Costco Product Reviews Scraper

Extract public customer reviews from Costco US product pages into clean, structured data.

Provide Costco product URLs or product IDs and get one dataset row per review, including ratings, review text, reviewer context, helpfulness, badges, media, product metadata, and aggregate rating statistics.

Use the Actor for recurring reputation monitoring, product research, quality alerts, competitor analysis, and sentiment pipelines—without browser automation or a Costco account.

What does Costco Product Reviews Scraper do?

The Actor turns Costco product review pages into export-ready records.

It can:

• 🛒 Process one or many Costco products in one run
• ⭐ Extract one through five-star customer reviews
• 📝 Return review titles and full review text
• 👤 Capture public reviewer nickname and location
• ✅ Identify verified-purchaser badges when available
• 👍 Include helpful and unhelpful vote totals
• 🖼️ Preserve public review photos and videos
• 📊 Add product-level average rating and rating distribution
• 📅 Filter by submission date
• ↕️ Sort by newest, highest rated, or lowest rated
• 🔁 Follow offset pagination automatically
• 📦 Export JSON, CSV, Excel, XML, RSS, or JSONL through Apify

Who is it for?

Brands and manufacturers

Track customer feedback about products sold through Costco and detect recurring quality complaints.

Marketplace and ecommerce teams

Compare ratings, recommendation rates, and review themes across competing products.

Reputation and customer-experience teams

Schedule review ingestion and feed new reviews into dashboards, alerts, or ticketing workflows.

Data and research teams

Build longitudinal datasets for sentiment analysis, category research, or product benchmarking.

Agencies and consultants

Deliver structured Costco review intelligence without manual copying or custom scraper maintenance.

Why use this Costco review extractor?

Costco product pages are designed for shoppers, not bulk analysis.

This Actor provides:

• Repeatable structured output
• Multi-product processing
• Stable review and product identifiers
• User-controlled limits
• Source-side sorting and filtering
• Automatic pagination
• Built-in Apify scheduling
• Dataset exports and API access
• Zapier, Make, Google Sheets, and webhook integrations

The implementation uses Costco's public review-data integration rather than fetching protected product-page HTML.

Quick start

1. Open the Actor input.
2. Add one or more Costco product URLs or numeric product IDs.
3. Set the total and per-product review limits.
4. Optionally select ratings, dates, or sorting.
5. Click Start.
6. Open the Dataset tab to preview or export reviews.

A low-cost first run can use product ID 100737171 with a limit of 20 reviews.

Input

Field	Type	Default	Description
productUrlsOrIds	string[]	sample product	Costco US URLs or numeric product IDs
maxReviews	integer	100	Global review-row limit
maxReviewsPerProduct	integer	100	Limit for each product
sort	string	newest	newest, highest-rated, or lowest-rated
ratings	integer[]	all	Optional ratings from 1 to 5
dateFrom	string	none	Inclusive ISO start date/datetime
dateTo	string	none	Inclusive ISO end date/datetime

Example input

{
  "productUrlsOrIds": [
    "https://www.costco.com/p/-/kirkland-signature-platinum-performance-ultrashine-dishwasher-detergent-pacs-115-count/100737171",
    "100737171"
  ],
  "maxReviews": 50,
  "maxReviewsPerProduct": 50,
  "sort": "newest",
  "ratings": [1, 2],
  "dateFrom": "2026-01-01"
}

Duplicate product IDs are processed only once.

Supported Costco URLs

The Actor supports Costco US product URLs whose host is costco.com or a Costco subdomain.

It extracts the numeric product ID from the URL path.

You may also provide the numeric ID directly.

Non-Costco URLs fail closed instead of silently producing unrelated data.

Costco Business Center search/category URLs are outside this Actor's scope.

Sorting and filters

Use newest for monitoring recently submitted reviews.

Use highest-rated to collect positive customer language.

Use lowest-rated to prioritize complaints and product risks.

The ratings filter accepts any combination of integers 1–5.

dateFrom and dateTo accept ISO dates or timestamps.

When both dates are supplied, dateFrom must not be after dateTo.

Output

Each dataset item is one public customer review.

