Everything this site claims rests on readings we took ourselves, at most one per listing per day: over the last week we read about 22,753 listings a day, roughly 18% of the 127,062 we track, so most listings are read every few days. Those readings are below, and the rows behind our main claim are a file you can download. We would rather you recomputed our figures than believed them.
127,790 listings appear in the record at least once; 728 of them have since left the catalogue we track, which holds 127,062 today. 2026-08-06 to 2026-10-23. That is 79 calendar days and 79 days of readings — no day is missing.
Of the 61,081 listings we have now read on 14 or more separate days, 40,459 (66.2%) show a “before discount” price higher than anything we ever recorded for them.
We count a listing as unmatched when the seller’s crossed-out price is more than 1.15× the highest price we ourselves recorded — the margin is there so an ordinary fluctuation is not called a false claim. It says the number cannot be matched against our record. It does not say the seller lied, and we do not know their intent.
Download the rows and check us
One line per listing, with our lowest, highest and average reading beside the seller’s figure. Apply the rule above and you should reach the same count we printed.
Download CSV (61,081 rows)A price record is only worth what its continuity is worth. Ours starts on 2026-08-06 and cannot be back-dated — a day we did not measure is gone for good, which is why this table is printed rather than summarised.
| Day | Readings | |
|---|---|---|
| 2026-08-06 | 16,238 | |
| 2026-08-07 | 16,459 | |
| 2026-08-08 | 16,146 | |
| 2026-08-09 | 19,128 | |
| 2026-08-10 | 21,012 | |
| 2026-08-11 | 21,093 | |
| 2026-08-12 | 19,871 | |
| 2026-08-13 | 23,813 | |
| 2026-08-14 | 19,981 | |
| 2026-08-15 | 22,168 | |
| 2026-08-16 | 19,224 | |
| 2026-08-17 | 19,330 | |
| 2026-08-18 | 20,694 | |
| 2026-08-19 | 16,568 | |
| 2026-08-20 | 18,394 | |
| 2026-08-21 | 19,533 | |
| 2026-08-22 | 22,336 | |
| 2026-08-23 | 21,746 | |
| 2026-08-24 | 20,041 | |
| 2026-08-25 | 19,667 | |
| 2026-08-26 | 19,357 | |
| 2026-08-27 | 20,481 | |
| 2026-08-28 | 19,721 | |
| 2026-08-29 | 20,867 | |
| 2026-08-30 | 22,371 | |
| 2026-08-31 | 47,849 | |
| 2026-09-01 | 25,713 | |
| 2026-09-02 | 36,809 | |
| 2026-09-03 | 50,327 | |
| 2026-09-04 | 42,897 | |
| 2026-09-05 | 37,386 | |
| 2026-09-06 | 39,767 | |
| 2026-09-07 | 36,580 | |
| 2026-09-08 | 27,919 | |
| 2026-09-09 | 28,393 | |
| 2026-09-10 | 48,188 | |
| 2026-09-11 | 50,290 | |
| 2026-09-12 | 41,412 | |
| 2026-09-13 | 50,756 | |
| 2026-09-14 | 45,080 | |
| 2026-09-15 | 47,894 | |
| 2026-09-16 | 45,304 | |
| 2026-09-17 | 47,954 | |
| 2026-09-18 | 41,297 | |
| 2026-09-19 | 37,455 | |
| 2026-09-20 | 40,361 | |
| 2026-09-21 | 37,488 | |
| 2026-09-22 | 37,965 | |
| 2026-09-23 | 43,120 | |
| 2026-09-24 | 32,913 | |
| 2026-09-25 | 27,069 | |
| 2026-09-26 | 24,308 | |
| 2026-09-27 | 25,209 | |
| 2026-09-28 | 22,557 | |
| 2026-09-29 | 20,190 | |
| 2026-09-30 | 23,126 | |
| 2026-10-01 | 20,191 | |
| 2026-10-02 | 21,716 | |
| 2026-10-03 | 22,041 | |
| 2026-10-04 | 22,060 | |
| 2026-10-05 | 24,212 | |
| 2026-10-06 | 20,604 | |
| 2026-10-07 | 20,516 | |
| 2026-10-08 | 23,178 | |
| 2026-10-09 | 22,476 | |
| 2026-10-10 | 23,989 | |
| 2026-10-11 | 23,044 | |
| 2026-10-12 | 24,005 | |
| 2026-10-13 | 22,251 | |
| 2026-10-14 | 23,173 | |
| 2026-10-15 | 22,534 | |
| 2026-10-16 | 22,909 | |
| 2026-10-17 | 23,800 | |
| 2026-10-18 | 22,142 | |
| 2026-10-19 | 22,343 | |
| 2026-10-20 | 22,896 | |
| 2026-10-21 | 22,322 | |
| 2026-10-22 | 22,862 | |
| 2026-10-23 | 20,250 |
We ask AliExpress for the current price of the listings we track on a recurring schedule, and write down what comes back with the date. When AliExpress returns no price for a listing that day, nothing is written rather than something guessed. We never edit a past reading. A listing that leaves the catalogue keeps the readings it already has.
We do not use the seller’s crossed-out price as evidence of anything, because AliExpress does not verify it. It appears in our data only as the claim being tested.
We place today’s price in its range only after 7 separate days of our own readings, and we test a seller’s “was” price only after 14. We do not describe any seller as dishonest. Our record either contains a price or it does not; that is the whole of what we know.
Found a figure that does not reconcile? Every price we have recorded is browsable per listing, and the file above is the same data in bulk. If our numbers are wrong we would rather hear it than keep printing them.