Capterra Reviews Scraper

Scrape every review from any Capterra software listing: overall and sub-ratings, pros and cons, reviewer job title, industry and company size, review dates, and verification details. Each review becomes one item in the Apify dataset.

Features

• Flexible input: product URLs, ID/slug pairs, or plain product names resolved through Capterra search.
• Full review text: the text is read from the page's embedded data, so it isn't cut off by "Continue reading".
• Fields that aren't shown on the page: company size, reviewer validation status and method, and review source codes.
• Automatic pagination by following Capterra's next-page link.
• Resilient: waits out Cloudflare challenges, rotates proxy sessions, retries failed pages, de-duplicates reviews, and falls back to parsing the rendered HTML if the embedded data changes.
• Limits on reviews or pages per product.

Input

Field	Type	Default	Description
products	array of strings	required	https://www.capterra.com/p/164283/Zendesk/, 164283/Zendesk, or zendesk
maxReviewsPerProduct	integer	0 (no limit)	Stop after N reviews per product
maxPagesPerProduct	integer	0 (no limit)	Stop after N pages (25 reviews each) per product
maxConcurrency	integer	3	Parallel browser pages
maxRequestRetries	integer	8	Retries per page, each with a fresh proxy session
proxyConfiguration	object	{ "useApifyProxy": true }	Apify Proxy settings. Residential proxies give the best results against Cloudflare.

Example:

{
  "products": [
    "https://www.capterra.com/p/164283/Zendesk/",
    "asana"
  ],
  "maxReviewsPerProduct": 500,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Output

One dataset item per review:

{
  "reviewId": "Capterra___7070758",
  "reviewUrl": "https://www.capterra.com/p/164283/Zendesk/#Capterra___7070758",
  "productId": "164283",
  "productSlug": "Zendesk",
  "productName": "Zendesk Suite",
  "productUrl": "https://www.capterra.com/p/164283/Zendesk/",
  "title": "Getting started with Zendesk from scratch",
  "date": "December 29, 2025",
  "dateIso": "2025-12-29",
  "summary": "As a self-taught Zendesk Admin, having an engaged community ...",
  "pros": "Excellent customer support platform, making it easy to centralize service channels ...",
  "cons": "Many simple but useful features are left out in order to focus on new products ...",
  "reasonsForChoosing": "Support history, native channels, integrations",
  "reasonsForSwitching": null,
  "adviceToOthers": null,
  "alternativesConsidered": ["Jira", "Movidesk"],
  "switchedFrom": null,
  "vendorResponse": null,
  "rating": {
    "overall": 4,
    "easeOfUse": 4,
    "customerService": 5,
    "features": 4,
    "valueForMoney": 3,
    "likelihoodToRecommend": 8
  },
  "reviewer": {
    "name": "Leonardo S.",
    "jobTitle": "Support Analyst",
    "industry": "Information Technology and Services",
    "companySize": "201-500 employees",
    "timeUsedProduct": "2+ years",
    "profilePicUrl": null,
    "linkedInVerified": false,
    "anonymous": false
  },
  "isVerified": true,
  "verificationMethods": ["ProofOfLink"],
  "incentivized": "VendorReferredIncentivized",
  "reviewSource": {
    "code": "VRI",
    "description": "Vendor Referred - Incentive Offered: This reviewer was invited by the software vendor ..."
  },
  "sourceSite": "Capterra",
  "page": 1,
  "scrapedAt": "2026-09-15T12:00:00.000Z",
  "dataSource": "payload"
}

Missing values are always null. rating.likelihoodToRecommend uses a 0–10 scale; the other ratings use 1–5.

dataSource is payload when the data came from the page's embedded JSON (all fields available). It is dom when the scraper fell back to the rendered HTML, where reviewId, companySize and isVerified are null.

Pages that still fail after all retries are saved as items with "#error": true, the URL and the error message.

Limitations

• Capterra serves at most 100 review pages (2,500 reviews) per product through its public pages. Later pages redirect to page 1, and the scraper stops there.
• Capterra is protected by Cloudflare. Runs without a proxy, or with datacenter proxies, may be blocked. Use Apify residential proxies for reliable results.

Run locally

npm install
npx playwright install chromium
mkdir -p storage/key_value_stores/default
echo '{"products":["164283/Zendesk"],"maxPagesPerProduct":2}' > storage/key_value_stores/default/INPUT.json
npm start

Results are written to storage/datasets/default/.

Tech

Node.js 20, Crawlee PlaywrightCrawler, and the Apify SDK. Capterra returns a Cloudflare 403 challenge to plain HTTP clients, so CheerioCrawler doesn't work and a real browser is required.