Eye creams: the most expensive skincare, and the one where price works best
I measured 3,782 price changes across 221 eye creams and compared each one against the products that held their price. This is the priciest category per millilitre I have measured — a typical 15 ml jar costs €22 — and it is also the one where discounts move rank the most. Cuts land at one for one. Increases cost noticeably less than they gain. And the gap between the cheapest and the most expensive third is the widest I have seen.
In short
- A 10 percent discount improved rank by about 10 percent — the strongest discount response of the five categories measured so far. Rank moves in step with the price.
- A 10 percent increase cost about 8 percent of rank (range 6 to 10). Demonstrably less than the price gained.
- In the premium third (€27–65), a discount buys 1.37 times its own size. In the budget third (€6–19) it buys 0.66. The two ranges do not overlap — a demonstrated difference.
- That pattern now holds in two of five categories and was absent in a third. It is real where it appears, and it is not a rule.
- Competitors answer about 8 percent more often than chance, usually within a day. Modest.
Price cuts
A 10% cut improved rank by roughly 9 to 12 percent.
Price increases
A 10% increase cost roughly 6 to 10 percent of rank.
The same discount, three price groups, a factor of two between them
Two things make this category unusual. The first is the level: a discount here moves rank in step with the price, which no other category in this series has done. The second is the spread. In the premium third a discount buys 1.37 times its own size; in the budget third 0.66, barely half. Those two ranges do not touch, so the difference is demonstrated rather than merely visible.
Read as a decision, that inverts the usual instinct. A €40 eye cream is the product where a discount pays for itself in visibility. A €12 eye cream is the product where it does not — there a cut gives away more margin than it wins in rank.
Increases behave the other way round and more evenly. In the budget and mid groups they cost demonstrably less rank than the price gained (0.56 and 0.77), and only at the premium end do they reach roughly one for one. For anyone passing on cost increases, the cheap end of this category is the safer place to do it.
The full picture
01 · 1 · Does the size of the discount matter?
Yes, and the first step is the cheapest one.
Rank after price changes of different sizes
A moderate discount of 15 to 25 percent already improved rank by 26 percent, and going deeper than a quarter added only five percentage points on top. The useful part of the discount curve therefore sits in the middle, not at the extreme — unlike face serums, where deeper cuts kept paying in proportion.
The increase side is the mildest first step of any category so far: a rise of 5 to 15 percent cost only 5 percent of rank. Beyond 15 percent the cost climbs quickly to 21 and then 33 percent. So a small increase is close to free here, and a large one is not.
02 · 2 · Will competitors follow?
A little, and quickly — but it is not a repricing war.
Time until the first competitor answers a price cut
Across 5,749 price changes from 234 products, rivals answered a cut 8 percent more often than their own repricing rhythm would produce, and an increase 12 percent more often. Six in ten answers land on the same day and most of the rest the next.
03 · 3 · Where could a new product enter?
€7.85 to €11.53 — the cheap end, where almost nobody competes.
Competition and demand by price range
Two adjacent ranges at the bottom of the category clear the demand baseline and almost nobody sits in them. Between €7.85 and €9.52 there are seven products, more than half of them in the top third of the ranking, with a typical review count of 40 and prices that hold for about fifteen weeks. Of three recent entrants, two are winning and one is building. Directly above, €9.52 to €11.53 holds thirteen products with a similar picture and four judged entrants, three of them winning.
The picture is unusually clear because the crowd sits elsewhere. The two most populated ranges, €24.87 to €30.14 with 47 products and €16.94 to €20.52 with 40, both fall below the demand baseline. Nine recent entrants tried the €25 to €30 range and none is winning. That is where this category is fought and where it is hardest to be seen.
One qualification belongs to the entry corridor: it sits in the budget tier, and this report shows that to be the segment where a discount buys the least. Entering cheap here means winning on product and visibility, and holding the price rather than discounting into the market.
For the category as a whole: of 309 priced products, 87 launched within the last two years, 14 have reached the top third and 2 disappeared.
04 · 4 · What the market looks like underneath
Clean data, and a category that is really about creams.
Of 5,757 price changes found, 948 had another change from the same product running in the same window, 458 had too little ranking data, 245 coincided with a burst of new reviews and 158 with a stockout. 244 events were dropped because the product was already sliding before the price came down.
One number stands out against the other categories in this series: only two events came from products outside this branch of Amazon's own category tree, against 854 in face serums and 1,299 in moisturisers. Eye care is a well-defined shelf — Amazon and the sellers agree on what belongs in it. Observed price changes run from 77 percent off to more than tripling.
Data appendix
362 products on amazon.de, taken from the eye-care listing and then narrowed: no eye serums (a separate format with its own price logic), no patches or masks, which are sold per piece, no rollers, applicators or devices, no tinted or concealing products, no lash or brow lines, no men's lines, no oils, no multipacks or gift sets, nothing below €8 or above €60, container sizes between 10 and 50 ml where stated, and at most three listings per brand so that a single catalogue cannot dominate the control groups. Only products with at least 20 reviews were kept. Products that Amazon does not place in this branch are flagged automatically and excluded from all figures. The product list is available on request for methodological review.
| Step | Events | Removed because |
|---|---|---|
| Price changes found | 5,757 | at least 5% and €0.30, held for 7 days |
| Gaps in the price history | −104 | the change cannot be located precisely |
| Too little ranking data | −458 | less than 70% of days covered |
| Overlapping price change | −948 | another change from the same product in the window |
| Deal or coupon markers | −0 | none recorded in this set |
| Stockout signals | −158 | rank moves for availability, not price |
| Burst of new reviews | −245 | more than 10% and at least 15 new reviews |
| Listing too young | −154 | less than 60 days of history |
| Too few control products | −139 | fewer than five comparable products held their price |
| Passed every check | 4,028 | an event can fail several checks, so the removals sum to more than the difference |
| Used for the figures | 3,782 | from 221 products, after removing 244 with a pre-existing trend and 2 from other category branches |
price change ≥ 5% and ≥ €0.30, held ≥ 7 days · comparison: 14 days before, days 3–14 after · control products: same category, price moved less than 2%, at least 5 of them · rank coverage ≥ 70% of days · listing older than 60 days · review burst = more than 10% and at least 15 new reviews · ranges are 90%, from a bootstrap clustered by product
Observed price changes: −76.6% to +228.3%.
Limits — I measure the ranking, not units sold, in the 14 days after a price change. Results hold only for the price range actually observed. Every figure is for the category as a whole or for one of its three price groups, never for an individual product. Where a range touches the level at which nothing happens, I say so instead of quoting a number.
- 362 products
- 5,757 price changes found
- 4,028 passed every check
- 3,782 used for the figures
- 221 products in the fit set
- ref 23dba463
The Monthly
Each weekly report measures a single category. The Monthly is where they get compared, and it carries the quarterly report when one is published.