Amazon dropshipping product finder: margin, weight, demand and competition

Give it a category, a bestseller list or a search term. Get back the four numbers that decide whether a product is worth selling: what it sells for, what it weighs, how many people bought it last month, and how many sellers are already on it.

Twenty-six fields per product, nineteen marketplaces, and a price that comes from Amazon's own buy box data rather than from anywhere else on the page.

The four numbers, and why most tools give you two

Product research tools are good at price and rating. Those are on the search page, so they are cheap to collect and everybody has them. They are also the two that tell you least.

What it weighs. weight and dimensions. This is the number that decides whether a product is a business. A twelve dollar item that weighs four pounds is a different proposition from a twelve dollar item that weighs four ounces, and the difference is your entire margin. It is on the product page and nowhere near the search page.

How much actually sells. boughtInPastMonth is Amazon's own figure, the one printed as "10K+ bought in past month". estimatedMonthlySales and estimatedMonthlyRevenue turn it into money. Rank fifteen in a category where everyone sells two hundred a month is a different business from rank two where the leader sells four.

How crowded it is. otherSellersCount is how many sellers are already on the listing. Four hundred reviews at 4.8 stars with nine sellers is a wall. The same product with one seller is an opening.

What the real price picture is. price with listPrice, savingsAmount, discountPercent and pricePerUnit. A product permanently at forty percent off has a list price that is decoration, and the per-unit price is what a customer is really comparing.

Every field it returns

Twenty-six per product, chosen for one question: can I make money on this.

Price and offer

Field	What it holds
price	Current price from Amazon's own buy box data (this is what you are charged for)
listPrice	The struck-through price, when Amazon shows one
savingsAmount	How much is off in money
discountPercent	How much is off as a percentage
currency	Three-letter code, from the page itself
pricePerUnit	Price per count, ounce, litre, whatever Amazon uses
otherSellersCount	How many other sellers offer it

Demand and rank

Field	What it holds
boughtInPastMonth	Units bought last month as a number
estimatedMonthlySales	Units, derived from the line above
estimatedMonthlyRevenue	Units times price
bestsellerRanks	Every rank with its category, not just the first
rating	Average stars, read in every shop language
reviewCount	How many ratings, as a number

Stock and delivery

Field	What it holds
inStock	True, false, or null when Amazon does not say
isPrime	Prime delivery on this offer

Identity and catalogue

Field	What it holds
asin	The product id
title	Full product title, shortened only past 1000 characters
brand	Brand name
marketplace	Which shop this row came from
url	The page this was read from
scrapedAt	Timestamp of the read

Description, media and variants

Field	What it holds
image	Main image at the largest size Amazon offers
category	The most specific category

Specifications, pulled apart

Field	What it holds
weight	Weight with its unit
dimensions	Package or product dimensions
unitCount	How many units in the pack

A real row

asin                     B0BZYCJK89
title                    Owala FreeSip Stainless Steel Water Bottle 24 oz
brand                    Owala
price                    27.99
currency                 USD
listPrice                34.99
discountPercent          20
pricePerUnit             null
otherSellersCount        3
boughtInPastMonth        30000
estimatedMonthlySales    30000
estimatedMonthlyRevenue  839700
bestsellerRanks          #1 in Water Bottles, #12 in Sports Water Bottles
rating                   4.7
reviewCount              89412
weight                   1.02 pounds
dimensions               3.5 x 3.5 x 10 inches
unitCount                1
isPrime                  true
inStock                  true

Thirty thousand units a month at twenty-eight dollars, one pound, three other sellers, and eighty-nine thousand reviews. Three of those numbers say yes and one says the door is closed. That is the whole job of this Actor: putting them on one row so you can see both at once.

How to read the estimates honestly

boughtInPastMonth is Amazon's own number and it is rounded: "10K+" becomes 10000, "50+" becomes 50. It only appears on listings with enough recent volume, which is roughly the top of a category. An empty value means Amazon did not print it, not that nothing sold.

estimatedMonthlySales is derived from that figure and inherits its roundness. estimatedMonthlyRevenue is that number times the current price, which assumes every unit sold at today's price. On a product that was on deal last week, it is an overestimate.

