Scrape Amazon reviews for voice-of-customer research
Pull thousands of Amazon reviews with rating, date, verified badge, and helpful votes. Feed them to Claude for sentiment analysis or category-leader benchmarking.
Why your current scraper doesn't cut it
Amazon reviews are the richest voice-of-customer data source on the web. Product managers, ecommerce founders, and agencies use them to find positioning gaps, surface feature requests, and benchmark category leaders. The problem is pulling them — Amazon's review pages are heavily A/B tested and most scrapers break within a week.
Stekpad re-learns the review DOM automatically. You click one review, Stekpad highlights the rest, and the recipe captures rating, title, body, verified badge, helpful votes, and date. Then Claude does the analysis.
How to do it with Stekpad
- 1
Open the reviews page of any Amazon product
Navigate to "See all reviews" on the product detail page. Stekpad works on the standard review list view. - 2
Click a review
Stekpad detects the review pattern and highlights every review on the page. Add fields: rating, title, body, verified badge, date, helpful votes. - 3
Let Stekpad paginate through all reviews
The recipe walks through every review page up to Amazon's hard limit (10,000 most recent). For top products this is a 20-30 minute run. - 4
Hand the corpus to Claude for analysis
Pipe rows to a Google Sheet or straight into Claude via the MCP server. Ask Claude to surface top complaints, feature requests, and positioning gaps.
What you get that other tools miss
Full review payload
Rating, title, body, date, verified badge, helpful votes — everything Amazon shows.
Survives DOM changes
Gemma selector engine re-learns the pattern when Amazon A/B tests.
Claude-ready
Pipe the corpus directly to Claude via MCP for sentiment and theme analysis.
Multi-product
Save the recipe once, re-run on every ASIN in your competitive set.
Related on Stekpad
More in this cluster
How to Scrape Google Maps Leads Without Code
Step-by-step guide targeting local agency owners, sales reps, and growth teams who want local business leads. Covers: opening the Stekpad side panel on a Google Maps search results page, clicking to select business name, address, phone, rating, and website fields, running a full-page scrape, exporting to Sheets. Includes tips for multi-city campaigns and handling Google Maps' infinite scroll. Ends with a link to the Google Maps use-case page.
Is Web Scraping Legal in 2026? A Practical Guide
Authoritative legal overview covering: hiQ v. LinkedIn (public data scraping is generally lawful in the US), CFAA limits (don't bypass authentication mechanisms you weren't authorized to access), GDPR considerations for EU personal data, robots.txt (advisory, not legally binding), and ToS clauses (civil risk, not criminal). Structured as a decision tree: if you're scraping public data → generally fine; if behind a login → only safe if it's your own session; if personal data of EU citizens → add anonymization. Positions Stekpad's browser-native model as the lowest-risk technical approach.
Recipes: Pre-Built Scraping Patterns
A catalogue of community-maintained and Stekpad-official scraping recipes. Each recipe includes: target site, fields extracted, expected schema, notes on auth requirements, and a "load this recipe" button. Organized by use-case category (leads, pricing, research, monitoring). Helps first-time users skip schema design entirely.
Amazon price tracker and review scraper. No API needed.
Stekpad tracks Amazon price changes and sends them to your Sheet on a schedule — click once on a price, and the extension learns the pattern for every product page, whether you're monitoring one product or building a full competitor pricing dataset.
Ready to stop fighting your scraper?
Free forever for personal use. Pro at €12/month or €99 lifetime.