Capterra Software Reviews, Products & Categories Scraper

How it works

All-in-one Capterra.com scraper — paste any mix of URL kinds in the same input list and the actor auto-classifies each one:

Input	Row(s) emitted
Product URL — /p/{id}/{slug}/	1 product row: name, description, application category, rating, vendor, pricing, features, screenshots, alternatives, review snippets, awards & badges (Best Value, Shortlist, Ease of Use), structured integrations (with logos + Capterra URLs), per-sub-rating averages with reviewer counts, Capterra-computed sentiment breakdown
Reviews URL — /p/{id}/{slug}/reviews/	N review rows — one per individual review (rowType: "review"). Each row carries a nested productAggregate object with the thread-level rating distribution, sub-rating averages, total reviews, etc. No separate summary row.
Category URL — /{slug}-software/ (e.g. /project-management-software/)	N category-product rows — one per product in the category (rowType: "category-product"). Each row carries a nested categoryAggregate with the category name, description, total count, and related categories. Same flat-with-nested-aggregate pattern as reviews.
Compare URL — /compare/{ids}/{slug-a}-vs-{slug-b}/	1 compare row: side-by-side products with ratings, pricing, deployments

Bare numeric product IDs (137005) and bare category slugs (project-management) also work. JSON + CSV.

Reviews are flat with the aggregate denormalized into every row. Every dataset row from a reviews URL is a single review. The thread-level aggregate (rating distribution, sub-rating averages, total reviews) lives under each row's productAggregate field — duplicated on every review row from the same URL, but that keeps each row self-contained and makes billing trivial: one row = one paid outputrecord event.

Pure HTTP. No Puppeteer, no Playwright, no headless Chromium, no third-party Cloudflare-bypass service.

---

Capterra is Cloudflare-gated with stronger deep-page protection than most B2B-software sites. The actor breaks past it without a browser via a session-aware HTTP pattern:

1. Warm session. Open one sticky-IP residential proxy session, hit a known-good product page, retry until 200. Cloudflare sets __cf_bm + _cfuvid cookies on success — captured automatically by the impit cookie jar.
2. Iterate URLs. Reuse the same Impit instance + sticky IP for every subsequent URL. Cookie + IP persistence pushes per-URL pass rate from a cold ~25% to a warm ~80%, with per-URL retry bringing cumulative success to ~95%.
3. Auto-refresh. On 3 consecutive 403s, drop the session and re-warm with a fresh IP. No human intervention needed.

Every page is then parsed via JSON-LD (the rock-solid SoftwareApplication schema + BreadcrumbList) plus targeted Cheerio walks of the visible HTML for sections (Features, Pricing, Integrations, Support, Alternatives, User reviews). No DOM execution; no DOMContentLoaded wait.

---

Input

Field	Type	Required	Notes
startUrls	string[]	yes	Any mix of product / reviews / category / compare URLs (or bare product IDs / category slugs). Auto-classified.
maxItems	integer	no	Safety cap on total dataset rows, not URLs. A reviews URL with 100 reviews emits 100 review rows; a category URL with 50 products emits 50 category-product rows. Default 1000. Free-tier users are capped at 100.
maxReviewsPerProduct	integer	no	Max individual review rows emitted per reviews URL. Default 100, 0 = no limit.
maxProductsPerCategory	integer	no	Max individual category-product rows emitted per category URL. Default 50, 0 = no limit.
maxRequestRetries	integer	no	Per-URL retry budget inside the warm session. Default 8.
proxy	object	no	Apify residential proxy recommended (US country). Required because Capterra rejects datacenter IPs.

Example input

{
  "startUrls": [
    "https://www.capterra.com/p/137005/SmartPM/",
    "https://www.capterra.com/p/137005/SmartPM/reviews/",
    "https://www.capterra.com/project-management-software/",
    "https://www.capterra.com/compare/137005-229614/SmartPM-vs-Notion/"
  ],
  "maxItems": 50,
  "maxReviewsPerProduct": 100,
  "maxProductsPerCategory": 50,
  "proxy": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"], "apifyProxyCountry": "US" }
}

That yields 90 rows on SmartPM: 1 product + 46 individual review rows (each with productAggregate nested) + 42 category-product rows from /project-management-software/ (each with categoryAggregate nested) + 1 compare. Use maxProductsPerCategory / maxReviewsPerProduct to cap the per-URL fan-out.

