Amazon Storefront Scanner: Extract Every Product, See What Changed

Every Amazon seller has done this at least once: you find a competitor who is clearly doing something right, you click through to their storefront, and you start scrolling. Page 3. Page 9. Page 17. Then page 21 tells you there are no more results — and you know, with certainty, that this seller has far more than 300 products.

That is the entire problem with storefront research. The data is public. Amazon just won't hand you more than a sliver of it, and it won't tell you when that sliver changes.

This article covers both halves of the job: how to extract a complete Amazon storefront — including stores with 100,000+ listings — and how to re-scan it later so you see precisely what changed. New products. Removed products. Price moves. Stock-outs. Buy Box flips.

Why an Amazon storefront only shows you about 300 products

A seller's storefront is just an Amazon search result with a me= filter. It inherits the same hard limit every other search result has: pagination stops at 20 pages.

We measured this against a live storefront while writing this article. On 9 July 2026 we walked amazon.com/s?me=A294P4X9EWVXLJ — AnkerDirect's storefront. Amazon reported last_page_number: 20 and served 16 products per page. Page 20 came back with just two products, labelled results 305–306. Page 21 returned nothing at all.

So the entire storefront, through the front door, is 306 products — and Amazon will not give you a 307th, no matter how large the catalog behind it actually is. (Per-page counts differ by layout; a keyword search page serves far more. What doesn't change is the 20-page floor under all of it.)

Worse, the 306 you can see aren't a trustworthy sample:

• The default "Featured" sort reshuffles. Amazon's own help page says Featured ranking factors in purchase frequency, availability, delivery speed and customer preferences. Load page 4 twice and you can get different products. Pages overlap and skip.
• Out-of-stock products are hidden by default. Amazon suppresses them unless you explicitly ask for them with the Include Out of Stock filter. A competitor's stock-outs are exactly what you want to see, and they're the first thing the page removes.
• Variation siblings collapse. A shoe in nine colours frequently appears as one representative tile, so a "306-product" store is often several times bigger even inside its visible window.
• Even the facets move. We asked that same storefront for its department list three times in five minutes and got back eight departments, then ten, then nine. The navigation you'd use to explore the store by hand isn't stable either.
• There's no snapshot. Even if you copy those 306 ASINs into a spreadsheet, you have nothing structured to compare against next month.

Put a number on it: a storefront with 100,000 listings shows you 0.3% of itself through the front door.

How to see all products in an Amazon storefront

You cannot page past the wall. Reaching the rest of the catalog means not walking in through the front door at all.

That is the job ASINSpotlight's Cloud Scanner does for you. Paste a storefront URL or a seller ID, and it returns that seller's complete catalog — whether the store has 300 products or 100,000, the workflow is identical. Getting there is harder than it looks, and the difficulties are worth knowing even if you never run a scan with us — they explain why a weekend scraping script hands back 60% of a store and never tells you which 60%.

• Completeness isn't observable from the inside. Amazon never tells you how many products a seller has. A crawler that stops when it runs out of pages cannot know whether it finished or was cut off — and neither can you. Cloud Scanner tracks its own coverage, and when a pass falls short it says so rather than presenting a partial list as the catalog.
• The catalog moves while you read it. Walking a large storefront takes time, and sellers add and retire listings while you're walking. A scan has to notice that the ground shifted and go back for what appeared, instead of quietly truncating.
• Out-of-stock products have to be asked for. They're excluded by default, and they're usually the ones carrying the signal you came for.
• Variation siblings have to be expanded. The nine-colour shoe that renders as one tile is nine ASINs, nine prices, and nine independent stock states.
• You should be billed for products, not requests. Everything above means fetching far more pages than the store has products. That is our problem, not yours — Cloud Scanner charges per unique product row delivered, and never for the machinery behind it.

Every product that comes back from a list pass carries: ASIN, title, image, price, shipping, Prime status, in-stock status, rating, review count, bought-in-past-month, coupon, and variation swatch data. Out-of-stock products included. You can stop there, or you can go deeper — more on that when we get to cost.

Storefront scans accept a URL from any Amazon marketplace ASINSpotlight supports: nineteen of them, from amazon.com to amazon.com.br.

Storefront stalking, done properly: the rescan

Pulling a competitor's catalog once is a snapshot. It tells you what they sell. It does not tell you what they're doing.

The interesting questions are all differential:

• Which products did they add since last month?
• What did they quietly stop selling?
• Which of their listings just dropped 15% in price?
• What ran out of stock — and therefore, what should I be sourcing right now?
• Where did they lose the Buy Box, and to whom?

So a scan in Cloud Scanner isn't a frozen result. It's a tracked list of ASINs you can re-run whenever you want. Each run records a full diff against the run before it.

The table always shows the latest known value for every product. History arrives as an overlay: a run strip along the top, a change digest that filters, and delta chips inside individual cells — the previous value struck through, the new one beside it, and the percentage move.

