Falabella Marketplace Scraper: Products, Prices & Sellers

Falabella Marketplace Scraper turns Falabella (falabella.com) into a clean product data feed. One actor covers three country storefronts: Chile, Colombia and Peru, each its own catalogue with its own currency. Run keyword searches or paste your own product and listing links, get price, discount, rating and seller data on every row, add full product detail reads on demand, then export to JSON, CSV or Excel, or pull the results straight into your app through the API.

Why This Scraper?

• Three storefronts in one actor. Chile, Colombia and Peru, each with its own catalogue, currency and connection.
• Two ways in. Run keyword searches on one storefront, or paste your own links: product pages, category pages, keyword collections, brand pages or search result pages.
• Full listing data on every row. Price, was-price, discount, rating, review count, seller and the card's own badges (official seller, sponsored, best seller, installments, shipping flags).
• Optional detail reads. Switch on the description, full specification table, category path, stock state, loyalty points, warranty code and the seller's registered business details, only when you need them.
• Built for monitoring. Recurring updates label every record NEW, UPDATED, REAPPEARED or EXPIRED, so a scheduled run returns only what changed.
• Resume support. Continue an interrupted run from its run id without paying for the same rows twice.
• A refused run fails loudly. A run that is refused everywhere stops with a clear message; a genuinely empty search still succeeds, with zero rows and a visible note.

Use Cases

• Price monitoring teams: track price and discount changes on a set of products or categories over time, with recurring updates flagging exactly what moved.
• Marketplace analysts: compare the same categories and keywords across the Chilean, Colombian and Peruvian storefronts.
• Brand and seller managers: watch a competitor's catalogue, ratings and seller badges across all three markets.
• Deal and content sites: surface discounted, sponsored and best-seller items by category or keyword for a feed or newsletter.
• Data teams: assemble structured product catalogues (price, brand, ratings, specifications) for pricing or demand models.

Data You Get

Sample shape: values are illustrative placeholders, not from a live record.

Field	Example
recordType	"product"
id	"300000001"
skuId	"300000002"
name	"Mochila de muestra"
brand	"MARCA MUESTRA"
url	Link to the product page
imageUrl	Main product image
currency	"CLP" (also "COP", "PEN" by market)
price / priceDisplay	22990.0 / "$ 22.990"
oldPrice / discountPercent	27990.0 / 18
ratingAverage / ratingCount	4.6 / 128
sellerId / sellerName	"S00042" / "Tienda de muestra"
officialSeller / sponsored / bestSeller	true / false / true (card badges)
installments	Interest-free installment offers shown on the card (null when none)
shippingOptions	Shipping flags the card shows, for example home delivery
market / marketHost	"CL" / "www.falabella.com"
description / longDescription	Short and extended product text (detail reads)
specifications	Named attribute list, for example color and material (detail reads)
categories	Category path from the breadcrumb (detail reads)
outOfStock / stockUnits	Stock state and per-variant counts (detail reads)
sellerDetails / sellerVerified	Seller's registered business name, id and verification flag (detail reads)
cmrPoints / warrantyId	Loyalty points and warranty code (detail reads)
changeType / changedFields	"NEW" / list of field names that moved (recurring updates)
firstSeenAt / lastSeenAt	First and most recent sighting (recurring updates)

Every row also carries the search or link it came from and its position in that walk (sourceType, sourceId, sourceUrl, position), when it was read (scrapedAt), and, on detail reads, the full list of extra product images (images).

How to Use

1. Pick a mode: keyword search on one storefront, or your own product and listing links.
2. Choose the market for a search, or paste links. Each pasted link is read, labelled (country, currency, market host) and connected through the country of its own storefront, so Chile and Colombia links can share one run; the Market setting only applies to keyword searches. If you leave the links untouched in links mode, the run scrapes one example of each supported link shape (see the table below).
3. Turn on detail reads if you need the description, specifications, stock and seller details; leave it off for a cheaper run. Pasted product links are always read from their own product page, so they always carry the detail surcharge.
4. Set Max records to control run size and cost, then click Start.

Minimal keyword search:

{
  "mode": "search",
  "market": "cl",
  "queries": ["mochila"],
  "maxItems": 20
}

Ordered search with detail reads:

{
  "mode": "search",
  "market": "co",
  "queries": ["zapatillas"],
  "sort": "price-asc",
  "fetchDetails": true,
  "maxItems": 50
}

Your own links:

{
  "mode": "urls",
  "urls": ["https://www.falabella.com/falabella-cl/product/300000001/producto-de-muestra"],
  "maxItems": 10
}