---

Output schema

Four row shapes, distinguished by the rowType field: product, review, category-product, compare.

rowType: "product"

{
  "rowType":             "product",
  "productUrl":          "https://www.capterra.com/p/137005/SmartPM/",
  "productId":           "137005",
  "name":                "SmartPM",
  "description":         "SmartPM is an AI-powered Automated Project Controls™ platform built by construction for construction…",
  "applicationCategory": "BusinessApplication",
  "categories":          ["Construction Software", "Project Management Software"],
  "rating": {
    "average":           4.9,
    "count":             46,
    "bestRating":        5
  },
  "vendor": {
    "name":              "SmartPM",
    "websiteUrl":        null
  },
  "pricingStarts":       "$49",
  "logoUrl":             "https://gdm-catalog-fmapi-prod.imgix.net/ProductLogo/7c88b5d2-…",
  "screenshotUrls":      [ /* up to ~22 imgix URLs */ ],
  "features":            [ /* visible feature list */ ],
  "deployments":         [ "Cloud", "Web-Based" ],
  "trainingOptions":     [ "Live Online", "Documentation", "Webinars" ],
  "supportOptions":      [ "Email/Help Desk", "Phone Support", "Chat" ],
  "customerTypes":       [ "Small Business", "Mid-size", "Enterprise" ],
  "reviewSnippets": [
    {
      "title":           "SmartPM, an amazing tool for schedule analysis",
      "pros":            "I had a 2 year experience using SmartPM…",
      "cons":            "Some advanced features have a learning curve…",
      "overallRating":   null,
      "reviewer":        { "name": null, "jobTitle": null, "industry": null, "companySize": null },
      "source":          "capterra"
    }
    /* up to 10 surfaced on the product page */
  ],
  "alternatives": [
    { "productId": "183213", "productUrl": "https://www.capterra.com/p/183213/Contractor-Foreman/", "name": "Contractor Foreman", "rating": { "average": null, "count": null, "bestRating": null }, "vendor": null, "logoUrl": null, "pricingStarts": null, "sponsored": null }
    /* + ~4 more */
  ],
  "breadcrumbs": [
    { "name": "Capterra", "url": "https://www.capterra.com/" },
    { "name": "Construction Software", "url": "https://www.capterra.com/construction-software/" },
    { "name": "SmartPM", "url": "https://www.capterra.com/p/137005/SmartPM/" }
  ],

  // ── RSC-derived rich data (from Capterra's embedded product JSON) ──
  "awards": [
    {
      "title":           "2025 Capterra Best Value for Money",
      "badgeUrl":        "/public-bx-capterra-v0/badge-best-value.png",
      "badgeAlt":        "Best Value Badge",
      "additionalInfo":  "and in",
      "linkText":        "+4 categories",
      "linkAction":      "best-value"
    }
  ],
  "valueBadges": [
    { "name": "Business Intelligence",   "year": "2025", "categorySlug": "business-intelligence-software" },
    { "name": "Construction Management", "year": "2025", "categorySlug": "construction-management-software" },
    { "name": "Data Analysis",           "year": "2025", "categorySlug": "data-analysis-software" },
    { "name": "Project Planning",        "year": "2025", "categorySlug": "project-planning-software" }
  ],
  "shortlistBadges":   [],                                  // populated when product is in Capterra Shortlist reports
  "easeOfUseBadges":   [],                                  // populated when product holds "Best Ease of Use" awards
  "integrations": [
    {
      "name":            "Microsoft Power BI",
      "id":              "e3df9c2d-1567-427c-acb2-a6d200b52931",
      "logo":            "https://gdm-catalog-fmapi-prod.imgix.net/ProductLogo/e48bb429-…png",
      "href":            "https://www.capterra.com/p/176586/Power-BI/",
      "reviewCount":     1878
    }
    /* + 7 more */
  ],
  "usedFor": [
    { "name": "Construction Scheduling", "slug": "construction-scheduling-software" },
    { "name": "Construction Management", "slug": "construction-management-software" }
    /* + 6 more */
  ],
  "pricing": {
    "startingPrice":         null,                          // raw value — may be blank when not disclosed
    "startingPriceCurrency": "$$",
    "startingPriceUnit":     "",
    "pricingModel":          "Other",
    "paymentFrequency":      "Per Month",
    "hasFreeTrial":          true,
    "hasFreeVersion":        false
  },
  "subRatingAverages": {                                    // Capterra's authoritative per-sub-rating means + counts
    "easeOfUse":       { "rating": 4.7, "reviews": 46 },
    "customerService": { "rating": 5,   "reviews": 45 },
    "valueForMoney":   { "rating": 4.9, "reviews": 42 }
  },
  "reviewSentiment": {                                      // Capterra-computed sentiment breakdown
    "totalReviews":  46,
    "overallRating": 4.87,
    "positive":      { "count": 96, "percentage": 96 },
    "neutral":       { "count":  4, "percentage":  4 },
    "negative":      { "count":  0, "percentage":  0 }
  },

