Trustpilot Scraper: company reviews with ratings, replies and TrustScore

Give it one or more company domains and get one row per review: the star rating, the title, the full text, who wrote it and from which country, whether Trustpilot verified it, when it was posted, when the experience actually happened, and the company's reply if there is one. Every row also carries the company's TrustScore and total review count, so you do not need a second lookup.

One thing trips people up on the first run. Trustpilot decides the exact domain string, and you have to match it. Amazon lives at www.amazon.com, not amazon.com. Copy whatever sits after /review/ in the Trustpilot address bar and paste that.

Input	Company domains, exactly as Trustpilot writes them
Output	One row per review
Ceiling	About 200 reviews per company per run, the 10 pages Trustpilot lists without an account
Account needed	None
Price	$0.45 per 1,000 reviews, flat on every plan

🔍 What Trustpilot Scraper does

It walks a company's review pages in order, about twenty reviews to a page, and turns each one into a row. Several domains in one run are handled one after another, and the per-company cap applies to each of them separately, so three domains at maxReviews: 100 is up to 300 rows.

Alongside the review itself you get the company context repeated on every row: companyName, trustScore, starRating, totalReviews, whether the profile is claimed, and the website Trustpilot has on file. That makes a single dataset enough to rank companies and read their reviews without joining anything.

experiencedAt is the date the reviewer says the thing happened, which is often days before publishedAt. It is the field to use when you are trying to tie a run of complaints to a bad week.

Three optional filters narrow the list: star ratings, a date range and the order. Trustpilot applies them itself, so a run for one-star reviews reads one-star pages and nothing else.

📥 What you give it

{
  "companyDomains": ["www.amazon.com", "nike.com"],
  "maxReviews": 200
}

Field	Default	What it is
companyDomains	none	The list version. Domains as Trustpilot writes them: www.amazon.com, nike.com, booking.com. The Console shows www.amazon.com as an example, but that is a prefill, so an API call has to send its own.
companyDomain	none	One box instead of the list, if that suits you better. Only read when companyDomains is empty.
maxReviews	100	Per company, 1 to 5,000. Trustpilot serves about twenty per page, so this sets how many pages get loaded.
starRatings	none	Only reviews with these ratings, as a list from "1" to "5": ["1", "2"] for the complaints. Empty means every rating.
dateRange	any	last30days, last3months, last6months or last12months. Counts from the day the review went up (publishedAt).
sortBy	recency	recency is newest first, Trustpilot's own default. relevance is Trustpilot's relevance ranking.
notionConnector	none	Optional. Writes every delivered review into your Notion. Authorise a connector once under Settings, API and Integrations, MCP connectors, then pick it here.
notionParentId	none	Optional. The Notion data source id to write into. Leave it empty and the pages are created privately in your workspace.
proxyConfiguration	on	Optional network setting. Leave it alone unless you want the run to leave through a network of your own, which is used exactly as you give it.

You can also paste the company's Trustpilot link, https://www.trustpilot.com/review/www.amazon.com, and the domain is read from it. That is the safest way, because the link already has Trustpilot's exact domain, www. and all. A link with filters in it, such as ?stars=1, is refused with a BAD_INPUT row rather than run without them. Paste the plain link and use the fields above instead.

📤 What you get back

A real row from a recent run, with the review body cut short here for length:

{
  "ok": true,
  "companyName": "Amazon",
  "companyDomain": "www.amazon.com",
  "trustScore": 1.6,
  "starRating": 1.5,
  "totalReviews": 49297,
  "categories": [],
  "isClaimed": true,
  "websiteUrl": "https://www.amazon.com",
  "reviewId": "6aaf168f35c963635c9f54b2",
  "rating": 1,
  "title": "My item was delivered damage and open",
  "text": "My item was delivered damage and open. It was not resolved. ...",
  "author": "Octavian Augustus",
  "authorCountry": "US",
  "authorReviewCount": 2,
  "verified": false,
  "likes": 1,
  "publishedAt": "2026-09-20T01:11:11.000Z",
  "experiencedAt": "2026-09-19T00:00:00.000Z",
  "language": "en",
  "replyText": null,
  "replyAt": null,
  "reviewUrl": "https://www.trustpilot.com/reviews/6aaf168f35c963635c9f54b2"
}

Field	What it is
trustScore, starRating	The company's weighted score and its rounded star value. Repeated on every row from that company.
rating	This one review's stars, 1 to 5.
verified	Trustpilot's own verification label, not a judgement about the reviewer.
experiencedAt	When the reviewer says it happened, as opposed to when they wrote it.
replyText, replyAt	The company's public reply, or null when there is none.
authorReviewCount	How many reviews that account has written in total. A 1 is worth noticing.
categories	Trustpilot's category ids, and an empty list on plenty of profiles.

title, text, author, authorCountry, likes, experiencedAt and replyText all come back null when Trustpilot does not publish them. That is normal, not a fault.

