pull every review for any company on Trustpilot — yours or a competitor. score the whole corpus with AI, surface the actual complaints behind the TrustScore, and catch review-bombing before it tanks your conversion rate.
Trustpilot gives you a single 0-5 TrustScore based on volume, recency, and star averages. It does not tell you what reviewers are actually saying. It does not surface a trend direction. It does not flag whether your responseRate is hurting you. It does not give you a per-complaint breakdown you can route to product or ops.
For your own brand, that gap is annoying. For competitor research, it's a blocker — there's no way to see structurally what their customers complain about without reading every review by hand.
The scraper closes that gap: it pulls every review for any company on Trustpilot — yours or theirs — and runs LLM sentiment across the whole corpus. You get back an eight-field structured report in 90 seconds for any company, at $0.03 per review.
point the scraper at any Trustpilot company URL or domain. comes back with every review plus an eight-field AI sentiment report per company. runs on Apify, $0.03 per review.
trustpilot.com/review/yourcompany.com) or just domain names. optional dateRange filter (30days / 90days / 365days / all) narrows the time window. each company is capped at 5,000 reviews per run.dateRange: "30days" to catch review-bombing or sudden trend reversals early. run on 3-10 competitors quarterly to find weakness patterns to position against. push the output to Google Sheets, Slack alerts, HubSpot custom fields, or your BI stack via the Apify API or n8n / Zapier — same pattern as the rest of the data-runner.dev catalog.eight structured fields per company, plus every individual review, plus the company response if it exists. all downloadable as JSON, CSV, or Excel.
sentimentScore (1-10) · overall sentiment derived from review words, not from TrustScore.
reputationRisk (1-10) · how risky is this company's current public reputation. 7+ = act now.
responseRateScore (1-10) · how actively the company responds to negative reviews. Low score = ignored complaints = compounding risk.
trendDirection · improving / declining / stable based on the last 90 days of reviews.
ratingTrend · rising / falling / flat based on star average drift.
topComplaints · 5 recurring complaint themes, ranked by frequency.
topPraises · 5 recurring praise themes, ranked by frequency.
executiveSummary · 2-3 sentence summary of what's driving the score, in plain English. paste it directly into a quarterly board update.
a B2B SaaS gets review-bombed after a botched price change. their TrustScore drops from 4.1 to 3.6 in a week but they don't see it for 10 days. running the scraper weekly with dateRange: "30days" would have flagged trendDirection: declining + reputationRisk: 8.2 on day 2 with the top complaint cluster ("unexpected price increase, no warning"). Slack alert routed to the CEO. crisis comms shipped 8 days earlier. demo signup loss halved.
run frequency: weekly · cost: ~$3-15/run depending on review volume
an ecommerce brand is launching a new product line and wants to know what customers complain about in the incumbent space. they run the scraper on 8 competitor Trustpilot URLs once. topComplaints across all 8 surfaces the same three themes: slow shipping, no real customer support, returns hell. the brand builds their landing page positioning around exactly those three: "ships in 24h, real human support, no-questions returns." cost: ~$45 for the entire competitor analysis. payback: 1 conversion.
run frequency: quarterly · cost: $30-80 depending on competitor count
a buyer is evaluating a $4M ecommerce acquisition. the seller's pitch: "Trustpilot 4.4, growing." running the scraper with dateRange: "all" surfaces a ratingTrend: falling over the last 18 months despite the 4.4 average, with reputationRisk: 6.8 driven by a complaint cluster around "product quality declining since 2024 product line update." the buyer renegotiates 18% off and writes a 90-day post-close quality clause. for cross-platform validation, the catalog also includes a Google Maps Review Sentiment scraper that runs the same playbook against any local-business listing.
run frequency: per-deal · cost: $15-50 depending on review volume
honest comparison against other Trustpilot scrapers on Apify and direct-to-Trustpilot review monitoring tools. checked their public listings as of 2026.
