IKEA Scraper — Product Prices, Discounts & Customer Reviews Across 50+ Markets

Extract structured IKEA product and review data for price monitoring, catalog builds and market research — across 50+ country markets, with no browser and no proxy required.

One API call returns up to 1,000 fully-structured products, including localised price, previous price, discount amount, stock status, category path, measurements, variants and images. Customer reviews and per-product rating summaries land in a second dataset.

Measured: 200 products from one search call in ~2 s; review batches of 200 at ~0.7 s per request, 10 concurrent. IKEA has no public developer API for catalog or review data, and www.ikea.com sits behind a Cloudflare wall that returns HTTP 403 to datacenter IPs — this Actor reads the two JSON microservices that feed IKEA's own storefront instead.

---

Main use cases

• IKEA price monitoring — track price cuts, NEW_LOWEST_PRICE tags and discount windows across countries
• Multi-market price comparison — same SKU priced in GBP vs EUR vs PLN, with sourceCountry provenance on every row
• Catalog and product-database builds — full category trees plus every variant SKU with its own price, rating and image
• Review and sentiment analysis — customer review text, star ratings, per-topic sub-ratings and rating distributions
• AR / configurator and room-planner pipelines — measurements, colours, category path and image sets
• Competitor assortment research — how IKEA ranges a category in a given market versus another

---

Which data sources are used, and why

Products — IKEA search microservice

GET https://sik.search.blue.cdtapps.com/{country}/{language}/search-result-page
        ?q={query}&size={1..1000}&types=PRODUCT

Example:

https://sik.search.blue.cdtapps.com/gb/en/search-result-page?q=sofa&size=200&types=PRODUCT

Only a normal desktop User-Agent and Accept: application/json are needed. No cookies, no API key, no session token, no CSRF token.

Reviews and ratings — IKEA ratings microservice

POST https://web-api.ikea.com/tugc/public/v5/reviews/{country}/{language}/{itemNo}
GET  https://web-api.ikea.com/tugc/public/v5/rating/{country}/{language}/{itemNo}

One static header authenticates both:

x-client-id: a1047798-0fc4-446e-9616-0afe3256d0d7

The review id comes from each product row's itemNo field.

Why these and not ikea.com

Surface	Result
www.ikea.com/{c}/{l}/p/{slug}-{itemNo}/	HTTP 403 — Cloudflare WAF blocks datacenter and residential IPs
Legacy www.ikea.com/.../products/search/ JSON	HTTP 403 — same wall
__NEXT_DATA__ / ld+json / dataLayer blobs	Unreachable — the HTML is never served, so there is no blob to parse
sik.search.blue.cdtapps.com	HTTP 200, clean JSON, no auth — the exact endpoint the ikea.com frontend calls
web-api.ikea.com	HTTP 200, clean JSON, one static header — the exact endpoint the frontend calls

Both microservices live on different hosts from www.ikea.com with their own edges, so they do not carry the storefront's WAF rules. This is the "find a structured source on a cheaper surface" bypass: the storefront blocks you, the backends that feed the storefront do not.

Why it is dramatically cheaper than a browser

• Up to 1,000 products per request. There is no page/offset parameter — size (max 1000) is the whole pagination mechanism.
• No browser at all. A run is plain JSON HTTP: CheerioCrawler over the search API, HttpCrawler over the review API. Zero rendering cost.

Validated API behaviour

• size must be 1–1000 (> 1000 → 400 "Size parameter cannot exceed 1000").
• There is no page/offset/start parameter, and moreToken is not accepted over HTTP. Use a larger size or more queries.
• Category keys work as queries — e.g. q=25219 returns the whole "Dining chairs" category. Category keys are published in every product's categoryPath[].key, so the whole tree is crawlable this way.
• IKEA category IDs are market-specific (25219 is a GB category).
• Review pages accept up to 500. IKEA pools some reviews at series level, so a product can return reviews for sibling variants; reviews are deduplicated globally by review id across the run.
• A product with no reviews returns 200 with an empty array; rating returns an empty array too.
• Errors arrive as {"status": 400, "title": "Bad Request", "detail": "..."} and are converted into Crawlee retries (3 attempts).

