Your barista team or your QA lead says the new lot "looks duller" or "tastes flatter" than the one before. Before you start shopping for a different Japanese supplier, you need to know where the change happened: at the source (a different harvest or a re-balanced blend), on the trans-Pacific and US inland leg, or inside your own distribution center after the bag was opened. This article is a cause-attribution workflow for US-landed matcha, written for buyers who already have a supplier and an approved standard.
Three neighboring articles cover adjacent ground and are worth reading instead if they match your actual question. Matcha standardization and recipe development is about designing drinks so they taste the same across sites — it assumes the powder is consistent and does not tell you how to attribute a drift you have already observed. Arrival quality and freight diagnosis on the Australia lane works through arrival damage for an AU importer on a long southbound ocean route, so the port and inland conditions it assumes are not yours. Matcha storage for businesses sets out general storage rules rather than a method for comparing two lots. What follows is US-specific and comparison-specific.
Why most drift complaints are decidable without a lab
Matcha changes in predictable directions. Dry color moves from vivid green toward olive or yellow-brown; the slurry loses depth; aroma flattens; bitterness and astringency come forward as sweetness and umami recede. The problem is that two completely different causes produce a similar-looking result. A later-harvest or re-balanced blend at source can land greener-but-different, or duller from the first cup. Heat and oxygen exposure after milling produce the same visual and sensory direction. Looking at one bag tells you nothing about which one you have.
What separates them is a sealed control. If you hold back a sealed retain from every lot and compare it side by side with the lot people are complaining about, the logic becomes simple:
- The drift shows up in the sealed retain of the new lot — the powder was already different when it arrived. The cause sits upstream: harvest period, blend composition, mill date, or handling before the container was sealed. You fix this in the purchase spec and the PO wording.
- The sealed retain of the new lot matches the approved standard, but the working bag does not — the powder arrived correct and changed in your possession. You fix this in receiving, storage, and dispensing.
- Both the retain and the working bag of the new lot differ, and older retains held at your DC have also drifted — your retain storage itself is not cold or sealed enough, and you have no reliable reference at all. Fix the retain program first; you cannot attribute anything until you have one.
In our own handling experience — an operating observation, not a measured study — buyers who switch suppliers over a lot complaint without a retain comparison frequently discover the same complaint with the next supplier, because the condition that caused it was in their own inbound and storage chain.
The retain protocol: two sealed samples per lot
Nothing downstream in this article works without retains. Set it up before the next shipment lands, not after the next complaint.
- Pull at receiving, not later. Take the retains when the pallet is broken down, from a bag you then reseal or consume immediately. A retain pulled from a bag that has been open for a week is not a record of arrival condition.
- Two retains, roughly 30 g each, in opaque foil pouches with as little headspace as you can manage, heat-sealed. Nitrogen flush if you have the equipment; if not, seal tight and small.
- Store one at your DC under the same conditions as your working stock and one in a freezer or chiller reference. The DC retain tells you what the lot became under your normal conditions; the cold retain is your closest thing to a frozen clock.
- Label both with lot number, mill date, harvest designation, receiving date, and container or PO number. A retain without a mill date cannot be compared meaningfully against anything.
- Keep retains at least as long as the shelf life you promise downstream, so that a complaint arriving late in the life of a lot still has a reference.
Also retain, in a file rather than a pouch: the supplier's spec sheet for that lot, the bill of lading, the container number, any temperature record the forwarder can produce, and the date and dock time of the inland delivery. These are what you will reach for when the retain comparison points at the freight leg.
The four-cell comparison
Run the comparison as a grid: sealed retain versus working bag, crossed with approved standard versus new lot. Score all four in the same session, under the same light, by the same two people. Scoring one cell on Tuesday and another on Friday introduces more variation than the drift you are chasing.
A workable scoring card, kept to a single page:
- Dry color. Spread a level spoonful on white paper under the same light source every time — daylight-balanced bulbs, never mixed daylight and fluorescent. Score against a fixed reference chip or against the approved standard sitting beside it.
- Slurry color after two minutes. Fix the ratio and water temperature, whisk the same way, and read the color two minutes after whisking. The two-minute wait matters: an oxidized powder often looks acceptable at the moment of whisking and turns duller as it stands.
- Aroma. Dry aroma and then the whisked aroma. You are listening for the sweet, marine, freshly-cut-grass note giving way to hay, straw, or nothing at all.
- Bitterness and astringency, scored separately from umami and sweetness. Drift usually shows as the first pair rising while the second falls.
- Foam — height and how long it holds. Relevant if your menu depends on it.
- Behavior in your actual application. One latte at your standard build, or one bakery mix at your standard dose. A drift that disappears at 2% in a muffin batter is a different commercial problem from one that is obvious in a straight ceremonial-style serve.
Two scorers, independently, then compare. If your two scorers disagree wildly, the lighting or the protocol is the problem, not the matcha. A colorimeter is a useful addition if you already own one, but disciplined visual scoring with a fixed reference and fixed lighting is enough to make the attribution call — the goal is to separate two causes, not to publish a number.
What the US lane actually does to matcha
If the comparison points at the freight and storage leg, these are the places to look, in the order they occur.
