Question 1

Which store is cheapest per unit?

Accepted Answer

price per unit, because the cheapest row is half the pack find <product> with litescrape google shopping on monid and open the product page endpoint. read pack size off the OFFER titles rather than the search results, and put any offer whose title carries no size into a third pile rather than dropping it: one merchant's title omits the size entirely, so a strict filter can throw away the cheapest matching pack. check that pile before you name a winner. most search titles carry no size at all, and filtering there throws away the right row and keeps the wrong battery the spread: $4.12 between the cheapest and the dearest shelf, which is 27.7%, on one identical pack in one country on one day what it cost: $0.0003 for fifteen merchant offers with direct store links, which is $0.00002 an offer per unit instead: $0.929 a cell at the bottom against $1.187 at the top, which is the number that actually decides the purchase the two that had no price: one big grocer's own page carried the product markup with its offers field set to null; one big hardware chain does not stock this size at all where not to filter: on the search list 12 of 20 titles carried no pack size at all, the correct top result among them, and one row that survived the filter was a different battery size entirely

Question 2

What does this product sell for on TikTok Shop?

Accepted Answer

what a product costs there against everywhere else get the tiktok shop products for category <id> with tikhub on monid several times, pool the results and de-duplicate, then sort by units sold yourself. there is no bestseller endpoint here and the feed is not just unordered, it is reshuffled per call: two identical requests came back with no products in common there is no ranking: nothing in this aisle returns a sales rank, and in our feed the top seller by units came back last on the page worse than unordered: three calls to the same category, two of them byte-identical requests, returned zero products in common; so pool several calls rather than paging one, because there is no stable sequence to page through two documented things that are not true: the list arrives under productList rather than the documented products, and a page is fifteen rows rather than the documented twenty, so the documented offset of twenty skips five rows every time what a month costs: $0.32 to refresh a whole category's prices and unit counts every day for thirty days, at seven calls a day listed stable, mostly not: the other provider's search endpoint answered 404 twice, and the category endpoint itself answered 400 on seven of eleven attempts while discover marked it stable; every one of those billed $0.0000

Question 3

What do the reviews really say about this Amazon product?

Accepted Answer

the rating spread is already inside the reviews pull the one-star reviews for <asin> with apify on monid, filtered to one star, and leave the page count alone unless you need more: it defaults to one page of ten and asking for two doubles the bill. this endpoint bills per review whatever its schema says two pages is not a cap: the default is one page of ten reviews at $0.0135; asking for two pages billed $0.027, exactly double, so the page-count clause spends money rather than saving it what it would have cost: $2.72 to pull all 2,014 reviews on this one product the filter is the point: one-star, filtered, is not a sample: it is the whole complaint set, and one page of it is the default the schema is wrong about the price: inspect says one result is returned per input query and that there is no result limit, and both are false: one query returned twenty results and twenty billed units. Trust the run's cost over the schema here free extra: every review carries the product-wide rating spread, so that job needs no second call

Channel3

Try it with your agents

How-to guides