Why we track vintages
Light through a paper window fell across the lot — a single 2003 Yìwǔ Chá (易武茶) cake, its wrapper faintly foxed, the leaves still carrying the quiet camphor note of well-judged dry storage. When the auctioneer’s gavel came down at a spring sale, the price outstripped the pre-sale estimate by a wide margin. Collectors who follow such lots on shop.puerh.app recognised the pattern; for everyone else it was a headline without context. That gap between headline and context is why the vintage tracker exists.
Auction data for aged Chinese teas — especially raw pu-erh cakes — is notoriously noisy. One house reports prices in Hong Kong dollars per gram, another in euros per kilogram, yet another bundles several cakes into a single “lot” of mixed condition and provenance. The free vintages page already shows this problem plainly: each row keeps the auction house’s own reported currency and unit rather than pretending to a false precision. The 2026 annual report is our attempt to take that same discipline further — a larger, hand-checked collection of realised prices for a full auction season, with a documented methodology note rather than a single normalised number presented without its working.
We are not promising “over two hundred vintages” or a finished interactive dashboard before we have built and verified them — the free page today lists thirteen dated, sourced lots, and the annual report’s job is to grow that number honestly, lot by lot, not to arrive pre-filled. What the report will add once it exists: a written methodology note (built with reference to the technical grounding on tea.school), provenance-chain notes per entry, and a link back to each original catalogue.
What a reservation gets you
- Priority access to the report the moment we publish it, rather than waiting for a public release.
- A written methodology note explaining exactly how each figure was sourced and why a given lot was or was not included — the same standard already visible on the free vintages page.
- A direct line to the tea.money research desk if you want to flag a lot, a source, or a correction before publication.
We deliberately do not promise a fixed count of entries, a specific statistical finding, or a delivery date more precise than “March” — those numbers do not exist yet, and we would rather under-promise here than repeat the mistake of describing a report as finished before a single verified row was collected.
Reserve a window and we will follow up as the season’s data comes together.