---

Input

Field	Type	Required	Default	Notes
searchQueries	array	no	["chair"]	Product keywords or category keys. One API call each.
country	string	no	us	IKEA market code: us, gb, de, fr, se, ch, ca, au, nl, pl, es, it, at, no, dk, fi, cz, pt, ie, jp, mx, br, in, zh, sg, my, tr.
language	string	no	en	Language segment for that market: en, de, fr, sv, nl, es…
resultsPerQuery	integer	no	200	Results per call. API max 1000.
useSeedQueries	boolean	no	false	Adds ~60 built-in seed queries (a–z letter pairs + furniture terms) for a broad catalog sweep. Adds one API call per seed.
sortOrder	string	no	"" (API default)	RELEVANCE, PRICE_LOW_TO_HIGH, PRICE_HIGH_TO_LOW, NEWEST, RATING, NAME_ASCENDING, MOST_POPULAR, WIDTH, HEIGHT, DEPTH, LENGTH.
includeVariants	boolean	no	true	Include sibling colour/finish variant SKUs as their own rows.
scrapeReviews	boolean	no	true	Fetch reviews for every scraped product into the second dataset. Requires 1–2 extra API calls per product.
reviewsPerProduct	integer	no	50	Max reviews kept per product (1–100000). Above 500, requests paginate with page.size=500.
reviewsSortBy	string	no	newest	newest, oldest, helpful, rating-desc, rating-asc.
includeRatingSummary	boolean	no	true	One free rating-summary row per product with star distribution and sub-ratings. Only used when scrapeReviews is on.
maxItems	integer	no	0 (uncapped)	Hard cap on product rows.
maxRequestsPerCrawl	integer	no	1000	Total API-call budget, shared by the product and review phases.
proxy	object	no	useApifyProxy: false	Standard Apify proxy picker. Not needed for normal volumes.

---

Output

Two flat JSON datasets. The product dataset offers an overview view and a prices price-history view.

Products dataset (default) — one row per product, plus one per variant when includeVariants is on

Field	Description
name, productName, productType, itemType	e.g. SANDSBERG / SANDSBERG Chair
itemNo, itemNoGlobal	IKEA article number. itemNo is the join key into the reviews dataset.
price, currency, priceFormatted	Localised: 35 / GBP / £35
previousPrice, previousPriceFormatted, previousPriceLabel	e.g. 40 / £40 / "Previous price"
discountAmount, discountPercentage, discountText	e.g. 5 / 12 / -12%
priceTag, priceTagText	NONE, NEW_PRODUCT, FAMILY_PRICE, DISCONTINUED_PRODUCT
priceValidFrom, priceValidTo, availabilityDisclaimer	Price-validity window — the fields price monitors key on
advertisement[]	IKEA Family / in-store-only offers
ratingValue, ratingCount	Average rating and review count
colors[], colorHex[], selectedColor	Colour options
measurements.text, measurements.variants	e.g. 40/35x38x7 cm
quickFacts[], features[]	Material and type chips
numberOfVariants, variantGroups[], variants[]	Variant matrix + every sibling SKU with its own price, rating, image and URL
mainImageUrl, mainImageAlt, contextualImageUrl, images[]	All image types (MAIN / CONTEXT / FUNCTIONAL / QUALITY)
categoryPath[], categoryPathText, businessStructure, filterClass	Full taxonomy, A > B > C
onlineSellable, lastChance, availability[], stockStatus	AVAILABLE / LAST_CHANCE / NOT_ONLINE_SELLABLE
sourceCountry, sourceLanguage, sourceQuery, scrapedAt	Provenance — essential for multi-market comparison

Reviews dataset (reviews, only when scrapeReviews is on)

Two row kinds, told apart by rowType:

• rowType: "review" — one row per customer review: reviewId, reviewItemNo, title, text, ratingValue + secondaryRatings[], isRecommended, verifiedPurchase, verifiedReviewer, anonymousReviewer, feedback counts, reviewerDisplayName, submissionOn / updatedOn, media[], ikeaResponse. Join keys (itemNo, itemNoGlobal, productName, pipUrl) are repeated on every review row. Billed per review.
• rowType: "rating-summary" — one row per product: overall ratingValue, ratingDistribution[] (per-star counts), subRatings[], totalRatingCount, totalRecommendedCount, firstReviewedOn / lastReviewedOn. Free.

Filter on rowType client-side when you only want one kind.

---

Example input

Price sweep across two markets:

{
  "searchQueries": ["sofa", "desk"],
  "country": "gb",
  "language": "en",
  "resultsPerQuery": 500
}

Whole-category crawl using a market-specific category key, with variants and reviews off for speed:

{
  "searchQueries": ["25219"],
  "country": "gb",
  "language": "en",
  "includeVariants": false,
  "scrapeReviews": false
}

Broad catalog sweep with reviews, deepest ratings first:

{
  "useSeedQueries": true,
  "country": "us",
  "language": "en",
  "resultsPerQuery": 1000,
  "maxItems": 5000,
  "scrapeReviews": true,
  "reviewsPerProduct": 100,
  "reviewsSortBy": "rating-desc"
}

---

Example output

Product row:

{
  "name": "SANDSBERG",
  "productName": "SANDSBERG Chair",
  "itemNo": "605.286.98",
  "itemNoGlobal": "60528698",
  "price": 35,
  "currency": "GBP",
  "priceFormatted": "£35",
  "previousPrice": 40,
  "previousPriceFormatted": "£40",
  "discountAmount": 5,
  "discountPercentage": 12,
  "discountText": "-12%",
  "priceTag": "NONE",
  "ratingValue": 4.6,
  "ratingCount": 1284,
  "measurements": { "text": "40/35x38x7 cm", "variants": [] },
  "categoryPathText": "Home office furniture > Office chairs > Desk chairs",
  "stockStatus": "AVAILABLE",
  "onlineSellable": true,
  "numberOfVariants": 4,
  "sourceCountry": "gb",
  "sourceLanguage": "en",
  "sourceQuery": "25219",
  "scrapedAt": "2026-09-29T10:00:00.000Z"
}

Review row:

{
  "rowType": "review",
  "reviewId": 7712345678,
  "reviewItemNo": "605.286.98",
  "itemNo": "605.286.98",
  "productName": "SANDSBERG Chair",
  "title": "Great value",
  "text": "Sturdy and comfortable for the price. Assembly took about ten minutes.",
  "ratingValue": 5,
  "isRecommended": true,
  "verifiedPurchase": true,
  "reviewerDisplayName": "M. K.",
  "submissionOn": "2026-08-14",
  "scrapedAt": "2026-09-29T10:00:01.000Z"
}

---

Build IKEA price monitors and product databases on a schedule

Every Apify Actor is callable over HTTP, so the natural setup is a recurring run against the same query set, diffed into your own store:

• Daily — same searchQueries + country; diff price, previousPrice, priceTag and priceValidFrom to catch cuts and new-lowest-price flags
• Weekly — rotate countries to build a cross-market price index for the same article numbers
• On demand — call the Actor straight from your backend via the Apify API and read the dataset as JSON, CSV or Excel
• Connect the tools you already use — Apify webhooks, Zapier, Make, Google Sheets and 100+ integrations fire on run completion

Run it on latest, pin a build number, or trigger it from your own scheduler. The prices dataset view is a ready-made price-history table with valid-from / valid-to dates.

---

Related Actors

• Ikea Scraper (crawlerbros/ikea-scraper) — multi-country IKEA storefront scraper with per-product and per-category modes
• IKEA Product Search Scraper (cirkit/ikea-product-search-scraper) — IKEA catalog search with optional per-store stock enrichment
• IKEA Reviews Scraper (shahidirfan/IKEA-Reviews-Scraper) — dedicated review-only run for large-scale review collection
• Amazon Bestsellers Scraper — the same price-monitoring and ranking workflow on Amazon
• Idealo Scraper — German price-comparison history across 6 markets, pairs well for EU assortment studies

---

FAQ

Why not use an official IKEA API instead?

IKEA publishes no public developer API for catalog, pricing or review data. The only documented services are internal IKEA Knowledge Graph endpoints (knowledge.api.inter.ikea.net), which cover product add-ons and range recommendations — not search, prices or reviews. Every third-party "IKEA API" is a wrapper around the same public JSON microservices this Actor reads directly, so you can skip the subscription layer.

What are the alternatives to this Actor?

• Browser-based IKEA scrapers — work, but roughly 50× the cost per run for the same JSON, and they still have to defeat Cloudflare.
• Managed third-party APIs (RapidAPI listings, data-API vendors) — add a subscription, a per-call price and a vendor dependency between you and the same source.
• The archived ikea_api Python client — requires generating a guest token through ikea.com cookies and expires every 30 days.

How do I scrape a whole IKEA category?

Read any product's categoryPath[] entries — each has a key — then pass that key as a searchQueries value (q=25219 returns the entire "Dining chairs" category in one call). Category IDs are market-specific, so resolve them from a product in the same country you are crawling. For the broadest sweep in one run, set useSeedQueries: true.

Can I get more than 1,000 products in one run?

Yes. resultsPerQuery caps a single call at 1,000, but the run continues across multiple queries and across useSeedQueries seeds. Use maxItems to cap total product rows and maxRequestsPerCrawl to cap total API calls.

Why are my reviews fewer than the rating count shows?

IKEA pools some reviews at series level, so reviews are deduplicated globally by review id across the whole run — two products in the same series share their pool. reviewsPerProduct also caps what is kept per product, and includeRatingSummary gives you the true per-product distribution for free.

Do I need a proxy, cookies or an API key?

No. The search and review microservices are read from any IP with a desktop User-Agent, and the review API authenticates with a single static public header. proxy defaults to useApifyProxy: false — enable it only if you specifically need a particular country exit.

