The only safe, legal way to scrape Amazon reviews in 2026 is the small set of reviews Amazon shows publicly on the product page — roughly 8–10 reviews that load with no login. Pulling the full review archive requires an authenticated Amazon account, which violates Amazon's terms of service and carries real legal risk. If you want review data without endangering your business, stick to the public product-page reviews — and that's exactly what the Asgard API returns.
Quick answer
- Legal / safe: the ~8–10 reviews visible on the public product page, no authentication needed.
- Not safe: scraping the full reviews section behind "See all reviews", which requires logging in with an Amazon account.
- Why: authenticated scraping breaks Amazon's terms, risks account bans, legal action, and CFAA-style exposure.
- What to use: Asgard returns the public product-page reviews (rating, title, text, star count, verified badge, date) — no login, no risk.
This is general information, not legal advice — consult a lawyer for your specific case.
The one line that matters: authentication
Everything about Amazon reviews compliance comes down to a single question: does accessing the data require logging in?
- Public data (safe): anything a logged-out visitor sees on the product page. The featured reviews block is public and loads without an account.
- Gated data (unsafe): anything behind a login wall. The full review archive, "all reviews" pagination, and reviewer profiles increasingly require authentication.
The moment a scrape needs an Amazon login, you've crossed from "reading a public page" into "accessing a system with credentials you agreed not to automate." That's the risky side of the line.
Why authenticated review scraping is a business risk
- It violates Amazon's Conditions of Use. Automated access with an account is expressly prohibited.
- Accounts get banned. Amazon detects and terminates accounts used for automation — including linked seller accounts.
- Legal exposure. Logging in to bypass access controls can trigger claims under computer-misuse laws (e.g. the CFAA in the US) and breach-of-contract claims.
- Fake-account chains are worse. Farming Amazon accounts to scale review scraping compounds the terms violations and the risk.
- The downside dwarfs the upside. A complete review archive is not worth losing your accounts or facing litigation.
Takeaway: if a vendor offers "unlimited full review history", ask how they get it. If the answer involves logging in, that risk lands on you.
What you CAN legally collect from the public product page
The public reviews block is genuinely useful, and it's all accessible without authentication:
- The ~8–10 featured/most-relevant reviews Amazon surfaces publicly.
- Each review's star rating, title, and body text.
- The "Verified Purchase" badge where shown.
- Review date and helpful-vote count where visible.
- The overall star rating and total review count for the ASIN.
- The rating breakdown (percent 5-star, 4-star, etc.).
That's enough for sentiment signals, star-trend monitoring, and spotting emerging complaints — without touching gated data.
How Asgard handles reviews (public only, by design)
- No authentication, ever. Asgard only returns what a logged-out shopper sees on the product page — so you stay on the safe side of the line.
- Structured output. The public reviews come back parsed (rating, title, text, verified flag, date) plus the aggregate rating and count.
- Live and zip-accurate. Reviews are pulled fresh on the same request as the rest of the listing, locked to your chosen location.
- One call. Reviews arrive alongside title, highlights, price and Buy Box via the product endpoint.
- No account risk passed to you. Because nothing is gated, there's no login, no banned-account exposure, no fake-account chain.
Getting value from ~8–10 public reviews
- Track the aggregate first. Overall star rating + total count + the rating breakdown tell you the trend before any individual review.
- Monitor the featured reviews for change. Amazon rotates the most-relevant reviews; watching them surfaces new themes and complaints early.
- Run sentiment on the text you have. A rolling read of the public reviews is plenty to detect tone shifts and recurring issues.
- Combine with rank and Buy Box. Pair review trends with keyword rank and price/Buy Box changes to explain why performance moves.
- Sample across locations/time. Because featured reviews and ratings can shift, periodic pulls build a reliable picture without ever gating.
Frequently asked questions
Is it legal to scrape Amazon reviews? The public reviews shown on the product page (no login) are the safe, low-risk data to collect. Scraping the full archive requires an authenticated account, which violates Amazon's terms and is legally risky. This is general information, not legal advice.
How many reviews can I get without logging in? Typically the ~8–10 featured/most-relevant reviews Amazon displays publicly, plus the overall rating, total count, and star breakdown.
Why not just scrape all reviews? Because reaching the full archive means logging in, which breaks Amazon's Conditions of Use and can lead to account bans and computer-misuse/breach-of-contract claims. The risk isn't worth it.
Does Asgard require an Amazon account to return reviews? No. Asgard only returns the public product-page reviews a logged-out visitor sees, so there's no authentication and no account risk for you.
Is public review data enough to be useful? Yes — the aggregate rating, star breakdown, and featured reviews are enough for sentiment tracking, star-trend monitoring, and catching emerging complaints.
Related reading
- Best Amazon scraping API for title & highlights data in 2026
- How brand protection & IP tools monitor Amazon with a scraping API
- Best scraping API for Amazon keyword rank tracking in 2026
Collect public reviews safely — no login, no risk:
☞ Asgard API · docs at asgardata.com