Supported link shapes

Link shape	Example	Notes
Category page	https://www.falabella.com/falabella-cl/category/CATG10137/Zapateros-y-Organizadores-de-Zapatos	Walked forward from page 1
Category page, paginated	.../Zapateros-y-Organizadores-de-Zapatos?page=2	The walk starts at the page in the link
Keyword search with ordering	https://www.falabella.com/falabella-cl/search?Ntt=zapatilla&sortBy=derived.price.search%2Casc	The pasted ordering is kept
Keyword search, Colombia	https://www.falabella.com.co/falabella-co/search?Ntt=televisor	Labelled COP, connected through Colombia
Keyword collection	https://www.falabella.com/falabella-cl/collection/iphone-15	Walked forward
Search that the site redirects to a collection	https://www.falabella.com/falabella-cl/search?Ntt=iphone	The redirect is followed
Product page	https://www.falabella.com/falabella-cl/product/<id>/<slug>	Not verified: product pages are often answered with a verification page; the run logs a warning per link
Peru search	https://www.falabella.com.pe/falabella-pe/search?Ntt=televisor	Not verified; the run logs a warning per link

In keyword search mode, term is accepted as an alias of queries.

Daily price monitoring:

{
  "mode": "search",
  "market": "pe",
  "queries": ["camiseta"],
  "incrementalMode": true,
  "stateKey": "camiseta-daily",
  "maxItems": 200
}

Run it from your code

Python:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("abotapi/falabella-marketplace-scraper").call(
    run_input={"mode": "search", "market": "cl", "queries": ["mochila"], "maxItems": 20}
)
for product in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(product["name"], product["price"])

JavaScript:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });
const run = await client.actor('abotapi/falabella-marketplace-scraper').call({
  mode: 'search', market: 'cl', queries: ['mochila'], maxItems: 20,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Or connect it to Make, Zapier, n8n, Google Sheets or webhooks from the Integrations tab.

Max records and max pages

Max records is the single cap that stops the whole run (0 means no limit; the walk then runs to the catalogue's natural end). Max pages per source is a runaway guard, not a target: leave it 0 to read as deep as the source serves. Either way a single search or listing walk stops after at most 500 pages, and a walk stopped by that guard counts as incomplete. Multiple search terms and pasted links share the record cap, split proportionally across the remaining sources as the run progresses.

Resume and recurring updates

• Resume (resumeFromRunId) continues ONE interrupted run: paste its run id (or dataset id) and every record it already collected is skipped, so you don't pay twice.
• Recurring updates (incrementalMode) is for scheduled runs over the same setup. Each record is classified NEW, UPDATED (with changedFields), UNCHANGED (suppressed and not billed unless emitUnchanged is on), REAPPEARED, or EXPIRED (only after a run that scanned its whole configured scope end to end, and only with emitExpired). stateKey names or shares the stored baseline. With recurring updates off, output is exactly as before.
• Without a stateKey, the baseline follows the mode, market, search terms or links, sort and the detail-reads switch: changing any of them starts a fresh baseline, while Max records and Max pages do not. Giving two different setups the same stateKey makes them share one baseline, so each can mark the other's products EXPIRED.
• EXPIRED only fires after a scan the actor considers complete: a resumed run, a run stopped by Max records or the page guard, or a run where a source failed never produces it.
• If the storefront answers "page not found" for page 2 or later of a search or listing before the results it declared are all read, the actor logs a warning and treats that search as incomplete: the rows already read are returned, and no EXPIRED rows are produced for that run. (A "page not found" on page 1 is reported as a failed source and also blocks EXPIRED.)

Send results into your apps (MCP connectors)

Optionally pipe results into the apps you already use, via Model Context Protocol (MCP) connectors. This is an extra delivery step after the scrape: the Apify dataset is never changed.

1. Authorize a connector once under Apify, Settings, API & Integrations (Notion, Linear, Airtable, or Apify).
2. Select it in the mcpConnectors input field. Notion gets a rich page-per-item export; other connectors get a best-effort write or a digest.
3. For Notion, also set notionParentPageUrl to the page under which item pages should be created.

maxNotifyListings caps how many items are written to each connector per run; it does not affect the dataset. Leave mcpConnectors empty to skip this step entirely.