How much does it cost?

Products are billed at $0.0007–$0.00075 per row depending on your plan tier, so roughly $0.70–$0.75 per 1,000 products; each review is billed as its own event and rating summaries are free. Product rows are dominated by plain JSON HTTP, not compute. Free Apify accounts are capped at 10 products per run; paying accounts are uncapped within maxItems / maxRequestsPerCrawl.

---

For AI Agents & LLM Apps

Callable via the Apify MCP server and the Apify API. Returns IKEA products (default dataset) and, when enabled, customer reviews plus per-product rating summaries (second dataset).

Minimal valid input — one call returns up to 1,000 products:

{ "searchQueries": ["chair"] }

Category crawl (category keys resolve to whole categories, market-specific):

{ "searchQueries": ["25219"], "country": "gb", "language": "en" }

Multi-market price comparison — run the same query in several markets; sourceCountry / sourceLanguage on every row tell you which shop each price came from:

{ "searchQueries": ["sofa"], "country": "de", "language": "de", "resultsPerQuery": 500 }

Products only, cheapest run — scrapeReviews: false removes all review events from the bill:

{ "searchQueries": ["desk"], "scrapeReviews": false, "includeVariants": false }

Output fields — products: name, productName, productType, itemNo, itemNoGlobal, price, currency, priceFormatted, previousPrice, previousPriceFormatted, discountAmount, discountPercentage, discountText, priceTag, priceTagText, priceValidFrom, priceValidTo, ratingValue, ratingCount, colors, colorHex, measurements, features, numberOfVariants, variantGroups, variants, mainImageUrl, images, categoryPath, categoryPathText, onlineSellable, lastChance, stockStatus, sourceCountry, sourceLanguage, sourceQuery, scrapedAt. Reviews dataset: rowType, reviewId, reviewItemNo, itemNo, productName, title, text, ratingValue, isRecommended, verifiedPurchase, reviewerDisplayName, submissionOn, plus rating-summary rows carrying ratingDistribution and subRatings.

Behaviors an agent should know

• itemNo is the join key between the products dataset and the reviews dataset. Reviews cannot be requested without first discovering products.
• There is no page/offset input. resultsPerQuery (max 1000) is the only pagination mechanism; searchQueries and useSeedQueries are how you go wider. Do not invent an offset field.
• scrapeReviews defaults to true. If you want a catalog-only run, you must set it to false explicitly — otherwise every product triggers 1–2 extra billed API calls.
• useSeedQueries: true adds roughly 60 additional API calls on top of your queries. Budget maxRequestsPerCrawl accordingly; it defaults to 1000 and is shared across the product and review phases.
• Reviews are deduplicated by review id across the entire run, so ratingCount on a product and the number of review rows for it will not match. Use rowType: "rating-summary" rows for the true per-product distribution — those rows are free.
• categoryPath[].key values are market-specific. A key scraped from the GB shop will not return the same category on the DE shop. Resolve keys per country.
• variants[] embeds sibling SKUs; with includeVariants: true (default) those siblings can also appear as their own top-level rows. Deduplicate on itemNo.
• Free accounts are capped at 10 products per run regardless of maxItems.

Billing model — pay per event: one Products event per product row ($0.0007–$0.00075), one Review event per review row, and rating-summary rows are free. The Actor stops fetching reviews as soon as the run's billing limit is reached rather than overrunning the budget.

---

Legal & Compliance Disclaimer

This Actor is an independent third-party tool. It is not affiliated with, endorsed by, sponsored by, or connected to IKEA, Ingka Group, or any of their brands.

It reads only publicly available JSON API responses from two microservices (sik.search.blue.cdtapps.com and web-api.ikea.com). It does not log in, does not bypass authentication, does not solve CAPTCHAs, does not use residential proxies by default, and does not collect any personal data about individuals. Extracted review text contains reviewer display names that IKEA itself publishes publicly — treat those as personal data and avoid sending unsolicited commercial outreach to them, as that can violate applicable law including GDPR, UK GDPR, CCPA, CAN-SPAM and TCPA.

You are responsible for your own compliance with IKEA's Terms of Service and with the data-protection law that applies to you. This Actor is a data-collection tool; it does not provide legal advice and makes no claim that any particular use is permitted.

---

SEO Keywords

ikea scraper, ikea api, ikea product data, ikea price monitoring, ikea price tracker, ikea catalog data, furniture price monitoring, ikea reviews scraper, ikea product database, ikea stock data, ikea variant data, ikea category scraper, ikea api alternative, ikea dataset, multi-country ikea pricing, ikea market research, ecommerce price monitoring, furniture catalog database, product review scraper, competitor assortment analysis, drop shipping furniture data, ikea bulk product data