The eight change signals

Every rescan computes exactly eight aggregate change classes:

Signal	What it means
new	Appeared in the latest list pass; wasn't in the previous one
removed	No longer returned by the storefront — but possibly still live on Amazon
delisted	The product page itself is gone from Amazon (a "dog page")
price changes	Listing price moved between runs
went out of stock	In stock last run, out of stock now
back in stock	The reverse
Buy Box changes	The Buy Box holder flipped, or its price moved
BSR changes	Best Sellers Rank moved (details layer)

Each chip is a filter. Click 9 went out of stock and the table narrows to those nine products.

Two design details matter more than they look:

"Removed" and "delisted" are different things, and the distinction is worth money. A removed product left the storefront but its Amazon page is fine — the seller chose to stop listing it, which is a sourcing signal. A delisted product's page is gone entirely — the listing died. Tools that lump these together tell you nothing.

Membership diffs are suppressed when they'd lie. If either list pass didn't run to full coverage, "appeared" and "removed" counts would read a partially-scanned catalog as a mass exodus. In that case the scanner hides those two counts rather than showing you a confident wrong number. Price, stock and delisting diffs are status-based and always reported.

Layers refresh independently

A storefront scan has three data layers, and each carries its own "as of run #N" freshness. You can refresh them separately:

• Search listing — membership, price, rating, reviews, stock. Cheap.
• Product details — BSR, brand, category, Buy Box price, seller count, bullets, dimensions, UPC/EAN.
• Seller offers — every seller on the listing, with price, shipping, stock quantity, FBA/FBM and seller rating.

Refresh only the listing layer and you get membership and price movement for a fraction of a cent per product. Refresh offers too and you get the competitive picture underneath each listing.

Who joined, who left, who took the Buy Box

Expand any row and the offers panel shows the seller-level diff against the previous offers fetch: sellers who joined the listing, sellers who left it, sellers who switched from FBM to FBA, price and shipping moves, stock-quantity changes, and whether the Buy Box was held or lost.

For brand owners, this is MAP monitoring that arrives by itself. For arbitrage sellers, it's the difference between "this ASIN looks good" and "this ASIN just went from six sellers to two, and the Buy Box price is up 14%."

What it costs to scan and rescan an Amazon storefront

Cloud Scanner bills per delivered record, in credits, from a one-time pack. There is no subscription and no per-store fee.

Layer	Credit cost
Search listing — storefront, brand, ASIN list	1 credit per 10 products
Search listing — keyword, category, best-sellers	1 credit per 20 products
Product details	1 credit per product
Seller offers	1 credit per offer page (≈10 offers)

And the packs:

Pack	Credits	Price	Per 1,000	Valid for
Starter	10,000	$12	$1.20	30 days
Small	30,000	$29	$0.97	30 days
Standard	100,000	$79	$0.79	60 days
Large	300,000	$199	$0.66	90 days
Bulk	1,000,000	$650	$0.65	180 days

Work the arithmetic on a shallow storefront scan and it's startling. At 10 products per credit, 100,000 credits covers a million delivered product rows. The $79 pack, in other words, buys:

Storefront size	Credits per full list pass	Cost inside the $79 pack
1,000 products	100	~$0.08
10,000 products	1,000	~$0.79
100,000 products	10,000	~$7.90

Scanning a 100,000-listing storefront costs roughly $7.90. Rescanning it every week for the pack's entire two-month life uses eight runs — 80,000 credits — and still fits inside that single $79 pack.

Before every rescan you see the estimate, per layer, and decide what you're paying for. Refreshing the listing layer on a 1,284-product storefront costs 129 credits — about ten cents.

Three honest caveats, because pricing pages tend to omit them:

• Details and offers are 1 credit each, not 1-per-10. Enriching all 100,000 products of that storefront with BSR and Buy Box data would consume the whole 100,000-credit pack. Depth costs; breadth is nearly free. Most storefront-monitoring workflows only need depth on the rows the digest flagged.
• Variation children count as delivered rows. Storefront scans expand sampled variation siblings, so a store that displays 100,000 tiles may deliver more than 100,000 rows. Treat 10,000 credits as a floor, not a quote.
• Credits expire and are forfeited. The $79 pack is valid for 60 days from purchase. The $7.90 figure is the amortised rate inside that pack — a standalone 10,000-credit purchase is the $12 Starter pack.

You can queue up to 100 storefronts in a single scan job.

Who actually uses an Amazon storefront scanner

Reverse sourcing / online arbitrage. Find sellers whose model matches yours, extract their whole catalog, then let the weekly rescan hand you their new ASINs. This is the classic "storefront stalking" workflow, and the new-product digest is the payoff.

Brand protection and MAP enforcement. Scan your own brand's ASINs, refresh offers, and read the seller-level diff: who joined your listings, who's underpricing, who switched to FBA to fight for the Buy Box.

Category and competitor intelligence. Track ten competitors' full catalogs. The removed-products list is the most under-used signal in Amazon research — it's every product a competitor tried and abandoned, which is a free read on what didn't work.

