Scrape Amazon Product Data to CSV (Free, No Code)

To get Amazon product listings into a spreadsheet without code, open the search or category page you want, let a pattern-detecting browser extension capture the repeating product grid — title, price, rating, review count, link — and export the rows as CSV. The file opens directly in Excel or Google Sheets. Keep it small-scale and personal: a few pages of results for your own research, collected at a polite, human pace.

This guide walks through the responsible way to do that: what Amazon's rules say, the exact capture workflow, which fields are realistically available, how to analyze the CSV, and the limitations you should know about before you start.

Is it OK to scrape Amazon at all?#

Only within narrow limits. Amazon's Conditions of Use restrict automated data collection from its sites, so the question is not just can you capture a results page, but how much and for what. Reading a handful of public pages you are already browsing, for your own price research or to study the niche your own listings compete in, sits at the low-risk end. Industrial harvesting, resale of the data, or anything that looks like a bot hammering the site sits at the other end — and Amazon actively defends against it.

Three sensible boundaries:

• Scale. Capturing the three pages of results you were going to read anyway is very different from sweeping entire categories nightly. Stay on the human side of that line.
• Data type. Stick to public, factual fields: titles, prices, ratings, review counts, links. Reviewer names, profiles, and review text bring personal data and copyright questions into play — leave them out.
• Purpose. Personal price research, comparing options before you buy, or studying how your own listings stack up against the page they appear on are reasonable uses. Building a commercial dataset is not what this workflow is for.

For commercial or large-scale needs, the right tool is an official route: the Amazon Product Advertising API for programmatic catalog data, Seller Central reports (Business Reports, Brand Analytics) if you sell on Amazon, or third-party market-intelligence and price-history services that license data properly. Those exist precisely so you do not have to scrape at scale.

For the broader legal picture — public data, terms of service, the major US cases, and GDPR — read Is Web Scraping Legal?. None of this is legal advice.

Step-by-step: export Amazon search results to CSV#

To export Amazon search results to CSV, open the results page, run a no-code pattern detector to capture the product grid, keep the columns you need, let it follow the "Next" pagination for a few pages, and download the file. An Amazon results page is one product card repeated dozens of times — exactly the structure a detector learns from a single example.

Here is the workflow with a pattern-detecting extension like Calquer:

1. Open the page you want. A search results page, a category page, or the results for a keyword your own product targets. Apply Amazon's own filters first (department, price range, Prime) so the capture starts from the slice you actually care about.
2. Run the detector. Open the extension and let it find the largest repeating group on the page — the product grid. A good detector proposes the columns automatically: title, price, rating, review count, badge, product URL, and image, with no field-by-field clicking.
3. Toggle and rename columns. Switch off what you don't need and rename the rest in plain English: Title, Price, Rating, Reviews, Badge, URL. Keep the Badge column even if you think you won't need it — it is how you separate sponsored placements later.
4. Collect across pagination. Start the run and let it follow the "Next" link page by page. This works while the pagination stays on the same amazon.* origin — which it does for normal search results, since each "Next" is a full page reload on the same site. Stop after a handful of pages; relevance drops fast and politeness matters.
5. Export. Review the rows, then download as CSV — or JSON or TSV — or copy the table straight to your clipboard for Excel or Google Sheets.

The pagination mechanics here are the same as on any paginated site — full-reload "Next" links on one origin. For the deeper version, including infinite scroll, see Scrape Infinite Scroll and Pagination to CSV. And if you want the general method behind all of this, start with Export Any Website to CSV Without Code.

What fields can you actually get from a results page?#

A results page gives you breadth: seven or eight factual fields for every product in the grid. A product detail page gives you depth: many more fields, but for one product at a time. Knowing which field lives where saves you from hunting for data that simply is not on the page you are capturing.

Here is what is realistically available and how to handle each field:

Field	Where it lives	Tips
Title	Results page + detail page	Grid view truncates long titles; list view shows more of them
Price	Results page + detail page	Capture the current price; strike-through "was" prices may come along — keep them in a separate column or drop them
Rating	Results page	Arrives as text like "4.5 out of 5 stars" — clean to the number 4.5 in your sheet
Review count	Results page	Strip thousands separators (1,234 → 1234) so the column sorts as numbers
Badge	Results page	"Best Seller", "Amazon's Choice", "Sponsored" — your filter column
Product URL	Results page	Contains the ASIN (/dp/ASIN); useful as a unique key per product
Image URL	Results page	Thumbnail-sized, fine for reference, not full resolution
Feature bullets, brand, availability, variations, seller name	Detail page only	One page per product — capture these manually for a shortlist, not in bulk

The practical pattern: build your dataset from the results page, then use it to shortlist. Sort the CSV, pick the five or ten products worth a closer look, and open those detail pages yourself. Bulk-capturing detail pages multiplies your page loads for fields you will mostly never use — the opposite of small-scale.

One field to deliberately skip: anything about reviewers. Names, profile links, and review bodies are personal data and creative content respectively. Review counts and star ratings are aggregate facts; that is the line.

How do you analyze Amazon product data in Excel or Sheets?#

Open the CSV, clean the numeric columns, and three simple analyses answer most research questions: price bands show where the market clusters, a rating-versus-reviews scatter shows which products are established versus emerging, and keyword frequency in titles shows the language successful listings use. This is where Amazon product data to Excel stops being a download and starts being research.

