Short answer
In the one systematic study, yes: the Science Museum Group analysed photographs of 7,083 objects from its collection, spanning roughly two centuries, and found the share of grey rising while brown and yellow fell. The authors attribute it mainly to materials — wood and brass giving way to plastic and metal — and warn that a museum collection is not a random sample. There is no comparable dataset for objects on sale or in homes.
The analysis was published in October 2020 by Cath Sleeman on the Science Museum Group's Digital Lab blog, as part of work funded by the Creative Industries Policy and Evidence Centre, and republished by that centre in 2021. It took photographs of 7,083 objects in 21 categories of everyday things — among them clocks and watches, cameras, telephones, lighting, domestic appliances and navigation instruments. For each photograph the background was removed by discarding pixels close to the most common edge colour, and photos without a plain background were excluded. The remaining pixels were pooled by the earliest date associated with each object, in 20-year groups, and the colour mix of each group compared. A second part of the project used a neural network to compare object shapes, finding a drift towards more rectangular forms.
The most common colour across the whole set was a dark charcoal grey, present in more than 80% of photographs, though usually as a small share of each image. Over time the share of grey rises and the shares of brown and yellow fall — the colours of polished wood, brass and leather giving way to the colours of steel, aluminium and grey or black plastic. A smaller counter-trend runs alongside: very saturated colours start to appear from the 1960s, the decade of cheap coloured plastics. Within categories the picture can be sharper. Telephones, for instance, are broadly colourful from the 1960s to the 1980s and turn grey and black from the late 1980s. The study gives these as visual trends and charts, not as percentages of objects.
A museum collects what curators judge significant, and what survives to be collected; it is not a random sample of the objects people bought. Photographs record the object's colour through lighting, camera settings and background removal, so a warm-lit Victorian clock and a flat-lit 1990s printer are not measured on equal terms. Pixel share mixes object colour with object size and shape. The study does not compute a chroma or saturation score, so 'greyer' means more grey pixels, not less saturated colours on average. Secondary coverage has added figures, such as monochrome objects rising from about 15% to 60%, which do not appear in the original analysis and should not be repeated.
The authors' materials explanation fits what is known about manufacturing. Before the twentieth century most everyday objects were made of wood, brass, leather, ceramic and paper, whose natural colours sit in the warm browns and yellows. Industrial materials — steel, aluminium, chrome, and later engineering plastics — are grey or silver in their raw state or are moulded in black and grey, because those pigments are cheap, hide recycled content and hide wear. Electronics added a further layer: from the 1980s office and consumer equipment was often grey or beige plastic, and from the 2000s black, silver and 'space grey' became the default finish for phones and laptops. None of this is measured by the museum study; it is the context its finding sits in.
| Aspect | Detail |
|---|---|
| Published | 8 October 2020, Science Museum Group Digital Lab; republished by Creative Industries PEC, 12 March 2021 |
| Author | Cath Sleeman |
| Sample | Photographs of 7,083 objects in 21 everyday-object categories |
| Period | Objects grouped in 20-year bands by earliest associated date, from the early 1800s |
| Method | Background pixels removed; remaining pixel colours pooled per date group |
| Main finding | Grey rises; brown and yellow fall |
| Secondary finding | Very saturated colours appear from the 1960s |
| Not measured | Chroma or saturation scores; objects outside the collection; sales or ownership |
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 Science Museum Group analysis (Sleeman, October 2020) used photographs of 7,083 objects from 21 everyday-object categories, grouping pixel colours by each object's earliest date.
Source: Colour & Shape: Using Computer Vision to Explore the Science Museum Group Collection
FactModerate evidence
The analysis found a substantial rise in grey and a fall in brown and yellow over time, which the author suggests reflects a move away from wood towards plastic.
Caveat: The author notes the photographs are a sample of a non-random museum collection.
Source: A little greyer and a bit more rectangular?; Colour & Shape: Using Computer Vision to Explore the Science Museum Group Collection
FactLimited evidence
The same analysis identifies a smaller trend of very saturated colours appearing among objects from the 1960s.
Caveat: Described qualitatively; no measure of its size.
Colourwise interpretationLimited evidence
Percentages circulating in press coverage of the study, such as monochrome objects rising to about 60%, are not given in the original analysis.
Based on: Comparison of the Science Museum Group Digital Lab post and its PEC republication, which give no such figures, with secondary coverage that does.
Caveat: Colourwise checked the original posts, not every piece of coverage.
Reviewed 29 September 2026. Colourwise summarises its sources in its own words and does not reproduce standards text or proprietary colour data. Spotted an error? Tell us.