Restock and stock-out sniping. "Went out of stock" on a competitor's listing is a demand signal with a clock on it.

Storefront Stalker Pro and the other alternatives

The dedicated storefront-stalking category is small. Storefront Stalker Pro is the best-known entrant, and it's genuinely good at the job it picked. In its own words, it "allows you to automate the process of reverse sourcing storefronts… You will be alerted every time new ASINs are added to the storefronts you're stalking."

That sentence is also the boundary. It is a new-ASIN alerting tool, priced by how many stores you watch, and it refreshes on a fixed 4-hour cadence — which is a real advantage if new-product alerts are all you want, since nothing is on-demand about it. Its published tiers, as of July 2026, are Entry $15/mo (3 stores), Basic $40/mo (10), Professional $70/mo (20) and Elite $140/mo (40), with a 7-day trial and 20% off annual.

	Storefront Stalker Pro	Cloud Scanner
Core job	Alerts when a tracked store adds an ASIN	Extracts the full catalog, diffs any run against the previous one
Priced by	Storefronts tracked	Delivered product records
Entry cost	$15/month	$12 one-time (10,000 credits)
Refresh	Automatic, every 4 hours	On demand
Change signals	New ASINs	New, removed, delisted, price, out of stock, back in stock, Buy Box, BSR, plus per-seller offer diffs
Full-catalog export	Not the stated purpose	CSV / Excel, whole store or filtered

Elsewhere in the market, the closest thing to a true competitor is Seller Assistant's Seller Spy: paste a storefront URL, it pulls the seller's full catalog every day and gives you a changes view of products added and removed. It's a $9.99/month add-on, and if a daily added/removed feed is all you need, it's a fair deal. What it doesn't give you is the rest of the diff — price moves, stock-outs, Buy Box flips, and the seller-by-seller offer changes underneath each listing.

Further out: Keepa's seller lookup returns a stored asinList of up to 100,000 ASINs per seller, though a single live storefront query refreshes only about 2,400 of them. SmartScout exposes seller catalogs from its own warehouse, with the Sellers Database starting at $97/mo. Chrome extensions like SellerAmp's ASIN Grabber or Helium 10's Xray will annotate or one-shot-grab a storefront page, but none of them re-scan and diff it over time.

And if you'd rather build it yourself: on published July 2026 pricing, 100,000 Amazon product records runs about $150 on Bright Data pay-as-you-go ($1.50 per 1,000 records), or roughly $50 in usage on Oxylabs, whose Amazon rate starts at $0.50 per 1,000 results on a $49/month plan. That's the raw fetching bill — before you solve any of the problems above, and before you build anything that diffs two runs. If raw data is genuinely what you want, that's what the ASINSpotlight API is for.

Frequently asked questions

How many products can an Amazon storefront actually show? Twenty pages, and no more. When we walked a real storefront in July 2026, amazon.com served 16 products per page and stopped at result 306. Seeing the rest of the catalog takes a tool that doesn't try to page through the storefront at all.

Can you export an Amazon seller's entire product list to CSV? Yes. Cloud Scanner exports the whole scan or just the filtered view, as CSV or Excel, with per-field column selection and an option to emit one row per seller offer.

Does the scan include out-of-stock products? Yes. Amazon hides them from storefront pages by default; the scanner asks for them explicitly.

Can it handle a storefront with 100,000+ listings? That's the case Cloud Scanner was built for. Cost scales linearly and gently: at 10 products per credit, a 100,000-product pass is 10,000 credits.

How do I see what a competitor removed from their store? Run the scan, run it again later, then click the removed chip in the change digest. Note the distinction from delisted: removed means they stopped selling it, delisted means Amazon's page is gone.

Is storefront stalking against Amazon's Terms of Service? Two separate questions get tangled here. On the law: US courts have held that automated collection of publicly accessible pages is not "unauthorized access" under the Computer Fraud and Abuse Act — the Ninth Circuit in hiQ Labs v. LinkedIn, reaffirmed in 2022 after the Supreme Court narrowed the CFAA in Van Buren. On the terms: Amazon's Conditions of Use grant only a limited licence to use the site and expressly exclude data mining and use of robots. Breaching a website's terms is a contract matter, not a crime, and Amazon's practical remedy is to block or rate-limit you. Cloud Scanner reads only public pages and never touches an account. None of this is legal advice — if the answer matters to your business, ask a lawyer.

Does a rescan cost the same as the first scan? At the listing layer, yes — it's the same per-product rate. But you choose the layers. A membership-and-price refresh of a 1,284-product store is 129 credits; re-fetching every product's details and offers on top of that is 2,697.

---

Storefront research has always been an exercise in guessing from a 300-product sample and hoping the sample was representative. It doesn't have to be.

Cloud Scanner is in beta and available now. See how it works, or run your first scan — credit packs start at $12, and there's no subscription to cancel.