Trustpilot Reviews API and MCP — Customer Ratings, Text & Author
Send a business domain, Trustpilot review URL, or company website URL and this source returns customer reviews for that business: rating, text and author, per reviewer.
- You send
- A business domain, Trustpilot URL, or company website URL
- You get back
- Reviews with rating, text and author
- Coverage
- Trustpilot businesses with a public review listing.
- Price
- 1 point per returned result, up to 300 per request — $1 per 1,000
- Billed per record actually returned, capped by max_items. A business with no reviews costs nothing.
Ways to get this data
Three approaches teams use today, and where each one runs into trouble.
| Approach | What it takes | Cost | Trade-off |
|---|---|---|---|
| Reading reviews by hand | Open the page, click through pages of reviews | Free, and very slow past a handful | Nobody reads hundreds of reviews this way |
| Building a scraper | Write it, host it, repair it when the layout moves | Engineering time | A maintenance commitment for something that isn't your product |
| NeuralVerge | One request per business | $1 per 1,000, misses not billed | Reviews are self-selected — strong opinions in both directions are over-represented |
What you get back
Every field in the example response below.
| Field | Type | Description |
|---|---|---|
partial | boolean | True if the lookup hit its time budget and stopped early — re-run with a smaller max_items for a complete result. |
business_found | boolean | Whether the business resolved on Trustpilot at all. |
total_items | number | How many reviews came back. |
reviews | array | The reviews, as Trustpilot orders them. |
reviews[].id | string | Trustpilot's internal review ID. |
reviews[].url | string | Direct URL to this review. |
reviews[].title | string | Review headline, when the reviewer gave one. |
reviews[].text | string | Full review text. |
reviews[].rating | number | Rating given by the reviewer. |
reviews[].written_date | string | When the review was written. |
reviews[].experienced_date | string | When the reviewer says the experience happened. |
reviews[].is_unprompted | boolean | Whether the review was left without an invitation from the business. |
reviews[].is_verified | boolean | Whether Trustpilot marks the review as a verified order/experience. |
reviews[].useful_count | number | How many people marked the review as useful. |
reviews[].language | string | Language code of the review text. |
reviews[].author | object | Reviewer details: id, name, country, review_count, image and profile url. |
reviews[].business_reply | object | null | The business's reply to this review, when there is one. |
proxy_traffic_bytes | number | Size of the raw page data collected to build this result. |
Example response
A real response, as returned.
{
"session_id": "11111111-1111-1111-1111-111111111111",
"kind": "trustpilot_business_reviews",
"business": "hubspot.com",
"max_items": 2,
"business_found": true,
"partial": false,
"human": "# Trustpilot business reviews\n\n- **Business:** hubspot.com\n- **Max items:** 2\n- **Records returned:** 2\n- **Proxy traffic:** 1.16 MB",
"machine": {
"total_items": 2,
"reviews": [
{
"id": "6abaecae2f9f38fb34c76ef3",
"url": "https://www.trustpilot.com/reviews/6abaecae2f9f38fb34c76ef3",
"title": "A Highly Flawed CMS with Even Worse Customer Support",
"text": "In theory, HubSpot CMS is decent, but not without its share of problems. We’ve experienced multiple instances where we make no changes to our site, but HubSpot seems to enjoy making updates for the sa",
"rating": 1,
"written_date": "2026-09-29T00:39:42.000Z",
"experienced_date": "2026-09-27T00:00:00.000Z",
"is_unprompted": true,
"is_verified": false,
"useful_count": 0,
"language": "en",
"author": {
"id": "6abaec2c00f873edfd48f3c4",
"name": "Ross",
"country": "US",
"review_count": 1,
"image": "https://user-images.trustpilot.com/6abaec2c00f873edfd48f3c4/73x73.png",
"url": "https://www.trustpilot.com/users/6abaec2c00f873edfd48f3c4"
},
"business_reply": null
},
{
"id": "6aba25a0b0c45dc9764c02e3",
"url": "https://www.trustpilot.com/reviews/6aba25a0b0c45dc9764c02e3",
"title": "To all new companies considering…",
"text": "To all new companies considering HubSpot: DON'T! Unless you're prepared to pay an outrageous amount.\n\nHubSpot is a good CRM system, but they milk their customers COMPLETELY dry. Nothing respectable ab",
"rating": 1,
"written_date": "2026-09-28T10:30:24.000Z",
"experienced_date": "2026-09-28T00:00:00.000Z",
"is_unprompted": true,
"is_verified": false,
"useful_count": 0,
"language": "en",
"author": {
"id": "5f71e406a3275caf89ab2953",
"name": "Thom",
"country": "NL",
"review_count": 7,
"image": null,
"url": "https://www.trustpilot.com/users/5f71e406a3275caf89ab2953"
},
"business_reply": null
}
],
"proxy_traffic_bytes": 1215970
},
"total_points": 2
}How to run it
The same lookup works from the app, the API, or as a tool an agent can call mid-task.
This endpoint is synchronous: the call blocks until the lookup completes and returns the reviews directly. business is required; max_items is optional (1-300, default 25). If a lookup hits its time budget before finishing, the response comes back with partial: true and whatever it collected so far — re-run with a smaller max_items for a complete result.
curl -X POST https://api.neuralverge.ai/functions/v1/run-trustpilot-reviews \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "business": "hubspot.com", "max_items": 25 }'What teams use it for
Where this field shows up in an actual workflow.
Mine the exact language customers use about a company's service.
Track how a competitor's review sentiment shifts over time.
See how companies in your space handle complaints in public.
Pull enough reviews to run real analysis instead of reading the first page.
Coverage, freshness and limits
What this source covers, how fresh it is, and where it stops.
Trustpilot businesses with a public review listing.
Read at request time.
This source returns publicly visible review information, stamped with the domain/URL it was read from and a session id you can trace.
- —Reviews are self-selected and skew towards strong opinions in both directions.
- —Businesses can influence how many reviews they collect, so volume isn't a neutral signal.
- —Businesses with few reviews return a thin, less representative sample.
Questions, answered
Yes — every Data Source endpoint, including this one, is synchronous: the lookup runs inside the request and the result comes back in the same response, no polling. If a lookup hits its time budget before finishing, the response comes back with partial: true and whatever records it collected so far — re-run with a smaller max_items to get a complete result instead of a partial one.
Related sources
Try trustpilot reviews on your own data
One request format across every source in the catalog.