  "scrapedAt":           "2026-05-12T13:35:00.412Z"
}

rowType: "review" (one per individual review)

Every dataset row from a reviews URL is a single review. Product info (productId, productName, reviewsUrl, productUrl, reviewUrl) and the thread-level aggregate (productAggregate) are denormalized onto every row, so each row is fully self-contained — no joining required for CSV consumers.

Per-review fields come directly from Capterra's embedded RSC JSON (one structured object per review, decoded brace-by-brace), giving 100% fill on every field the source provides: reviewer name / job title / industry / company size / profile pic / LinkedIn verification, all four sub-ratings, recommendation score (0-10), incentivization status, plus structured alternativeProducts[] and switchedProducts[] arrays for the competitor products each reviewer mentioned, plus the vendor's reply when one exists.

The actor auto-paginates through every review page (following each page's ?page=N link). productAggregate.totalReviews and productAggregate.pagesScraped reflect the full thread up to maxReviewsPerProduct. On SmartPM that's 46 reviews across 2 pages; on bigger products it can be hundreds across 10+ pages.

{
  "rowType":             "review",
  "reviewsUrl":          "https://www.capterra.com/p/137005/SmartPM/reviews/",
  "productId":           "137005",
  "productName":         "SmartPM",

  // ── Per-review fields ────────────────────────────────────────────
  "reviewId":            "Capterra___6325867",                // stable primary key
  "reviewUrl":           "https://www.capterra.com/p/137005/x/#Capterra___6325867",
  "productUrl":          "https://www.capterra.com/p/137005/x/",
  "title":               "SmartPM Elevates Company Schedule Performance & Health",
  "overallRating":       5,
  "recommendationScore": 10,                                  // 0–10 NPS, when provided
  "subRatings":          { "easeOfUse": 5, "customerService": 5, "features": 5, "valueForMoney": 5 },
  "text":                "Our SmartPM experience has been fantastic - integration, support, implementation - everything!",
  "pros":                "SmartPM has already proved itself to be invaluable to our company…",
  "cons":                "Some advanced features have a learning curve…",
  "adviceToOthers":      null,                                // Optional "Advice to others" text
  "reasonsForChoosing":  null,                                // "Reasons for Choosing" (~20% fill)
  "reasonsForSwitching": null,                                // "Reasons for Switching" (~11% fill)
  "switchedFrom":        "Deltek Acumen",                     // Name of previous tool (= switchedProducts[0].productName)
  "switchedProducts": [                                       // Full structured list (rare — 4% fill on SmartPM)
    { "productId": "10001903", "productSlug": "Deltek-Acumen", "productName": "Deltek Acumen" }
  ],
  "alternativeProducts": [                                    // Competitors the reviewer ALSO evaluated (9% fill on SmartPM)
    { "productId": "10012584", "productSlug": "Schedule-Validator", "productName": "Schedule Validator" }
  ],
  "publishedAt":         "May 31, 2024",
  "timeUsed":            "1-2 years",                         // "2+ years" / "6-12 months" / "Less than 6 months" / "1-2 years"
  "frequencyOfUse":      null,
  "reviewer": {
    "name":              "John M.",                           // null when reviewer is anonymous ("Verified User")
    "jobTitle":          "Scheduler",
    "industry":          "Construction",
    "companySize":       "201-500 employees",                 // "Self-employed" / "1-10 employees" / … / "10,001+ employees"
    "profilePicUrl":     "https://reviews.capterra.com/cdn/profile-images/linkedin/e7e3268d2b...jpeg",
    "verifiedLinkedIn":  false,
    "validationsPassed": ["ProofOfLink"]                      // Methods Capterra used to validate the reviewer
  },
  "verified":            true,                                // Capterra "isValidated" flag
  "incentivization":     "vendor-referred-incentive",         // "none" | "vendor-referred" | "nominal-gift" | "vendor-referred-incentive" | "unknown"
  "source":              "capterra:incentivized",             // back-compat string form of `incentivization`
  "reviewSourceTooltip": "Vendor Referred - Incentive Offered: This reviewer was invited by the software vendor…",
  "vendorResponse": {                                         // null when the vendor didn't reply (typical)
    "date":              "July 11, 2022",
    "text":              "Thanks Joe! It's been great working with you!"
  },

