented range |
|---|---|---|---|
| Decorative paint | At the shop counter, by tinting | Tint bases and tinters | About 12,000 shades offered across one country's industry |
| Trucks and commercial vehicles | In the paint shop, to the fleet's order | Unpainted cabs and bodies | Typically more than 600 colours |
| Passenger cars | In the paint shop, but with colour-matched parts made beforehand | Bumpers, mirrors and trim in every colour | Rarely more than 40 per model; five tones over 90% of volume |
Why: Tinters add water, glycols and surfactants, and a pale base tinted hard hides poorly.
Fix: For deep or bright colours, ask which base is used and whether a tinted undercoat is specified.
Why: The shop holds bases and tinters; the shades exist only as recipes.
Fix: Count bases, can sizes and product lines to see what is physically stocked.
Why: Colour that runs through the material, or through many matched parts, has to be committed early.
Fix: Ask at which production step the colour is fixed; that step sets the lead time and the minimum.
Each statement is labelled by kind — established fact, a standard’s requirement, observed market data, a convention, or Colourwise’s own interpretation or analysis — with the strength of the evidence behind it.
FactStrong evidence
The EU reference document for vehicle painting states that for passenger cars five tones generally cover more than 90% of product volume and that a maker's colour portfolio for a model rarely exceeds 40 colours.
Caveat: A general statement with no year, maker or market; 'tones' are not defined, and the count is not a market-share survey.
ConventionModerate evidence· New Zealand· 2004
A paint maker's technical memorandum describes point-of-sale tinting as producing a total colour range from a small range of factory-made tint bases and added tinters, without carrying a large number of stock units, and puts the shades formally offered by the New Zealand paint industry at about 12,000.
Caveat: One maker's account with a round figure and no method; it counts shades offered, not shades sold.
Source: Point of Sale Tinting Systems – the 6th Decade of Use: A User's Perspective
FactModerate evidence
The same memorandum states that competition for shelf space pushes paint makers towards tinting systems with fewer bases, often as few as two, at a significant cost in quality for deep, bright colours.
Caveat: Written by a maker that sells through its own shops and favours more bases; no test data are given.
Source: Point of Sale Tinting Systems – the 6th Decade of Use: A User's Perspective
FactModerate evidence
Lee and Tang's 1997 model treats delaying the point of product differentiation as a trade-off between the cost of redesigning products and processes and the benefits of keeping variants common for longer.
Caveat: Only the abstract was read; it does not mention colour, and the model is theoretical.
Source: Modelling the Costs and Benefits of Delayed Product Differentiation
Colourwise interpretationLimited evidence
The width of a colour range tracks where undifferentiated stock can be held: at a shop's tinting machine, in a made-to-order factory, or not at all once finished goods are in a showroom.
Based on: Colourwise's comparison of the tinting memorandum, the EU reference document's figures for car and truck colour ranges and the assembly-plant study, as recorded in the observation dataset.
Caveat: A pattern across three unlike industries, not a tested relationship; no inventory data by colour were found.
Reviewed 6 October 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.