Give it Google Shopping product links - or just the product ids - and get their reviews back as rows: who wrote it, the rating, the date, the headline, the review itself and where it came from. Seven columns, keyed on the product id rather than on your query. One thing to know before you start: Google blocks datacentre addresses here, so this scraper needs your own residential proxy.
The input is the product, not the review - you name the items you want feedback on and the job collects what shoppers wrote about each one.
Rows come back keyed on product_id, so that is the column you join your input list on - this scraper does not echo your query back.
Paste the full Google Shopping product URL if that is what you have, or the bare product id if your list is already normalised. Either goes one per line, and either can arrive as a CSV, XLSX, TXT or Parquet upload instead of being typed.
Google Shopping is regional, and the country selector decides which market's reviews you are reading. It defaults to the United States, so set it deliberately rather than inheriting it - the same product can look very different two borders away.
Google blocks free and datacentre addresses on Shopping, so this scraper runs through your own paid residential proxy. Without one a run comes back empty - and so does a product that genuinely has no reviews yet, which is why the data dictionary tells you how to tell those two apart.
Each row is one review of one product: the author, the rating, the date, the headline, the review text and the source it came from - tied to the product by its id.
The column list below is the header row of real run exports, corroborated by the column array the run UI renders - the two agree exactly. What it is not is a sample: read the note under the table before you write code against any of these fields, because it says plainly what we have and have not seen this scraper return, and it explains the one column that is missing compared with every sibling scraper here.
The names are exact - they are the workbook header row and the run UI's column array, which match. The descriptions say what each column is for. They do not describe its format, and the note under the table explains why not.
query column, so this is what ties a row back to your input list.Read this before you plan around the table above. Every run export we hold for this scraper came back empty - three runs across 2026-06-30 to 2026-07-15, each a workbook with this header row and no data row beneath it. So the seven names are solid, and nothing else here is: we publish no example values and no format for any field, not whether rating is a number or a string, not whether date is absolute or relative, not what source holds. Two more things worth planning for. There is no query column - join on product_id, and normalise your input list to ids first if you submitted links. And an empty result has two possible causes here, which the tool itself spells out: a missing residential proxy, or a product that simply has no reviews. Check the proxy first, then check the product on Google Shopping in the country you selected before concluding anything.
Two controls here rather than one: how many reviews to take per product, and which country's Google Shopping to read. Both are chosen before the job starts, alongside how you hand over the product list and the proxy the job runs through.
A few examples of how teams use shopping reviews to answer a question they actually have.
Pull every review on the products you sell and read the low-rated text together rather than one at a time. Recurring complaints show up as repeated language, which is far more actionable for a product team than an average that moved slightly.
Take the products you compete against and export the lot in one run. Ratings and verbatim text side by side show where a rival genuinely wins and where they simply have more reviews.
The same item is reviewed differently market to market. Run the product once per country and compare - the country selector is what makes that a job rather than a research project.
No subscription, no minimum, no recurring bill. Your first 500 rows are on us - after that, pay-as-you-go at the same flat rate as every other scraper here.
Every new account, one-time. No credit card required. Per-product limits, the country selector, file upload and every export format included. The residential proxy this scraper needs is yours to supply.
The same flat rate as every other scraper on the platform. The pre-flight estimator shows the row count and credit cost before a run starts - no surprise bills, no compute units to translate.
Volume pricing, dedicated workers and an SLA for continuous monitoring or very large historical pulls. Tell us your numbers and we will quote.
LIVESCRAPER10 at checkout.The same product judged on the other big shelf. Worth having next to these rows before you draw a conclusion from one marketplace's ratings.
Reviews of the seller rather than the product. A good item from a bad merchant is a pattern you only see when you read both.
Short answer: yes for the public review content - and this export does carry an author name, so plan for that.
Reviews on Google Shopping are published to be read. The rating, the headline, the text, the date and the author name are shown to anyone who opens the product page, signed in or not. Collecting publicly visible feedback for research is long-established practice, and nothing here touches a login or a paywall.
One column deserves care rather than a footnote. author identifies the person who wrote the review, which makes the file personal data under the GDPR and comparable regimes. Reading the corpus in aggregate, or quoting a review the way any shopper could, is ordinary use. Building a profile of a named reviewer across products is not, and it is not what this is for.
Google's terms restrict automated access, and Shopping in particular is protected against it - which is why the residential proxy is a requirement rather than a suggestion. That makes this a terms question as well as a technical one. We run no third-party trackers on the data layer, and your exports auto-delete after 30 days.
The questions we hear most. Anything else? Talk to us - humans, not bots, write the answers.
500 one-time free rows on every new account - no expiry. After that it is $0.002 per row, pay-as-you-go - no card on file until you say so. Bring your own residential proxy.
Livescraper's Google Shopping Reviews Scraper turns a list of products into review data. You submit Google Shopping product links or bare product ids - typed one per line, or uploaded as a CSV, XLSX, TXT or Parquet file - set how many reviews to take per product, pick the country, and download what shoppers wrote as a clean CSV, Excel or JSON file.
Each row is one review: the author, the rating, the date, the headline, the review body and the source it came from, tied to the item by its product id. That last point is the one to note before you write a parser - this scraper does not echo your query back in its own column the way the rest of our catalogue does, so product_id is the join key, and a list of URLs is worth normalising to ids before you run.
Product teams pull every review on the items they sell and read the low-rated text as a corpus, where recurring complaints appear as repeated language rather than as a score that drifted. Marketing teams export a competitive shelf in one run and compare ratings and verbatim text side by side. Researchers run the same product against different countries, because Google Shopping is regional and the country selector makes that a job rather than a project.
Two requirements are stated here rather than buried. Google blocks free and datacentre addresses on Shopping, so this scraper needs your own paid residential proxy to return anything. And an empty export has two possible causes - a missing proxy, or a product with no reviews yet - which the tool itself distinguishes and this page repeats rather than flattening. This page also publishes no example values and no format for any field, because every run export we hold returned zero rows, and a value we have not seen is not a value worth printing. The seven column names are exact and cross-checked; everything else is for your first free-tier run to tell you. Start free: your first 500 rows cost nothing and need no credit card.