Trustpilot Reviews Scraper + Insights Report

Most Trustpilot scrapers pull the page over plain HTTP. Trustpilot blocks that a lot of the time, which is why so many of them fail part way through a run or only bring back half the reviews. This one runs a real browser, gets past the bot check, and reads the reviews straight from the page data. You get the whole set, not a random slice.

You also get more than rows. Every run turns the reviews into a short report: how the rating moved over time, what people complain about most, what they praise, how often the company replies, and a side by side table when you scrape more than one company.

To try it, drop in a domain like pipedrive.com and hit Start. First results come back in a few minutes.

Trustpilot review insights report: complaint analysis, trends, and competitor comparison

The built-in insights report turns raw reviews into answers: a complaint and praise keyword analysis (the terms that show up most in low- and high-star reviews), rating trends by month, the company response rate, fake-spike detection, and a side-by-side competitor review comparison when you scrape more than one brand.

What you get

Two outputs per run.

A dataset with one row per review. Download it as JSON, CSV, Excel, or XML, or pull it through the API into your own tools.

A report in the key value store:

• REPORT is an HTML page. Open it in any browser.
• INSIGHTS is the same numbers as JSON, in case you want to process them yourself.

The report covers:

• Average rating, the full star breakdown with percentages, and how many reviews are verified.
• How often the company replies to its reviews.
• Rating by month, so trends are easy to see.
• Months where review volume jumps to three times the usual. These can point to fake or paid-for review bursts, so they are worth a look before you trust a score.
• The 15 words and phrases that show up most in 1 and 2 star reviews. The complaints.
• The 15 that show up most in 4 and 5 star reviews. The praise.
• A comparison table across companies when you scrape more than one.

How it compares

The two things that set it apart are reliability and the report.

	This actor	Typical scraper
Gets past Trustpilot's bot protection	yes, real browser	often not, plain HTTP
Brings back the full review set	yes	not always
Insights report	yes	no
Complaint and praise keywords	yes	no
Fake spike detection	yes	no
Compare several companies in one run	yes	no
Runs on the free plan	yes	usually

Common uses:

• Tracking your own brand's reputation over time.
• Benchmarking against competitors.
• Pulling product problems out of the low star reviews.
• Checking a vendor's score for manipulation before you buy.
• Client reporting at agencies.

How to use it

1. Open the Input tab.
2. Put one or more domains or Trustpilot URLs in Company URLs, for example pipedrive.com or airbnb.com.
3. Set Max reviews per company. Leave it at 0 for all of them, or set 40 to 200 for a quick test.
4. Filter by language, star rating, or date if you want.
5. Hit Start. Reviews land in the Dataset tab. The report lands in the Key value store tab under REPORT.

Input

Field	Type	Default	What it does
companyUrls	array (required)	["pipedrive.com"]	Trustpilot URLs or bare domains, one per line
maxReviewsPerCompany	integer	0 (all)	Cap per company
stars	array	all	Keep only these star ratings
languages	array	["en"]	ISO language codes like en, de, fr. Leave empty for all
sort	string	recency	recency or relevance
date	string	all time	last30days, last3months, last6months, last12months
verified	boolean	false	Only verified reviews
withReplies	boolean	false	Only reviews that got a company reply
includeCompanyInfo	boolean	true	Add trust score and category to each row
generateInsights	boolean	true	Build the report after scraping
maxConcurrency	integer	5	Parallel page fetches per company
requestDelayMs	integer	300	Wait between fetches. Raise it for politer crawling
proxyConfiguration	object	Apify datacenter proxy	Proxy settings

Example input

{
  "companyUrls": ["pipedrive.com", "hubspot.com"],
  "maxReviewsPerCompany": 200,
  "languages": ["en"],
  "generateInsights": true
}

Output

One row per review:

{
  "reviewId": "abc123",
  "companyName": "Pipedrive",
  "companyDomain": "pipedrive.com",
  "companyTrustScore": 4.4,
  "companyTotalReviews": 1872,
  "rating": 5,
  "title": "Great CRM for small teams",
  "text": "Easy to set up and the pipeline view is excellent...",
  "language": "en",
  "country": "US",
  "publishedDate": "2024-11-03T08:22:00Z",
  "isVerified": true,
  "replyMessage": null,
  "authorName": "John D.",
  "reviewUrl": "https://www.trustpilot.com/reviews/abc123"
}

Download the dataset as JSON, CSV, Excel, or XML from the Dataset tab.

Key value store:

Key	Type	What it is
INSIGHTS	application/json	The full analytics object, per company plus the comparison
REPORT	text/html	The report. Open it in a browser

Fields

Field	Type	Description
reviewId	string	Trustpilot review ID
companyName	string	Company display name
companyDomain	string	Trustpilot identifying domain
companyTrustScore	float	Trust score (0 to 5)
companyTotalReviews	integer	Total reviews on Trustpilot
rating	integer	Star rating (1 to 5)
title	string	Review title
text	string	Full review text
language	string	ISO 639-1 language code
country	string	Reviewer country code
publishedDate	ISO 8601	Review published date
experienceDate	ISO 8601	Date of experience
isVerified	boolean	Whether the review is verified
verificationLevel	string	Verification label
authorName	string	Reviewer display name
authorReviewCount	integer	Total reviews by this author
replyMessage	string	Company reply text, null if none
replyPublishedDate	ISO 8601	Company reply date
reviewUrl	string	Direct link to the review

Pricing

$0.0015 per review. No start fee.

Reviews	Cost
100	$0.15
1,000	$1.50
10,000	$15.00
100,000	$150.00

The report is always part of the run. You pay for reviews, not for the analysis. Apify's free plan gives you $5 a month in credits, which is around 3,300 reviews, and you don't need a card to start.

Tips

• To benchmark, put in 3 to 5 competitor domains, set maxReviewsPerCompany to 300, keep languages at ["en"]. The comparison table in the report gives you the side by side in a couple of minutes.
• For trends, set date to last12months and maxReviewsPerCompany to 0 to pull a full year.
• For complaint mining, look at topComplaintTerms in the INSIGHTS output to spot recurring problems fast.
• For fake review checks, suspiciousSpikeMonths in INSIGHTS flags months with three times the normal volume.
• To track reputation over time, schedule the run monthly with Apify's scheduler.

FAQ

Is scraping Trustpilot legal? Trustpilot is a public website. Scraping public, non-login data for research, competitive analysis, or your own use is generally fine in most places. Don't use this to collect personal data without a legal basis or to break Trustpilot's terms.

Why are some fields empty? Trustpilot's page layout changes by company, region, and A/B test. The actor returns null for anything it can't find instead of crashing.

I got fewer reviews than expected. Trustpilot may rate limit you. Raise requestDelayMs to 500 or 1000 and run again. The default datacenter proxy is usually enough. If you keep hitting 429s, switch on residential proxy, which needs a paid plan.

Does it work in languages other than English? Yes. Drop the languages filter to get every language. The keyword extraction has stopword lists for English and German. Other languages still work, but the keyword lists may include more filler words.

Issues or feature requests? Use the Issues tab on the actor page. Custom or white label setups are available on request.