First, clean the two numeric columns. In Google Sheets, =VALUE(SUBSTITUTE(A2,"$","")) turns $129.99 into 129.99, and =VALUE(SUBSTITUTE(B2,",","")) fixes review counts. In Excel, Power Query's Replace Values plus Change Type does both during import. Then:

• Price bands. Build a histogram of the Price column (Insert → Chart → Histogram in Sheets, or FREQUENCY with band edges like 0–10, 10–20, 20–35). For buyers, the cluster shows what "normal" costs so you can spot genuine outliers. For sellers, a thin band between two crowded ones is a positioning gap worth investigating.
• Rating vs. review count scatter. Plot Reviews on the x-axis and Rating on the y-axis. Top-right products are established incumbents. High rating with few reviews often means a newer product gaining traction — the interesting quadrant if you sell in the niche. Low rating with many reviews flags demand that current products satisfy poorly.
• Keyword frequency in titles. Split titles into words (SPLIT in Sheets, Text to Columns in Excel), stack them into one column, and count with COUNTIF. The terms that repeat across the first pages of results are the vocabulary of the niche — useful both for search and for writing your own listing titles.
• Sponsored share. Filter the Badge column to count how many of the captured rows are sponsored placements. A results page where a third of the visible products are ads reads very differently from one that is mostly organic.

Two natural follow-ups use the same capture, repeated. Re-running it on a cadence and appending dated rows turns the file into a price history — the full workflow is in Track Product Prices in a Spreadsheet (No Code). And if you sell, pointing the same columns at the listings you compete against is the basis of monitoring competitor prices without code.

What are the honest limitations?#

A no-code capture of Amazon has real limits: bot checks can appear, layouts differ across country sites, sponsored results mix into the grid, and the prices you capture are the prices you see. None of these break the workflow for small personal research, but pretending they don't exist leads to bad data and bad behavior.

• Bot checks and CAPTCHAs. If Amazon shows a verification page, stop the capture. It means your pace or volume crossed a line. Solve the check as a human, go back to normal browsing, and come back later with fewer pages and longer pauses. Never try to automate around a CAPTCHA — that crosses from "reading a page" into circumventing a technical barrier.
• Layouts vary by country site. amazon.com, amazon.de, and amazon.co.jp render the grid differently and format prices and review counts differently. The capture works per site, but re-check your columns when you switch domains rather than assuming the same setup carries over.
• Sponsored placements are mixed in. Ads sit inside the organic grid and inflate counts if you ignore them. This is why the Badge column earns its place: filter rows containing "Sponsored" out — or analyze them separately, which is its own insight.
• Prices are personalized. What you capture can depend on your login, location, Prime status, and active deals. Your CSV is a snapshot of your view, which is exactly right for personal price research and roughly right — not authoritative — for market analysis. For canonical historical pricing, price-history services and the official APIs are the better source.
• Pagination has a boundary. Cross-page collection continues while "Next" keeps you on the same amazon.* origin. If a click ever lands you somewhere else, the run ends there — by design.

One thing that is not a limitation: privacy. A capture that runs 100% on-device, like Calquer's, reads only the page already rendered in your browser and writes the file locally. Nothing you collect is uploaded anywhere, which keeps your research between you and your spreadsheet.

FAQ#

Is scraping Amazon legal?#

It is nuanced. Reading public, factual data you can already see — prices, titles, ratings — is generally the lowest-risk category, and US case law treats public-page scraping as unlikely to violate computer-access law. But Amazon's Conditions of Use restrict automated collection, and violating terms can still mean a contract claim, blocks, or a closed account. Keep it small-scale and personal, collect no personal data, stop at any bot check, and use the Product Advertising API or Seller Central reports for anything commercial. See our plain-English legality guide — and none of this is legal advice.

Can I export Amazon search results directly to Excel?#

Yes, via CSV. Capture the results grid, export the CSV, and open it in Excel or import it through Data → From Text/CSV so Power Query can clean prices and review counts on the way in. Tools that copy the table to the clipboard let you paste straight into a sheet, and JSON or TSV exports work if you prefer those formats.

How many pages of results should I capture?#

Three to five pages covers most research questions — Amazon's ranking puts the relevant products early, and relevance drops fast after that. Fewer pages also keeps the capture polite. If you genuinely need broader coverage than a handful of pages, that is the signal you have outgrown this workflow and should look at the official API or a market-intelligence service.

Why don't my captured prices match what someone else sees?#

Because Amazon prices are personalized and time-sensitive. Login state, delivery location, Prime membership, and running deals all shift what renders on your screen, and prices change frequently regardless. Your CSV records what you saw at capture time — ideal for your own buying decisions, but always pair a Date column with the price so later comparisons stay honest.

Can I capture review text or reviewer names?#

Don't. Reviewer names and profile links are personal data, and review bodies are creative content — both carry legal weight that titles and prices do not. Stick to the aggregate facts: the star rating and the review count. They tell you most of what reviews can tell you at the research stage, without collecting anything about an identifiable person.

---

Ready to turn a results page into research? Add Calquer to your browser — Chrome, Firefox, or Edge. It auto-detects the product grid, collects across pagination, and exports CSV with everything staying 100% on your device.