{
  "reviewId": "252764979",
  "productId": "100737171",
  "rating": 5,
  "ratingRange": 5,
  "title": "Superior cleaning",
  "reviewText": "This product cleans as well as the name brands",
  "userNickname": "Ljr1",
  "userLocation": "St Louis MO",
  "submissionTime": "2026-07-13T13:02:09.000+00:00",
  "isRecommended": true,
  "isVerifiedPurchaser": true,
  "helpfulVotes": 0,
  "photos": [],
  "videos": [],
  "productName": "Kirkland Signature Platinum Performance UltraShine Dishwasher Detergent Pacs, 115-count",
  "averageRating": 4.58,
  "totalReviewCount": 1644,
  "scrapedAt": "2026-07-20T00:00:00.000Z"
}

Data fields

Field group	Included data
Identity	reviewId, productId, authorId
Review	rating, ratingRange, title, reviewText
Reviewer	userNickname, userLocation, contentLocale
Time	submissionTime, lastModificationTime, scrapedAt
Signals	isRecommended, isVerifiedPurchaser, isSyndicated
Feedback	helpfulVotes, unhelpfulVotes, totalFeedbackCount
Rich context	badges, contextData, secondaryRatings, photos, videos
Product	productName, productUrl, productImageUrl, productBrand
Statistics	averageRating, totalReviewCount, ratingDistribution

Optional source values are represented as empty arrays/objects or null, not invented values.

How much does it cost to scrape Costco product reviews?

The Actor uses pay-per-event pricing.

• A small start fee is charged once per run.
• A result event is charged for each review saved.
• Per-review prices decrease by Apify subscription tier.

The exact active prices are shown in the Actor's pricing panel before a run starts.

Control spend with maxReviews and maxReviewsPerProduct.

Start with 20 reviews, inspect the output, then increase the limit for production workflows.

Scheduling review monitoring

Use an Apify schedule to run daily or weekly.

Recommended monitoring input:

{
  "productUrlsOrIds": ["100737171"],
  "maxReviews": 100,
  "maxReviewsPerProduct": 100,
  "sort": "newest",
  "dateFrom": "2026-07-01"
}

Send the completed dataset to a webhook or downstream automation.

Deduplicate across scheduled runs by reviewId.

Quality-alert workflow

To focus on negative reviews:

1. Set ratings to [1, 2].
2. Set sorting to lowest-rated or newest.
3. Schedule the Actor daily.
4. Route new reviewId values to Slack, email, or a ticket queue.
5. Group complaints by product and recurring phrases.

This creates an actionable quality-monitoring feed rather than a manual review-page check.

Sentiment and analytics workflow

Export review text to your preferred NLP or BI system.

Useful metrics include:

• Rating distribution over time
• Negative-review share
• Recommendation rate
• Verified-purchaser share
• Helpful-vote leaders
• Complaint topics by product
• Review velocity after a launch or packaging change

Use submissionTime for trends and scrapedAt for pipeline auditing.

Integrations

Google Sheets

Use an Apify integration or Make scenario to append new dataset items to a sheet.

Zapier

Trigger a Zap after a successful run and route low ratings to Slack or email.

Make

Schedule multi-product extraction, filter by rating, and update a product dashboard.

Webhooks

Attach a run-succeeded webhook to notify your ingestion service when data is ready.

Cloud storage and databases

Fetch dataset items through the API and load them into BigQuery, Snowflake, PostgreSQL, S3, or a data lake.

API usage with cURL

curl -X POST \
  "https://api.apify.com/v2/acts/automation-lab~costco-product-reviews-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"productUrlsOrIds":["100737171"],"maxReviews":20}'

Use the returned run ID to inspect status and the default dataset.

API usage with JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('automation-lab/costco-product-reviews-scraper').call({
  productUrlsOrIds: ['100737171'],
  maxReviews: 20,
  sort: 'newest',
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

API usage with Python

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ['APIFY_TOKEN'])
run = client.actor('automation-lab/costco-product-reviews-scraper').call(run_input={
    'productUrlsOrIds': ['100737171'],
    'maxReviews': 20,
    'sort': 'newest',
})
items = client.dataset(run['defaultDatasetId']).list_items().items
print(items)

Use with Apify MCP and AI agents

Claude Code

$claude mcp add --transport http apify https://mcp.apify.com?tools=automation-lab/costco-product-reviews-scraper

Claude Desktop, Cursor, and VS Code setup

Use this mcpServers JSON in Claude Desktop or Cursor. VS Code users can add the same HTTP MCP server through the MCP configuration panel.

