Draft v0.1 — 2026-06-19 | Status: pending analyst review | Layer 1 piece #2 Sequel to:
01_the_hidden_inventory_clock.mdTarget length: ~1900 words. Tone: lucid, plain language, business-friendly. Audience: broad (Amazon sellers, sourcing operators, importers, generalists). No brand-specific claims. SEO target: “amazon inventory recovery signal”, “importyeti amazon sellers”, “amazon shipment data sourcing”, “bsr recovery prediction”
In the previous piece, I wrote about how Amazon BSR works like a depletion gauge — not a popularity score. When a product’s rank crashes, it usually means inventory ran out, not that demand collapsed.
That framing leads to an obvious follow-up question: if inventory depletion shows up in BSR, does inventory replenishment show up in anything?
It does. And it shows up three to six weeks before BSR recovers.
That lead time is the point of this piece.
A product's inventory is like a water tank with two flows: sales drain it, and new shipments fill it. BSR measures the drain rate, while the fill rate appears separately in public US Customs records — shipper, consignee, ports, weight, containers, and shipment date — making both flows of one system measurable.
Think of any Amazon product as a water tank. The tank is inventory at Amazon’s fulfillment centers. One pipe drains it — sales. Another pipe fills it — new shipments arriving from the manufacturer or supplier.
BSR measures the drain rate. When the drain rate is fast relative to other products in the category, BSR rank is low (good). When the drain slows — whether because demand dropped or because the tank is running empty — BSR rank rises (bad).
But here’s the thing: the fill pipe is also measurable. It’s just measured somewhere else entirely.
When a Chinese manufacturer (or any overseas supplier) ships goods to Amazon sellers in the United States, those shipments clear US Customs. Customs records — technically called Automated Export System / Import Entry data — are public information under US trade law. The data covers the shipper name, consignee, port of loading, port of entry, weight, number of containers, and shipment date.
Services like ImportYeti aggregate these records and make them searchable by brand, company, or HS code.
The result: you can watch the fill pipe in real time.
When a seller or importer is restocking a product that sold well, a few things show up in the shipment records: Cadence : A healthy brand restocking a healthy product ships on a regular cycle. Roughly every 30, 60, or 90 days, depending on how fast the product sells and how much buffer the brand keeps.
When a seller or importer is restocking a product that sold well, a few things show up in the shipment records:
Cadence: A healthy brand restocking a healthy product ships on a regular cycle. Roughly every 30, 60, or 90 days, depending on how fast the product sells and how much buffer the brand keeps. Irregular cadence is one of the first signs of trouble.
Volume: The number of cartons or the container size tells you roughly how many units are inbound. If a brand typically ships a 20-foot container and suddenly ships a 40-foot container, they’re either anticipating higher demand or catching up from a stockout.
Timing relative to the BSR crash: This is the signal that matters most for spotting recoveries. When a product’s BSR crashes — the rank goes from 150 to 12,000 in two weeks — you can check whether a new shipment departed in the days or weeks before the crash. If there was a departure 30-45 days before the crash, the inbound goods are probably already on the water. They’ll hit Amazon’s warehouses in three to six weeks depending on origin port and freight routing.
Ocean freight from major Chinese manufacturing hubs (Yiwu, Yongkang, Ningbo, Guangzhou) to the US East Coast takes roughly 25 to 35 days in transit. From departure to being checked into an Amazon fulfillment center, add another one to two weeks for customs clearance, domestic freight, and receiving.
Ocean freight from major Chinese manufacturing hubs (Yiwu, Yongkang, Ningbo, Guangzhou) to the US East Coast takes roughly 25 to 35 days in transit. From departure to being checked into an Amazon fulfillment center, add another one to two weeks for customs clearance, domestic freight, and receiving.
That’s a 6-to-10 week lag between when a factory ships a container and when that inventory shows up as available units on Amazon.
