Foundation ranges used to be narrow, and a substantial proportion of buyers could not find a shade that matched them. That has changed considerably, and how it changed is instructive.
The situation before
Ranges of a dozen or so shades were standard, clustered in a narrow band, with a small number at either extreme that were frequently poor matches.
The industry explanation was commercial — that darker and lighter shades sold slowly and were unprofitable to produce and stock.
That reasoning was somewhat circular. Shades that were poor matches sold slowly, which was taken as evidence that the demand was absent.
Meanwhile a substantial population mixed products, used shades that did not match, or bought from specialist brands that had served them for decades and received little mainstream attention.
What actually shifted it
Argument had been made for years without much effect. What changed the industry was a commercial demonstration.
A launch with a very wide range at once, marketed on that basis, which sold extremely well and specifically sold well in the shades the industry had claimed were unprofitable.
That provided evidence rather than an argument, and the response across the sector was rapid.
Within a few years, wide ranges became a competitive expectation, and launching narrow attracted criticism that was commercially damaging.
Why matching is technically difficult
Worth understanding because it explains why more shades alone was insufficient.
Skin colour varies along more than one dimension. Depth is the obvious one. Undertone — the warm, cool or neutral cast — is the second and is what makes a shade of the correct depth still look wrong.
A range that extends depth without covering undertones at each depth produces the situation where a buyer finds something the right darkness that reads as ashy or orange.
The better ranges are constructed as a grid rather than a line, with multiple undertones at each depth, which multiplies the number of products required.
That is the actual cost of a proper range and it is why the count alone is a poor measure.
The stock problem
The failure that persists after the ranges expanded.
A wide range on a brand's website is not a wide range in a shop. Retailers stock a selection, and the selection is frequently narrower at both ends.
Which means a buyer can be told a range covers them and find nothing available to test.
Testing matters here more than in almost any other category, since matching by photograph is unreliable, and the inability to test is a substantial barrier that shade counts do not address.
How to match without testing
Practical, since online buying is now the norm.
Match to the jaw and neck rather than to the hand or the inner arm, since those differ in colour from the face.
Test in daylight, since retail lighting distorts substantially.
Allow it to sit for several minutes, because many formulations oxidise and shift toward orange after application.
Use shade-matching tools cautiously. They work reasonably for depth and are less reliable for undertone.
And take advantage of sample programmes, which several brands now run and which are the only genuinely reliable method at a distance.
The direction now
Some developments worth noting.
Adjustable products — drops that shift depth, or bases designed to be mixed — reduce the number of stock units required and give the buyer control.
Ranges built around undertone families rather than a single line of depths, which is the technically correct structure.
And a general reduction in the emphasis on coverage, with lighter formulations that are more tolerant of an imperfect match, which sidesteps some of the problem.
What the episode demonstrates
Beyond the specific category.
An industry claimed for decades that demand did not exist, and the claim was not tested because nobody had an incentive to test it.
One party tested it, the claim was false, and the sector adjusted quickly once the evidence existed.
Which suggests the barrier was never really commercial reasoning but the absence of anybody willing to find out, and that is a pattern worth recognising elsewhere.
The photography problem
A related failure that persists.
Product imagery and swatches are photographed under lighting and with processing that alters colour substantially, which makes online matching unreliable even with a wide range.
Swatches on a model's arm, in unstated lighting, are close to uninformative.
What helps is imagery shot in daylight without retouching, video rather than stills, and images of the product on a range of people rather than on one.
Several brands now do this deliberately, and it is a reasonable signal about whether a range was designed seriously or assembled to meet an expectation.
Undertone, identified simply
Since it is the part people find hardest.
The common methods — looking at vein colour, testing against jewellery metals, considering how skin responds to sun — are rough heuristics rather than reliable tests, and they contradict each other frequently.
What works better is comparison. Placing two shades of the same depth side by side on the jaw makes the difference visible in a way that assessing one alone does not.
Daylight is essential for this, and the answer is generally obvious once two options are compared directly.