{
  "mcpServers": {
    "apify": {
      "type": "http",
      "url": "https://mcp.apify.com?tools=automation-lab/costco-product-reviews-scraper"
    }
  }
}

Example prompts:

• “Extract the 50 newest reviews for Costco product 100737171.”
• “Collect one- and two-star reviews for these Costco product URLs.”
• “Compare negative-review themes across these three Costco products.”
• “Schedule a weekly Costco review monitor and summarize new complaints.”

For Claude Desktop, add the Apify MCP server using your Apify token and enable this Actor tool.

Performance and limits

The Actor uses lightweight HTTP requests and does not launch a browser.

It requests up to 100 reviews per source page and follows offsets automatically.

A run can process up to:

• 100 submitted products
• 10,000 total review rows
• 5,000 review rows per product

Source availability and the selected filters determine the actual result count.

Reliability and errors

Requests use a 30-second timeout and bounded exponential-backoff retries.

The Actor deduplicates review IDs inside each run.

Invalid inputs fail with a clear message.

A complete source-request failure makes the run fail instead of reporting misleading success.

An empty dataset can be valid when a product has no matching non-rating-only reviews or filters exclude all records.

Tips for best results

• Use numeric product IDs when you already store them in a catalog.
• Keep first-run limits low.
• Use newest sorting for recurring monitors.
• Use explicit date windows to reduce repeated extraction.
• Deduplicate scheduled exports by reviewId.
• Preserve productId as the stable join key.
• Check both totalReviewCount and returned rows: aggregate counts can include ratings-only submissions excluded from textual reviews.
• Avoid submitting search or category URLs; this Actor expects product URLs.

Troubleshooting

Why is the returned count lower than Costco's displayed rating count?

Costco's aggregate count can include ratings-only submissions. This Actor returns public reviews with review content and excludes ratings-only records.

Why did my URL fail validation?

Only Costco US product URLs on costco.com and direct numeric product IDs are accepted. Search URLs, unsupported country sites, and unrelated hosts fail closed.

Why did I receive no rows?

Check the product ID and remove rating/date filters. The product may have no public text reviews matching the selected range.

Why are some reviewer or media fields empty?

Those fields are optional source data. Not every customer publishes a location, photo, video, or secondary rating.

Data freshness

Results reflect the public Costco review data available when the run executes.

submissionTime is the review's source timestamp.

lastModificationTime shows the latest public modification when supplied.

scrapedAt records when the Actor transformed the record.

Schedule repeated runs when you need ongoing freshness.

Legal and ethical use

The Actor extracts publicly accessible product-review data.

You are responsible for using output in accordance with applicable laws, Costco's terms, Bazaarvoice terms, and privacy requirements.

Do not use public reviewer information for harassment, discrimination, identity resolution, or unsolicited outreach.

Collect only what your workflow needs, secure exported datasets, and apply suitable retention policies.

This Actor is not affiliated with or endorsed by Costco or Bazaarvoice.

Related scrapers

Explore other ecommerce and review tools from Automation Lab:

• Walmart Reviews Scraper
• Amazon Reviews Scraper
• Steam Reviews Scraper
• Costco Business Center Product Scraper

Use the Business Center actor for product search, prices, stock, and catalog cards. Use this Actor when individual Costco.com customer reviews are the primary output.

FAQ

Does it require a Costco account?

No. The supported review scope uses public review data.

Does it use a proxy or browser?

No proxy or browser is required for the current public review route.

Can I scrape several products?

Yes. Add multiple URLs or IDs to productUrlsOrIds and control per-product and total limits independently.

Can I filter for negative reviews?

Yes. Set ratings to [1, 2] and choose newest or lowest-rated sorting.

Can I export to CSV or Excel?

Yes. Open the dataset and select CSV, Excel, JSON, XML, RSS, or JSONL.

Can I run it on a schedule?

Yes. Apify schedules can run the same input hourly, daily, weekly, or with a custom cron expression.

Are aggregate statistics included?

Yes. Rows include the average overall rating, total aggregate review count, and rating distribution when supplied by the source.

Are review images included?

Yes. Public photo and video metadata is preserved when a review contains media.

Is every aggregate rating a text review?

No. Aggregate counts may include ratings-only submissions. Dataset rows represent non-rating-only public reviews.

Support

If a valid Costco US product stops returning expected public reviews, open an Actor issue with:

• The product URL or ID
• The approximate expected count
• Your input without secrets
• The run ID
• The relevant log message

That evidence makes source changes easier to reproduce and fix.