BSR reacts to sales velocity. Sales velocity can’t recover until inventory is available. So if you know a container departed Ningbo on a specific date, and you know the approximate transit time, you can estimate when BSR will start moving back toward the product’s historical range.
This is the structural gap between BSR (which is real-time) and replenishment data (which is 30-day-lagged but visible now).
Not every BSR crash has an inbound shipment behind it. Some crashes are genuine demand collapses — the product got a wave of negative reviews, a competitor undercut it on price, or a category trend reversed. In those cases, looking at shipment records tells you the same thing as BSR: the brand isn’t replenishing.
Not every BSR crash has an inbound shipment behind it. Some crashes are genuine demand collapses — the product got a wave of negative reviews, a competitor undercut it on price, or a category trend reversed. In those cases, looking at shipment records tells you the same thing as BSR: the brand isn’t replenishing.
But “brand isn’t replenishing” is actually useful information in its own right. Here’s what it tells you by case:
No recent shipments, no sign of an inbound: The brand may be struggling. Stockouts with no visible restock are often the first signal of a brand winding down a SKU, clearing excess inventory, or facing cash flow problems. The product may not recover.
No recent shipments, but there was a regular cadence before: The cadence break is meaningful. Something changed — supplier relationship, demand outlook, or the brand is waiting for a revised product. This is worth watching. The cadence break happened for a reason.
Shipment arrived, but BSR hasn’t moved: This usually means the goods haven’t been processed at the fulfillment center yet, or they arrived but the seller is FBM (fulfilling from their own warehouse) rather than FBA (using Amazon’s warehouses). FBM restocks don’t necessarily show up in Amazon’s in-stock data the same way.
New entity appearing as consignee or shipper: Sometimes a brand switches suppliers, consolidators, or 3PLs (third-party logistics providers). A new entity name appearing in the shipping chain can mean a supply-side change, not a demand-side one. The product may be undergoing a sourcing transition — which could mean a temporary gap in stock even if demand is healthy.
Most Amazon sellers who use shipment data use it competitively — watching when a competitor’s supply is thin as an opportunity to gain sales rank. That’s one use. Sourcing operators and importers use it differently. They’re watching for structural signals: Category-level concentration : If you scan shipment records across all the major brands in a category, you can see who.
Most Amazon sellers who use shipment data use it competitively — watching when a competitor’s supply is thin as an opportunity to gain sales rank. That’s one use.
Sourcing operators and importers use it differently. They’re watching for structural signals:
Category-level concentration: If you scan shipment records across all the major brands in a category, you can see who manufactures where. When most brands in a small appliance category are sourcing from the same three factories in Yongkang or Shunde, that’s a concentration signal. A supply disruption — factory shutdown, freight capacity crunch, regulatory change — affects all of them simultaneously.
Supplier relationships: When a major brand switches from one manufacturer to another, it often shows up in the shipper field. If Brand A’s containers used to come from “Zhejiang XYZ Electronics Co.” and now they’re coming from “Guangdong ABC Hardware,” something changed in the supply relationship.
Quote leverage: If you’re negotiating with a factory and you can see that the factory is currently shipping to multiple competing brands, you know their capacity is constrained. If the factory appears idle in the shipment records for the past 60 days, you have more negotiating room.
None of this is secret intelligence. It’s public customs data. But almost nobody reads it systematically.
ImportYeti aggregates US import records, searchable by company name or brand. Its free tier returns recent shipment data for any entity, letting you find the US consignee for a brand as listed on Amazon. Panjiva and Flexport offer more structured, better-filtered data at enterprise prices. Direct customs records are public via FOIA requests, but commercial aggregators are more usable.
ImportYeti (importyeti.com) aggregates US import records and makes them searchable by company name or brand. The free tier gives you recent shipment data for any entity. You can search the brand name as it appears on Amazon listings and usually find the US consignee in the records.
Panjiva (owned by S&P Global) and Flexport’s market intelligence products offer more structured data with better filtering, but they’re priced for enterprise use.