The ocean leg. Matcha moving from Japan to a US west-coast port has a shorter ocean exposure than matcha routed to the east coast, whether via the Panama route or transshipped. Ask your forwarder for the routing and, if reefer or temperature-controlled service was used, for the temperature record. If the powder shipped as ambient dry container, it experienced whatever the container experienced — and a dry container parked in the sun is a hot box, regardless of the sea temperature.
Port and deconsolidation. Time sitting on a terminal or at a CFS, where the pallet may be outdoors, is exposure that rarely appears on any document. Ask what the free time was and whether it was used up.
The inland leg. This is where US routing genuinely diverges. A west-coast port to a west-coast DC is a short dray. A west-coast port to an inland or east-coast DC in July is days of trailer interior temperature that no one recorded. If your complaints cluster in the lots that landed between late spring and early fall, the inland leg is a strong candidate — check the mill and arrival dates on your retains against the complaint dates before you conclude anything.
The DC itself. "Ambient" in a US DC means very different things in Sacramento in August and in a climate-controlled facility in the Midwest. Put a logger in the rack where the matcha sits — not on the office wall — and read it for one summer. Cartons on a dock door aisle see something quite different from cartons in the middle of the building.
After the bag is opened. This is the largest and most controllable source of drift, and the easiest to overlook because the lot number never changes. A 1 kg foil bag opened daily at a busy store, folded over with a clip, sitting under the counter next to an espresso machine will drift within weeks. The same bag decanted into small daily-use containers, resealed, and held cold between uses behaves very differently. If your working-bag score is poor while the sealed retain of the same lot is fine, this is almost always where it happened.
When the source is the cause: what to change in the spec and the PO
If the sealed retain of the new lot differs from the approved standard, the conversation belongs with your supplier — and it is much more productive if you arrive with the retain comparison rather than an adjective. Matcha is an agricultural product, and blends are re-balanced as crops change; the question is not whether that happens but whether your spec constrains it.
Clauses worth having in writing on the spec sheet and referenced in the PO:
- Harvest designation and period. First harvest, later harvest, or a stated mix. A blend that quietly shifts toward later-harvest leaf will read as more bitter and less sweet, and the spec is where you catch it.
- Blend window. If your product is a blend of cultivars or regions, ask for the ratio to be stated with an allowed range rather than left open.
- Color reference. A named reference — a physical chip, a retained standard sample held by both parties, or a numeric color value if both sides measure — with a stated tolerance. "Vivid green" is not a spec.
- Particle size range. Particle size drives both perceived color and mouthfeel; a coarser grind reads duller and rougher even when the leaf is identical.
- Mill date and maximum age at shipment. Ask for the mill date printed on every bag and agree a maximum number of days between milling and departure. This single clause removes a large share of ambiguity, because it separates "the leaf was different" from "the powder was older."
- Packaging format. Bag size, foil construction, nitrogen flush or not, and whether cartons are lined. If your opened-bag drift is the problem, a smaller bag size is often the cheapest fix available and belongs in the PO.
- Retain exchange. Ask whether the supplier keeps a retain of the same lot. Comparing your retain against theirs separates "the lot shipped different" from "the lot changed in transit" with more confidence than anything else on this list.
Then re-approve the standard deliberately rather than by drift. If the supplier's new blend is genuinely acceptable to your menu, approve it, retain it, and date it — so that the next comparison is against something current instead of a sample nobody can find.
When the US leg is the cause: what to change in receiving and handling
- Receiving check at the dock. Record carton condition, any temperature record available, mill date, and pull the two retains. Five minutes per shipment.
- FEFO by mill date, not by receiving date. Lots do not always arrive in the order they were milled. Rotating by arrival date can put the oldest powder at the back of the rack.
- Move the matcha off the hot part of the building. Cold or chilled storage for reserve stock, and at minimum away from dock doors, exterior walls, and anything that generates heat.
- Write a rule for opened bags. Decant into a day container, reseal the mother bag immediately, purge headspace as best you can, and set a maximum number of days an opened bag stays in service. Whatever number you choose, write it down and make it visible at the station — an unwritten rule is not a rule across multiple sites.
- Chilled powder needs to come to room temperature sealed before opening, or condensation will form on it. This is the single most common self-inflicted damage in cold-storage programs.
- Re-score at the store, not just at the DC. If the DC retain is fine and complaints keep coming from one location, the problem is that location's handling.
And if it is simply seasonal
Sometimes the answer is that the crop changed, the supplier disclosed it, the new lot is within the agreed tolerance, and the drink is still within your menu's intent. Accepting that is legitimate — but accept it explicitly: approve and date a new standard, retain it, and tell the stores what changed and why. An undocumented acceptance becomes next quarter's complaint, because the reference everyone compares against quietly moved.
Running the whole thing once
The practical sequence for your next complaint: pull the sealed retain of the complained lot and the retained approved standard; score all four cells in one session with two people; read the conclusion off the grid; then pull either the freight and DC records or the supplier's lot documentation depending on which way the grid points. Keep the scoring card and the conclusion with the lot file. After three or four lots you will have something more valuable than any single diagnosis — a record of how your particular lane and your particular DC behave across the year.
If the comparison points upstream, the useful next move is a reference point: matcha from a supplier who states harvest period, mill date, blend window, and particle size on the spec sheet, so your retain comparison has something concrete to sit beside. Request samples with the lot details stated and run them through the same four-cell grid you just built.
