Sephora Product Scraper for US Catalog Research

Extract structured product records from public Sephora US product pages or keyword searches for beauty catalog research, price observation, ingredient analysis, and variant comparison. Each dataset row represents one normalized product with canonical identity, public price and availability, rating aggregates, images, ingredients, variants, source URL, and collection time.

Best fit for this Actor

Choose this Actor when your workflow starts with Sephora US search terms or public product URLs and needs product-level catalog data. It supports broad keyword discovery and detail-first extraction through the same normalized product contract.

Focused standalone workflow

This Actor is designed as a focused standalone workflow for public Sephora US product discovery and product-detail research.

A practical assortment scenario

A beauty category analyst starts with selected search terms and collects a product dataset with brand, price, availability, rating, ingredients, images, and variants. The analyst then filters the rows by brand or ingredient, compares variant breadth, and exports source-linked products into an assortment research sheet.

Quick start input

Search Sephora US products:

{
  "searchQueries": ["rare beauty blush", "vitamin c serum"],
  "maxResults": 20,
  "includeIngredients": true,
  "includeVariants": true
}

For a selected product set, place canonical Sephora US product pages in productUrls.

What data you receive

Field	Meaning
productId, skuId, title, brand	Product and selected SKU identity
productUrl, sourceType, sourceUrl, scrapedAt	Canonical URL, discovery route, provenance, and collection time
price, listPrice, currency, availability	Public price and availability fields
rating, reviewCount	Public product-level review aggregate
description, ingredients, categories	Product presentation and composition fields
imageUrls, variants	Public media and shade, size, or SKU options

{
  "productId": "P97989778",
  "skuId": "2362168",
  "title": "Soft Pinch Liquid Blush",
  "brand": "Rare Beauty",
  "productUrl": "https://www.sephora.com/product/example-P97989778",
  "price": 25,
  "currency": "USD",
  "availability": true,
  "rating": 4.7,
  "ingredients": "Water, Dimethicone, Iron Oxides",
  "imageUrls": ["https://www.sephora.com/productimages/example.jpg"],
  "sourceType": "search",
  "sourceUrl": "https://www.sephora.com/search?keyword=blush",
  "scrapedAt": "<ISO-8601 collection time>"
}

Run through the Apify API

curl -X POST "https://api.apify.com/v2/acts/khadinakbar~sephora-product-scraper/runs" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchQueries":["rare beauty blush"],"maxResults":20,"includeIngredients":true,"includeVariants":true}'

Read the default dataset for product records and the default key-value store for OUTPUT and RUN_SUMMARY.

Use with AI agents through Apify MCP

Example prompt:

Research public Sephora US products for these search terms. Return product ID, brand, title, current public price, availability, rating, ingredients, variants, canonical product URL, source URL, and collection time. Read the dataset and summarize the assortment with source provenance.

Give the agent search terms or canonical product URLs, a focused maxResults scope, and the detail fields needed for the decision. Ask it to read the dataset, retain source URLs, and report the outcome and Apify cost.

Pricing

This Actor uses Pay per event plus Apify platform usage. A product event is associated with a validated normalized product record saved to the dataset. Open the live Pricing tab for current billing details and use maxResults or Apify run cost controls to manage the research scope.

Best results

Provide specific beauty product phrases for discovery or canonical Sephora US product URLs for detail-first research. Enable ingredients and variants when composition or shade analysis supports the project, and retain sourceUrl with scrapedAt when comparing observations over time.

Builder's note

I built the extractor to score the product-shaped data embedded in each page because a retail page can contain several structured objects. Selecting the strongest product candidate before normalization keeps identity, price, ingredients, and variants attached to the same source-linked record.

Responsible use

Collect public product data you are authorized to access and follow applicable laws, Sephora terms, and your organization's catalog and market-research policies.