What the lipstick index is
The lipstick index is the claim that spending on small, affordable indulgences holds up or rises when the economy weakens, while big-ticket discretionary spending falls. Leonard Lauder of Estee Lauder is credited with popularising it, pointing at lipstick sales that rose 11% in the final quarter of 2001. It is a nice story, it gets retold every downturn, and it is not what you should bring to a category review.
Bring the test instead. Trade-down is a price and mix phenomenon, which means it is fully measurable in transaction data without asking anyone how they feel about the economy. Two reads settle it in about an hour.
The evidence is genuinely mixed, which is the point
The lipstick index has been checked and it does not hold up as a general law. The 2001 spike was later attributed to a wave of celebrity-designed cosmetics brands rather than to recession behaviour, and during the late-2000s recession cosmetics sales fell along with everything else.
NielsenIQ went further in 2022 and measured it directly. Over the 12 weeks ending 13 August 2022, lower-income shoppers cut beauty spending 3% in dollars and 11% in units, and NielsenIQ concluded that "the 'Lipstick Index' theory may no longer apply to the Beauty and Personal Care industry." Fragrance in the same window went the other way, up 16% in dollars and 3% in units, driven by higher-income and Gen Z shoppers.
Read those two together and the useful version emerges. It was never about lipstick. It is about where in a category's price ladder the units go when budgets tighten, and that varies by category, by banner and by income mix. Which is why you run it on your own data rather than quoting 2001.
Test one: units by price tier
Split a category into price tiers, hold the tier boundaries fixed, and compare units by tier against the same weeks a year ago. Dollars will mislead you here because list prices moved; units by tier will not.
Packaged confection at a regional grocery banner, 13 weeks, tiers set on shelf price.
| Price tier | Avg price, year ago | Avg price, this year | Units, year ago | Units, this year | Unit change |
|---|---|---|---|---|---|
| Opening (under $2.50) | $2.09 | $2.15 | 184,000 | 226,000 | +22.8% |
| Mid ($2.50 to $3.99) | $3.29 | $3.35 | 262,000 | 249,000 | -5.0% |
| Premium ($4.00 to $5.99) | $4.79 | $4.95 | 138,000 | 112,000 | -18.8% |
| Super-premium ($6.00 and up) | $6.99 | $7.29 | 46,000 | 33,000 | -28.3% |
| Category | $3.54 | $3.41 | 630,000 | 620,000 | -1.6% |
Look at what the last row does. Every single tier raised its average price, and the category's average selling price still fell, from $3.54 to $3.41, a drop of 3.6%. That gap is the whole finding. Nobody discounted; the units walked down the ladder. Category units are down only 1.6%, so this is not a demand collapse. It is a mix shift, and it is the mix shift that costs you: dollars fell 5.1% on a 1.6% unit decline, and the missing 3.5 points is entirely composition.
The arithmetic that makes it visible: 184,000 x $2.09 and so on down the tiers gives $2,229,100 a year ago against $2,115,020 this year, on 630,000 and 620,000 units. Divide and you get the $3.54 and the $3.41.
Super-premium down 28.3% while the opening tier is up 22.8% is the lipstick index in its measurable form. The small indulgence held, and it held at the cheap end of the aisle.
Test two: the small-indulgence categories in the basket
The second read is cross-category and needs basket data rather than category totals. Pick the categories that behave as affordable treats in your stores, single serve confection, packaged cookies, ice cream novelties, salty snacks, premium coffee, and track two things through the period: units, and their incidence in the basket.
In the same banner and the same 13 weeks as the table above, single-serve confection units ran 6.2% ahead of a year ago while boxed gift confection ran 19% behind. Same category, same shopper, opposite direction. The affordable treat did what the lipstick index predicts, and the giftable version did the opposite, which is a far more actionable read than either a category total or a consumer survey.
Two guardrails. Hold distribution constant, since a tier that gained facings will show unit growth that has nothing to do with trade-down, and check promoted share by tier, because an opening tier that ran four extra deal weeks manufactured its own trade-down.
Premiumisation is the same measurement, reversed
Run the identical test in an expansion and the pattern inverts: units migrate up the ladder, the super-premium tier outgrows the category, and average selling price rises faster than any individual tier's price. That is premiumisation, and it is one measurement, not two. The same tier table answers both questions, so build it once and read it every quarter rather than reaching for it when the news gets bad.
The corollary is worth stating plainly. Premiumisation and trade-down often run at the same time in the same category, because the mid tier is the one that empties in both directions. A category that looks flat is frequently a barbell forming.
Where Scout fits
Scout builds the tier table off your own POS or syndicated file, holds the tier boundaries fixed across periods so the comparison stays honest, and splits promoted from base weeks so an opening tier that just ran a deal does not read as a trend. One boundary to state plainly: POS records transactions, not people. It shows units moving down a price ladder. It cannot tell you that shoppers felt anxious, and any page that turns a mix shift into a claim about consumer sentiment has left the data behind. Attributing the movement to household income needs loyalty-linked demographics or a survey, neither of which is a scan file.
The short version
- The lipstick index is the claim that small indulgences hold up when budgets tighten. As a general law it has failed repeated checks, including a direct NielsenIQ test in 2022.
- As a measurement it works. Split a category into fixed price tiers and compare units year over year.
- The signature is every tier raising its price while the category's average selling price falls. In the worked example that was $3.54 to $3.41, with super-premium units down 28.3% and the opening tier up 22.8%.
- Confirm it in the basket. Single-serve confection ran 6.2% ahead while boxed gift confection ran 19% behind, in the same stores and the same weeks.
- Reversed, the same table measures premiumisation. Build it once and read it every quarter.
Sources: Wikipedia, "Lipstick effect"; NielsenIQ, "Is it time to reassess the 'Lipstick Index'?".