Nordstrom API

Nordstrom does not publish a self-serve developer API for its product catalog, prices or reviews. The Scavio Nordstrom API fills that gap with 8 POST endpoints that return nordstrom.com (US) data as structured JSON: search, category, brand, product detail, reviews, store-pickup availability, a store locator and search suggestions.

It is a REST API that returns public nordstrom.com product data as structured JSON: keyword search with six sorts, and product detail with every color and size as a sku with its own price, UPC and units available, plus ratings, the size-fit distribution and reviews. It also pulls a product's full review history, browses categories, sale sections and brand pages, checks in-store pickup by size and color, and finds stores. Every call costs 5 credits, except availability and reviews at 10.

Yes. Call /api/v1/nordstrom/product for each style on your watchlist on a schedule and store each sku's price, original_price, percent_off and price_type (regular, clearance or promotion). A change in price or price_type is your price drop alert.

Pull a sale section with /api/v1/nordstrom/category, such as sale/women or /browse/sale/women/clothing/dresses, or search, with sort set to percent_off to get the deepest markdowns first. Every result carries on_sale, percent_off, price, original_price_range and badges such as new_markdown. On product, clearance_price_range shows the clearance price when there is one.

Nordstrom marks down colors separately. In our capture of a Betsy & Adam gown, BLACK and PERIWINKLE were $279 at the regular price while PLUM was $195.30 on clearance. The product's price is the lowest current price, price_range spans all of them, and each sku carries its own price.

Yes. Product returns every size of every color as a sku with sku_id, size, size_type, color, upc, price, original_price, percent_off, price_type, in_stock, quantity and is_final_sale, plus color_count and sku_count. The gown had 3 colors and 22 skus.

Product returns in_stock and quantity for each sku. quantity is the number of units available to ship online; in_stock is true when the sku can be shipped or picked up in a store, so a size held only in stores reads in stock with quantity 0. Search and category results carry a low_inventory flag.

Yes. /api/v1/nordstrom/availability returns, for every size and color, whether it can be picked up, the pickup quantity across the chain and each store that offers in-store pickup, with address and phone. Nordstrom says yes or no per store and does not publish a unit count per store. The Maeve Midi Dress could be picked up at 85 stores in our capture. It costs 10 credits.

Yes. /api/v1/nordstrom/category takes any /browse path or URL, including departments (/browse/women), sale sections (sale/women) and new arrivals (/browse/new), and returns 24 products a page with six sorts, the subcategories and the filter sidebar with product counts. Women's Dresses held 40,339 products in our capture.

Yes. Pass the brand page path, such as /brands/astr-the-label--53360, or a product's brand_url to /api/v1/nordstrom/brand. It returns 24 products a page with six sorts and the brand's filter counts: ASTR the Label had 371 products, 200 of them on sale. The brand id after the double dash is required.

Yes. /api/v1/nordstrom/stores finds Nordstrom stores and Nordstrom Local near a ZIP code or coordinates within up to 100 miles, or by store number, with address, phone, weekly and holiday hours, the next 8 days' opening times, pickup and curbside flags, and restaurants and services. ZIP 10001 returned 14 stores within 100 miles.

Yes. /api/v1/nordstrom/reviews returns a product's review history, up to 200 reviews per call (60 by default), each with title, text, rating, size rating, author, date, helpful votes, verified-purchase flag, reviewer badges, size and color purchased, the reviewer's height, weight, body type and usual sizes, and photos. It also returns the average rating, total reviews, star histogram, size and width rating, the most praised and criticized features with a quote each, and Nordstrom's AI review summary. It costs 10 credits. Product still includes the reviews shown on the product page, up to about 13.

Up to 200 per call. Set max_reviews from 1 to 200; total_results and has_more tell you whether more reviews match. There is no page parameter, so for a product with more than 200 reviews, split the history by star rating with stars (the star histogram gives each rating's count) to reach more of it. Rating-only reviews, with a star rating and no text, are included with text null: 78 of the Maeve Midi Dress's 216 reviews were rating-only.

Yes, the same ways a shopper can on nordstrom.com. sort takes most_helpful (the default), most_recent, highest_rated or lowest_rated; stars limits to chosen ratings, such as [1, 2] for the critical ones; verified_only and with_photos narrow to verified buyers or reviews with photos; and search returns reviews mentioning a word. search comes back in Nordstrom's own order, so sort cannot be set with it: a request with both is a 400.

Longer than the other Nordstrom calls: typically 20 to 60 seconds, up to about 100 at max_reviews 200 (18 reviews took 13.8 seconds in our capture). Set a client timeout of at least 120 seconds. If an attempt fails on Nordstrom's side the API retries automatically, and a call that still fails is a 502 that is not charged. When too many review calls are running, it returns a 503 with Retry-After, also not charged.

size_fit has the average and the distribution from too small (1) to too big (5), fit_info has Nordstrom's fit notes, and each review has a size_rating on the same scale. For the gown, 13 of 22 shoppers said true to size and the fit note said it runs small.

Yes. is_marketplace is true for items sold by a third-party marketplace seller, and seller is that seller's name. In our dress search, 3 of the 24 results were marketplace listings, two sold by Mac Duggal and one by Kiyonna.

24 products per page, up to page 100. Every search returns total_results and page_count: "dress" matched 47,123 products over 1,964 pages, so narrow the query, browse a subcategory with /api/v1/nordstrom/category, or change the sort to reach products past page 100.

POST the nordstrom.com product URL, such as https://www.nordstrom.com/s/maeve-midi-dress/7737947, as url to /api/v1/nordstrom/product, or just the style id at the end of it (7737947) as style_id. An unknown style id is a 404 and is not charged.

Every Nordstrom endpoint costs 5 credits per successful call, except availability at 10, because it reads the product page and then looks up the pickup stores, and reviews at 10. An unknown product, category or brand (404) or a temporary outage (502) is not charged. New accounts get 50 free credits on signup, one time, with no card required: 10 Nordstrom calls at 5 credits. Paid plans start at $15 per month for 3,000 credits.

Every request fetches live from nordstrom.com at the moment you call it, and nothing is cached. Each response carries fetched_at, the time the data was retrieved. Measured in production: search and product answer in 3 to 7 seconds (the captured search sample took 16), stores and suggestions in about 3, availability in about 11 because it makes two lookups, brand pages in 9 to 27 seconds, category pages in 7 to 60 seconds (large sale sections can take up to about a minute), and reviews in 20 to 60 seconds, up to about 100 at 200 reviews. Set a client timeout of 120 seconds.

nordstrom.com in the United States, with prices in US dollars. The store locator and pickup availability cover Nordstrom's US stores, both full-line stores and Nordstrom Local. Nordstrom Rack is a separate site with its own catalog, prices and reviews, covered by the Scavio Nordstrom Rack API at /api/v1/nordstromrack.

Every endpoint is a plain HTTPS POST with a JSON body, so any language works. Scavio also ships Python and JavaScript SDKs, a remote MCP server at mcp.scavio.dev for AI agents, and an n8n community node, all on the same API key.

Scavio returns publicly available nordstrom.com data and does not sign in to Nordstrom customer accounts or read order history. You remain responsible for how you use the data, including compliance with Nordstrom's terms. Talk to your own counsel for your specific use case.