🧾 Reading the output

Real reviews carry ok: true. Everything else carries ok: false and an errorCode, and none of those rows are charged.

Code	What it means
BAD_INPUT	No domain was sent, a filter value Trustpilot does not know, or a pasted link with its own filters. Nothing was loaded.
NO_RESULTS	The page loaded but there was no company there. Nine times out of ten the domain is missing its www., or has an extra one. With filters set it can also mean no review matched them, and the row's note says so.
EMPTY_REVIEW	A review with no rating and no text. Kept so your count reconciles, never billed.
BLOCKED	Trustpilot refused the page after the run kept trying. Nothing was delivered for that domain.

The Console's Overview table hides text, trustScore, verified, replyText and experiencedAt, and shows diagnostic rows as mostly blank lines. Download the dataset as JSON, CSV or Excel, or switch the table to all fields.

▶️ How to run it

1. Find the company on Trustpilot and copy whatever follows /review/ in the address bar.
2. Open Trustpilot Scraper and click Try for free.
3. Paste the domain into Company domains. Add more lines for more companies.
4. Set Max reviews per company. Try 20 first to check the shape.
5. Click Start, then download the dataset.

💰 How much does it cost?

$0.45 per 1,000 reviews. Flat on every Apify plan, no volume tiers.

You are billed per genuine review row. Empty review shells, repeated reviews inside a run and every diagnostic row are not charged. With filters set, a review that does not match them is never delivered, so it is never charged either.

💡 What people use it for

• Watching a competitor's rating and reading what changed, on a schedule.
• Grouping complaints by experiencedAt to find the week something went wrong.
• Checking how often a company replies, and how quickly, across a list of rivals.
• Pulling one-star text into a summariser to get the recurring objection in the customer's own words.

🚧 What it does not do

• Date ranges are fixed windows. The last 30 days, or the last 3, 6 or 12 months, counted from the day a review went up. For any other window, filter on publishedAt or experiencedAt afterwards.
• No reviewer profiles. You get the display name, country and their total review count, nothing more.
• Not the whole history for a big company. Trustpilot lists 10 pages, about 200 reviews, of any one view to a visitor without an account, so that is what one run gets per company. Filters are how you reach further: the newest 200 one-star reviews and the newest 200 five-star reviews are two different sets.
• A stopped walk does not always explain itself. If a run was serving reviews and then stopped early, you get the shorter set without a diagnostic row telling you why.
• The domain must be Trustpilot's exact one. Getting it wrong gives you an empty result, not a helpful correction.
• Null is common. Several fields depend on what the reviewer filled in and what Trustpilot chooses to publish.
• Rows are a snapshot. Reviews get edited and removed, and TrustScore moves.

🧭 Which reviews scraper do you need?

If you want	Use
Trustpilot reviews and TrustScore	This one
Software reviews from G2	G2 Reviews Scraper
Software reviews from Capterra	Capterra Reviews Scraper
Reviews of a physical place	Google Maps Reviews Scraper
Employee reviews instead of customer ones	Glassdoor Reviews Scraper

❓ Questions people ask

Which domain do I type? The one in Trustpilot's own URL, after /review/. If their page says www.amazon.com, then amazon.com will find nothing.

Can I get reviews from a date range? Within fixed windows, yes: set dateRange to the last 30 days or the last 3, 6 or 12 months. For any other range, filter on publishedAt or experiencedAt afterwards.

Can I run several companies at once? Yes. Put each on its own line, and remember the cap is per company.

Why is text empty on some rows? Star-only reviews exist. If there is no rating either, the row comes back as EMPTY_REVIEW and is not billed.

How do I avoid paying for the same review twice? Dedupe on reviewId, which is stable across runs.

Is this legal? These are public reviews on public company pages. They are written by identifiable people, so GDPR and similar laws apply and you want a reason for collecting them. Apify's write-up on scraping and the law is a good starting point, and we are not lawyers.

🆘 If something breaks

Open the Issues tab on the actor page. Send the exact domain string you used and the run id. The errorCode on the diagnostic row usually names the problem by itself.