| data-runner.dev | Apify alternatives | Trustpilot Business | |
|---|---|---|---|
| AI sentiment included | ✓ 8 fields per company | raw reviews only | basic charts, no AI |
| Cost per review analyzed | $0.03 | $0.005-0.02 (no AI) | $259+/mo subscription |
| Competitor analysis | ✓ any company | ✓ if scraping works | only your own profile |
| Multi-language reviews | ✓ scored in original | raw text only | basic translation |
| Review-bombing detection | ✓ trendDirection field | manual analysis only | limited alerts |
| Verified-flag preserved | ✓ isVerified per review | typically yes | ✓ |
| Export formats | JSON / CSV / Excel / API | varies | CSV only on higher tier |
no subscription. no minimum. pricing scales linearly with review volume per company.
includes every review (text, rating, date, isVerified, company response) + the full 8-field AI sentiment report per company. example: a SaaS competitor with 500 reviews ≈ $15. empty profiles cost nothing.
new to Apify? you get $5 in free credits on signup — that's ~166 reviews analyzed before you spend a cent.
run on Apify →how it works, what it costs, what's legal, and how it handles edge cases.
The scraper pulls every review on the company's Trustpilot profile — review text, star rating, date, verified-purchase flag, and the company's response — then sends the full corpus to an LLM that analyzes the batch. You get back a per-company report with eight fields: an overall sentimentScore (1–10), a reputationRisk score (1–10), a responseRateScore (1–10) measuring how actively the company replies, a trendDirection (improving / declining / stable) and ratingTrend (rising / falling / flat), the top 5 recurring complaint themes, the top 5 recurring praise themes, and a 2–3 sentence executive summary. Sentiment is computed from the actual review words across the whole corpus, not from Trustpilot's TrustScore.
Yes — it's one of the primary use cases. Drop in 3-10 competitor Trustpilot URLs, run once, and the output includes the top complaints cluster per company so you see exactly what their customers complain about. Most SaaS, ecommerce, and service operators run this once per quarter and use the output to position landing pages, sales decks, and ad copy against the gap. Cost scales at $0.03 per review analyzed — typical SaaS competitor with 500 reviews costs $15.00 per analysis.
Yes — every review carries an isVerified flag in the output so you can filter or weight verified reviews separately if you want. The default analysis includes both verified and unverified reviews because the public TrustScore Trustpilot displays also includes both. If you want a stricter analysis, filter to isVerified === true in your downstream pipeline before scoring trends.
Yes — Trustpilot has profiles in 20+ countries and the scraper analyzes reviews in their original language (English, Spanish, Portuguese, French, German, Italian, Dutch, Japanese, and more). No translation step, so nuance is preserved. The dateRange filter (30days / 90days / 365days / all) lets you narrow to recent reviews if you're tracking a sentiment shift in a specific market.
The analysis is tuned specifically for online reviews and produces consistent, repeatable scores. Because the AI sees every review for a company as one batch — not each review in isolation — it picks up patterns and recurring themes that single-review scoring misses. Results align closely with how a human reviewer would summarize the same set of reviews, except it scales to 1,000+ reviews in seconds and gives you a structured output you can pipe into a dashboard.
Run the scraper weekly on your own Trustpilot URL with dateRange: '30days' and watch the trendDirection field. If it flips from 'stable' to 'declining' with reputationRisk crossing 7+ inside one week, set a Slack alert. Combine with the top complaints cluster to see if the bombing has a real cause (legit product issue surfaced) or a coordinated pattern (same complaint text in multiple reviews). Trustpilot will eventually act on coordinated fraud, but you want to know first.
No hard cap on companies per run — pass a list of Trustpilot URLs and each one is analyzed and pushed as a separate dataset item. Each company is capped at 5,000 reviews scraped per run. Pricing scales linearly at $0.03 per review: a company with 200 reviews ≈ $6.00, with 1,000 reviews ≈ $30.00. If a company has zero reviews, you pay nothing for that result.
$0.03 per review. structured AI report per company. pay only for what you analyze.