Neither is a forecast and neither is a substitute for Amazon's own reports. They are there so a category sorted by revenue puts the interesting rows near the top, which is what you actually want from a research pass.

Reading boughtInPastMonth correctly took a repair worth mentioning: Amazon splits the figure across two separate page elements, so an earlier version found it on one product in six. It now finds it wherever Amazon prints it.

It will not guess a price

Amazon puts up to nine price-shaped numbers on a product page. Measured on this engine: prices read from the markup were correct zero times out of six, prices from the buy box data Amazon itself uses were correct nine out of nine.

So the price comes from that data and nowhere else. When it is absent you get no price rather than a wrong one, and the row still arrives with the weight, the rank and the demand figure.

On a research pass that matters twice over: a wrong price does not just mislead you about one product, it corrupts estimatedMonthlyRevenue and moves that product up your sorted list.

You are not charged for a row without a price.

What to put in

A category or bestseller link. This is the main way in. Paste https://www.amazon.com/gp/bestsellers/kitchen or any category link and it walks the list.

A search term. insulated water bottle becomes an Amazon search and every result gets the full read.

A filtered search URL. Build the search in your browser with the price band and the filters you want, paste the address, and this Actor continues from there. That is usually how a real research pass starts: under twenty-five dollars, four stars and up, Prime only.

ASINs, when you already have a shortlist and want the full picture on it.

Settings worth knowing about

Setting	What it does
marketplace	One of nineteen. Read the marketplace note below before researching a European shop.
proxyCountry	Leave empty and it follows the marketplace.
maxItems	Hard ceiling on delivered products. Zero means no ceiling. Worth setting on a research pass so a broad search cannot run away with your budget.
maxItemsPerList	How many products to take from each search or category. Prefilled at ten so your first click is quick and cheap; raise it for a real pass.
maxListPages	How deep to walk each list.
lanes	How many products are read at once. Eight by default. Sixteen roughly halves the run time and costs a few prices.
maxAttempts	Attempts per product, each from a different address. Twelve by default.
residentialFallback	Off by default. On product pages it buys almost nothing: measured 2026-08-26, it cost roughly six times as much and returned two more products out of sixty. On the search and category pages this Actor starts from, it is a different story. Measured 2026-08-30, during hours when Amazon refused every listing page with a two kilobyte stub, twenty datacenter retries returned nothing and residential returned a full page in eight attempts. Turn it on if a run comes back empty.
economical	Off by default. It would hang up before the weight arrives, which is the field you came here for.

How it works, in plain terms

A research pass is two jobs. First it walks the list you gave it, which is a search page or a category page, and collects the ASINs. Then it opens every one of those products and reads it properly.

That second half is why the weight is here and not on the search page. It is also why this takes minutes rather than seconds: sixty products is sixty page reads, each of which Amazon refuses roughly six times in ten on a fresh address.

So every product gets up to twelve attempts, each from a different exit address, and the same address is never used twice in a row. Addresses that work are remembered for the run and tried first. When refusals climb the runner slows down. Anything still refused goes into a second pass at half speed.

The listing page gets the same treatment, and it needs it: a refused listing page costs you every product inside it. On 26 August 2026 a category run came back with zero products because Amazon answered the bestseller link with four kilobytes and no error code, and the engine read that as an empty category. It now recognises a refusal without an error code and retries it.

Reading the run log

Done: 16 charged (+3 delivered without a price, free), 1 refused by Amazon, 0 other. 1055 KB per charged result, 223 attempts in total.
Exit addresses: 184 used, 2 proven good, 14 burned.
Attempts by outcome: {"refused by Amazon":146,"page did not carry the field this Actor needs":56,"http_500":3,"ok":16,"network error":2}

charged is products that came back with a price. The bracketed number arrived complete without one and cost you nothing. On a broad search that second number is normally a handful: searches turn up subscription pages, apps and listings Amazon does not price.

If a run ever delivers rows and charges for none of them, the log says so in capitals at the bottom.

Reliability, measured

Three consecutive runs of twelve products on 26 August 2026, at the default settings:

run 1   12 of 12 delivered, 7 with a price
run 2   10 of 12 delivered, 6 with a price
run 3   10 of 12 delivered, 8 with a price

And one run of sixty products the same day: 28 delivered, 23 with a price, in 6.0 minutes.