Direct customs records: The US Customs import data is technically public and can be accessed through Freedom of Information Act requests, though the commercial aggregators are considerably more usable.
For a basic read on a single product, ImportYeti’s free search is usually enough to answer the question: “Has this brand shipped recently, and is there something inbound?”
It won't capture domestic manufacturers or air-freight brands, which appear inconsistently or not at all in ocean-freight records. It lags 30 to 60 days, so you see what happened, not what's happening now. And brand names don't cleanly match shipper or consignee names, requiring manual investigation or a product-to-supplier database to reconcile.
A few honest caveats:
Shipment data covers ocean freight. Many brands — especially domestic manufacturers or brands using air freight — won’t show up in US import records at all, or will show up inconsistently.
The data lags by 30 to 60 days depending on the provider and how frequently they update. You’re looking at what happened, not what’s happening right now.
Brand names don’t always match shipper or consignee names cleanly. “Keurig” as a brand might ship under multiple legal entities or through a consolidator whose name bears no resemblance to the consumer brand. Matching brand → import record requires some manual investigation or a product-to-supplier database.
And finally: shipment volume tells you units are inbound, not that they’ll sell. A large inbound shipment after a BSR crash can mean recovery is coming. It can also mean the brand miscalculated and is about to sit on excess inventory.
The signal is useful. It’s not infallible.
BSR and import cadence together reveal both sides of the stock-flow system: how fast demand drains inventory and when shipments refill it. Neither signal alone suffices. Combined, they let analysts read the supply chain in ways Amazon's listings don't surface—using only public data accessible to anyone with a few hours and a free account.
What BSR and import cadence together give you is a read on both sides of the stock-flow system: how fast the tank is draining and when it’s being filled. Neither signal alone is sufficient. BSR without replenishment data tells you about the symptom (the crash) but not the cause (demand vs. supply). Replenishment data without BSR context tells you goods are moving, but not whether demand exists to absorb them.
The combination is more useful than either alone — and it’s available to anyone with a few hours and a free account.
The analysts who use both signals routinely — watching BSR patterns across categories while scanning shipment records for restocking activity — are, in effect, reading the supply chain in a way that Amazon’s own product listings don’t surface.
That’s what this platform is built to do systematically. But the underlying data is public. You can start reading it yourself today.
The next piece covers how to read what happens when multiple brands in a category run out of stock at the same time — and what that synchronization tells you about the underlying supply structure.
This appendix lists the article's SEO target queries. Primary informational queries cover ImportYeti use, Amazon inventory replenishment tracking, BSR recovery after stockout, and import shipment data. Secondary long-tail queries address stock-recovery prediction and shipment cadence. The rationale: these low-competition queries claim the "BSR + supply chain signals" semantic space before competitors build topical authority.
Primary target queries (informational, Layer-1 educational): - “how to use importyeti for amazon selling” - “amazon inventory replenishment tracking” - “bsr recovery after stockout” - “amazon import shipment data sellers” - “importyeti explained for sellers”
Secondary (long-tail, lower volume, high fit): - “how to predict amazon stock recovery” - “us customs import data amazon brands” - “amazon shipment cadence analysis” - “reading bsr and import data together”
Keyword rationale: These are low-competition informational queries. Piece #1 (“hidden inventory clock”) targets BSR understanding; this piece targets the import/replenishment complement. Together they claim the “BSR + supply chain signals” semantic space before any direct competitor has built topical authority there.
Internal link target: Piece #1 (hidden inventory clock) → this piece → TBD piece #3 (category-level synchronization / the stress read)
Claim safety check (per L1 content rules spirit): - No capability/pricing/factory claims — PASS - No brand-specific product endorsements — PASS - No sourcing promises or OEM recommendations — PASS - ImportYeti/Panjiva referenced as public tools, not as our products — PASS - All quantitative claims (transit times, lag estimates) are industry-standard ranges, not proprietary data — PASS - CTA is soft: platform described in last section, no sales language — PASS