Input Parameters

Parameter	Type	Default	Description
mode	string	search	Keyword search on one storefront, or your own product and listing links.
market	string	cl	Country storefront: Chile, Colombia or Peru. Searches run on it; in links mode it sets the connection country and the labels on listing-link rows, so match it to your links.
queries	array	(none)	Search terms for search mode, one entry per search (alias term); all searches share the record cap.
sort	string	relevance	Ordering per search: most relevant (relevance), lowest price (price-asc), highest price (price-desc), brand A to Z (brand-asc) or best rated (rating-desc).
term	string		Alias of queries for a single keyword (used only when queries is empty).
urls	array	one example of each supported shape	Product or listing links for links mode. Each link is fetched, labelled and connected through its own storefront.
fetchDetails	boolean	false	Read each product's own page for description, specifications, stock and seller details (a per-product surcharge applies; pasted product links always carry it).
maxItems	integer	20	Stop after this many records across the whole run (0 = no limit; the page guard still applies).
maxPages	integer	0	Runaway guard: upper bound on pages read for any single search or listing walk (0 = as deep as the source serves, up to 500 pages).
resumeFromRunId	string	(none)	Run id or dataset id of an earlier run to continue; already-collected records are skipped.
incrementalMode	boolean	false	Turn on for recurring monitoring: later runs return only new, updated and reappeared records.
stateKey	string	(none)	Optional name for a monitoring campaign, to keep its memory stable or to share it across setups.
emitUnchanged	boolean	false	Also return (and bill) records that did not change, labelled UNCHANGED.
emitExpired	boolean	false	Also return (and bill) records that are gone, labelled EXPIRED. Only produced after a complete scan.
ignoreFieldsForChanges	array	empty	Field names (dotted paths for nested values, for example sellerDetails.isVerified) whose changes do not make a record UPDATED. The fields are still returned.
mcpConnectors	array	(none)	Optional: send results into apps you authorized under Integrations. Never changes the dataset.
notionParentPageUrl	string	(none)	Notion connector only: page under which item pages are created.
maxNotifyListings	integer	50	Cap on items written to each connector per run.
concurrency	integer	6	Accepted for compatibility; pages are read one at a time to stay polite to the storefront.
retries	integer	3	Accepted for compatibility; the connection ladder decides its own retries.
timeoutSecs	integer	60	Wall-clock budget for one page across all retries and connection fallbacks (5-120). The fallback connection tier takes about 35-40 seconds for a product page, so values below 45 cannot serve pages the site refuses on the first connection.
proxyConfiguration	object	Apify Proxy, residential group	Connection used for every request; pinned to the selected market's country when residential. See Plan Requirement.

Output Example

Sample shape: values are illustrative placeholders, not from a live record.

{
  "recordType": "product",
  "id": "300000001",
  "skuId": "300000002",
  "name": "Mochila de muestra",
  "brand": "MARCA MUESTRA",
  "url": "https://www.falabella.com/falabella-cl/product/300000001/producto-de-muestra",
  "imageUrl": "https://media.falabella.com/falabellaCL/300000002_01/public",
  "currency": "CLP",
  "price": 22990.0,
  "priceDisplay": "$ 22.990",
  "oldPrice": 27990.0,
  "discountPercent": 18,
  "ratingAverage": 4.6,
  "ratingCount": 128,
  "sellerId": "S00042",
  "sellerName": "Tienda de muestra",
  "officialSeller": true,
  "sponsored": false,
  "bestSeller": true,
  "market": "CL",
  "marketHost": "www.falabella.com",
  "description": "Descripcion del producto de muestra.",
  "specifications": [{ "name": "Color", "value": "Azul" }],
  "changeType": "NEW"
}

Plan Requirement

Connection needs differ by page type, so the default connection is a residential-grade proxy pinned automatically to the market's country.

• Listing, search and collection pages (Chile, Colombia): often load on a standard datacenter connection, but not reliably: the storefront sometimes answers with a small "Access denied" page. A single refusal only makes the actor retry from a fresh exit on the same tier; after two refusals in a row on one link it moves that link up to the residential-grade tier, and the next link starts from the configured tier again.
• Product pages and the Peru storefront: treat these as needing escalation. Product pages are frequently answered with a verification page (even on residential connections) and Peru has only been seen refusing datacenter connections. The actor sends refused product pages through an alternate connection tier, but success is not guaranteed: it logs a warning per pasted product or Peru link, and detail reads can come back empty.

Running the actor reliably needs a plan that includes residential access, or your own proxy links pasted into the Connection section. Connection tiers are chosen from what your plan actually allows. Each page has a 60 second budget (the timeoutSecs input) across all retries and connection fallbacks; a product page that needs the fallback connection takes about 35-40 seconds of it.

If your plan has no residential access, the run logs this warning and continues on a backup connection when one is available:

The selected connection is unavailable to this run. Apify free plans do not include residential proxies. Continuing on a backup connection; some pages may still be unavailable.

and, when the backup serves a page, "Served over the backup connection (the default connection was unavailable)." If no page can be loaded the run ends as failed (never as an empty success) with one of these messages:

• Proxy refused for the account: "The proxy connection was refused for this account, so no page could be read. Your Apify plan does not include proxy access; a paid plan is required." Check your own proxy credentials if you entered any.
• Free plan, backup also failed: "No results: residential proxies are unavailable on this run (Apify free plans do not include them), and the backup connection could not load the requested Falabella pages."
• Connection problem: "Could not load any results page (the requests failed), so nothing was collected. This is a connection problem, not an empty search." Try again later.