One thing is worth reading out of that, and it is not the one you would expect. The sixty product run delivered 28, which sounds worse than the 32 of 36 above until you notice what it was asked for: a category page, and Amazon only offered 28 products on it that day. A list Actor is bounded by what the list holds, so "delivered" here means "everything that was there", not "everything we wanted".

The second is the residential switch, and the honest version is less dramatic than it sounds. Measured 26 August on the same sixty products, twice: with the switch off, fifty-one came back with the field this Actor charges on, for $0.0205 of platform usage. With it on, fifty-three came back, for $0.1192. Two more products for roughly six times the cost.

That is why it is off by default. It is in your input form, so if a marketplace is refusing almost everything you can turn it on and pay for the difference.

What you can build with it

A sourcing shortlist. Walk a category, sort by revenue, filter to products under a pound with fewer than five sellers, and you have a week of work narrowed to an afternoon.

A margin model. price, weight and dimensions are the three inputs every freight and fulfilment calculator asks for. Export to a spreadsheet and the model is one formula.

A saturation map. otherSellersCount across a whole category tells you where the crowd already is. The interesting products are the ones with demand and without a crowd.

A competitor's catalogue. Search a brand name and you have every product they list, with volume and rank on each.

A repeat pass. Run the same category monthly into one dataset and you can see which products are gaining sellers, which is the earliest sign a niche is closing.

Working it into your stack

Every run writes to a dataset you can download as CSV, JSON, Excel or XML, or pull through the API. For research the spreadsheet export is usually the point: the output is flat apart from bestsellerRanks, which is a list.

Schedules take a cron expression. A monthly pass on the same categories, into one named dataset, turns a snapshot into a trend without any extra work.

Marketplaces

Nineteen, from amazon.com to amazon.co.jp, read in their own language.

Prices do not travel everywhere, and for this Actor that matters more than for most, because price feeds the revenue estimate. Measured on 2026-08-24:

Marketplace	Products priced
amazon.com	57 of 60
amazon.ca	5 of 5
amazon.co.uk	5 of 6
amazon.fr	4 of 5
amazon.it	1 of 5
amazon.es	1 of 5
amazon.de	0 of 8
amazon.com.au	0 of 5

Weight, rank, ratings and demand still arrive on all of them, and rows without a price are free. If you are researching a European marketplace, run twenty products first and see what you get.

Pricing

Four dollars per thousand products with a price, plus three cents to start a run.

A research pass over five hundred products costs about two dollars. You are not charged for rows without a price or for products that never arrived, and platform usage is included, so there is no separate compute or proxy bill.

The rest of this family

Eleven Actors on one engine. After a research pass:

• Amazon Niche Research when the question is whether a whole category is worth entering rather than which product in it to pick.
• Amazon Price Tracker on the shortlist, to watch what the competition does before you commit.
• Amazon Product Specs when you need the full catalogue detail to brief a manufacturer.
• Amazon Variants Scraper because a product that looks like one listing is often fourteen.
• Amazon Bestsellers Scraper for the same depth with all sixty-four fields.

Questions people ask

Why is boughtInPastMonth often empty. Amazon only prints it on listings with enough recent volume. Empty means it did not print, not that nothing sold.

Are the sales estimates reliable. They are Amazon's own rounded figure times the current price. Good enough to sort a category by, not good enough to build a forecast on. Nothing public is.

Why is weight empty on some products. Amazon prints it in a detail table that not every listing carries, and electronics carry it least. Every Actor in this family reads it the same way, so the specs Actor will not find it either; what it adds is the raw specification table, where you can see what Amazon did print.

Can I filter by price or rating. Not in the input, but you can build the filter into a search URL in your browser and paste that. Amazon's own filters are better than anything this Actor could reimplement.

How many products should a research pass cover. Two to five hundred is typical for one category. There is no built-in ceiling; on the measured run 28 products took 6.0 minutes, so sixty is around 13.

What it does not collect

No review text, no reviewer names, no customer questions, no seller names. You get the aggregate rating and the count, which are facts about a product rather than about a person.

Legal note

This Actor reads public pages and takes product facts: prices, weights, ranks, demand figures, offer counts. Business information about an item, not information about a person.

You are responsible for what you do with the output. Check Amazon's terms for your own use case.