Output views and run summary

The dataset has three views: Results (every product record), Detail (description, specifications, stock and seller details from detail reads) and Changes (change type, changed fields and first/last seen, for recurring updates). A run with several searches or links also saves an OUTPUT record in the key-value store with items per source, sources that succeeded, failed or were truncated, whether the cap was hit and the change-type counts.

Notes and limits

• Reviews are not included: the review data sits on product pages, which the storefront answers with a verification page, so no review text or limit is offered.

• Links default changed: the default urls value is now one example of each supported link shape (it used to be a single link); clear it or replace it with your own links.

• Product ids and category ids are per country: a Chilean product id does not exist on the Peruvian storefront.

• Max records is the only cap that stops a run; max pages is a runaway guard.

• Resume and recurring updates are mutually exclusive on the same setup: use one or the other.

• Keyword searches can be answered by a keyword collection or a category page; the actor follows the storefront's own redirect and walks that route.

FAQ

How much does it cost?

You pay per product record returned, plus a per-product detail surcharge on every returned product when detail reads are switched on, and on every pasted product link (a product link is always read from its full product page). Records suppressed as unchanged by recurring updates are never charged. The Pricing tab on this actor's page shows the current rates for your plan. Use Max records to cap the cost of any run.

Is it legal to scrape Falabella?

This actor collects only publicly available marketplace data: product listings, prices and seller information as shown on the storefront pages. You are responsible for how you use it: follow Falabella's terms and the laws that apply to you, and get legal advice if you plan commercial redistribution. Prices and product facts are generally not protected, but product photos, descriptions and brand names may be subject to third-party rights.

Can I get only new or changed products on a schedule?

Yes. Schedule the actor from the Schedules tab and turn on Recurring updates. Each run then returns only new, updated and reappeared records; unchanged ones are not returned or billed unless you turn that on too.

Why did a run fail instead of returning an empty dataset?

If every page is refused, the run stops with a clear message so "no results for this search" is never confused with "nothing could be read". Run it again in a few minutes. A genuinely empty search still succeeds, with zero rows and a visible note on the run.

Why does a resumed or capped run never report EXPIRED products?

Recurring updates only marks a product gone after a run that scanned its whole configured scope end to end. A resumed run, a run stopped by Max records, or a run where a source failed never qualifies, because none of them actually confirmed the product is gone; the stored baseline for that product is left untouched until a later, complete run says otherwise.

Can a product be marked EXPIRED by mistake?

No. If the storefront answers "page not found" for page 2 or later of a search or listing, the actor logs a warning and treats the walk as incomplete, so no EXPIRED rows are produced for that run. A product is only marked gone after a run that read its whole search or listing to the end.

What is the difference between the Chile, Colombia and Peru links?

Each market is its own catalogue: product and category ids are not portable across storefronts. A product id from the Chilean site will not resolve on the Colombian or Peruvian one. Links are fetched from the storefront in the link, and the market, marketHost and currency written on each row and the connection country follow that link's own storefront, not the Market setting. Pick the matching market when you search by keyword.

Can I use it with AI agents or MCP?

Yes. Call it from any Apify integration or MCP client, and use the mcpConnectors field to push results into Notion, Linear or Airtable.

🔗 More scrapers you might like

Pair this actor with these related scrapers from the same team:

🧩 Jumia Marketplace Scraper Scrape Jumia, Africa's largest marketplace: products with local prices, discounts...	🧩 Digitec Galaxus Scraper Scrape Digitec Galaxus, Switzerland's largest online retailer: products with local...
🧩 VidaXL Scraper Scrape the vidaXL storefronts by keyword or URL across 6 European markets. Every row...	🧩 Walmart Scraper Scrape Walmart.com by keyword, URL, or item ID. Extract prices, was-prices, sellers...
🧩 Fruugo Scraper Scrape Fruugo marketplace products by keyword or URL across country storefronts. Extract...	🏷️ Wildberries Scraper Scrape Wildberries marketplace: product search with prices (wallet vs retail), discounts...

👉 Browse all abotapi scrapers

💬 Support

• 🐞 Found a bug or a missing field? Open a ticket on the Issues tab. We usually reply within hours.
• 🛠️ Need another site, extra fields or a private build? Email contact@abotapi.com.
• ⭐ Enjoying it? A quick review on the actor page helps other users find it.

Extra listing columns

Each record also carries offering_id, discount_label, badges, promotions, category_id, total_reviews, search_term, shipping_labels and a raw_product JSON string of the listing card, plus snake_case copies of the main fields (product_id, sku_id, seller_name, seller_id, is_sponsored, is_best_seller, rating, image, scraped_at). internet_price is the online price the card publishes as its internet or event price and normal_price is the card's normal (list) price; each is empty when the card does not publish that price type, and the card-holder price is never used for either. They are left out of the recurring-update comparison, so adding them does not mark existing products as UPDATED.