  // ── Thread-level aggregate (same on every review row from the same URL) ──
  "productAggregate": {
    "rating":            { "average": 4.9, "count": 46, "bestRating": 5 },
    "subRatingAverages": { "easeOfUse": 4.7, "customerService": 5, "features": 4.8, "valueForMoney": 4.9 },
    "ratingDistribution": [
      { "stars": 5, "count": 21, "percentage": 84 },
      { "stars": 4, "count":  2, "percentage":  8 },
      { "stars": 3, "count":  2, "percentage":  8 },
      { "stars": 2, "count":  0, "percentage":  0 },
      { "stars": 1, "count":  0, "percentage":  0 }
    ],
    "commonPros":        [],
    "commonCons":        [],
    "totalReviews":      46,
    "pagesScraped":      2
  },

  "scrapedAt":           "2026-05-12T10:55:14.123Z"
}

rowType: "category-product" (one per product listed in the category)

Same flat-with-nested-aggregate pattern as reviews. Every dataset row from a category URL is a single product listed in that category. The category-level aggregate (name, slug, description, total count, related categories) is denormalized under categoryAggregate on every row.

{
  "rowType":             "category-product",
  "categoryUrl":         "https://www.capterra.com/project-management-software/",
  "categorySlug":        "project-management",

  // ── Per-product fields (different on every row) ──────────────────
  "productId":           "120390",
  "productUrl":          "https://www.capterra.com/p/120390/Teamwork-Projects/",
  "name":                "Teamwork Projects",
  "description":         null,
  "rating":              { "average": 4.5, "count": 1234, "bestRating": 5 },
  "vendor":              null,
  "logoUrl":             "https://gdm-catalog-fmapi-prod.imgix.net/…",
  "pricingStarts":       null,
  "sponsored":           false,

  // ── Category-level aggregate (same on every category-product row from the same URL) ──
  "categoryAggregate": {
    "categoryUrl":       "https://www.capterra.com/project-management-software/",
    "categorySlug":      "project-management",
    "categoryName":      "Project Management Software",
    "description":       "Project management software helps teams plan, track, and deliver work …",
    "totalProductsOnPage": 42,
    "relatedCategories": [
      { "name": "Task Management Software", "url": "https://www.capterra.com/task-management-software/" }
      /* up to 30 */
    ]
  },

  "scrapedAt":           "2026-05-12T..."
}

rowType: "compare"

{
  "rowType":     "compare",
  "compareUrl":  "https://www.capterra.com/compare/137005-229614/SmartPM-vs-Notion/",
  "products": [
    { "productId": "137005", "name": "SmartPM", "rating": null, "vendor": null, ... },
    { "productId": "229614", "name": "Notion",  "rating": null, "vendor": null, ... }
  ],
  "summary": {
    "ratings":     [ { "average": 4.9, "count": 46, "bestRating": 5 }, { "average": 4.7, "count": 5000, "bestRating": 5 } ],
    "pricingStarts": [ "$49", "Free" ],
    "vendors":     [ "SmartPM", "Notion Labs" ],
    "deployments": [ ["Cloud"], ["Cloud", "Mobile"] ]
  }
}

Compare rows are populated by per-product fetch. Capterra's /compare/... pages are 100% client-rendered SPAs — their HTML carries no product data. Instead we read the product IDs from the URL itself and fetch each compared product's own page through the warm session, giving full ratings + vendor + pricing + deployments per side-by-side product. Cost: N extra HTTP calls per compare URL (where N = number of products compared, usually 2–3).

---

Pricing

Pay-per-event. Both reviews and category products are billed flat — every review row is one outputrecord, every category-product row is one outputrecord. Cost = rows × outputrecord-price. No nested-array arithmetic.

Event	When	Suggested rate
outputrecord	Once per dataset row pushed — product, review, category-product, compare. The thread/category aggregate is nested INSIDE each flat row, not charged separately.	configured on the actor page
additional-data	Once per nested item inside a product or compare row (features, screenshots, alternatives, breadcrumbs, compared products). Reviews and category-products are NEVER billed here — each is its own paid outputrecord row.	$0.75 per 1,000 items ($0.00075 each)

What counts as additional-data

Only the URL kinds that still keep nested arrays bill here. Reviews and category products never do.

Row type	Items billed as additional-data
Product row	every feature + screenshot + review snippet + alternative + breadcrumb category
Compare row	every compared product

---

What makes this richer than the competition

We surveyed all 16 Capterra actors on the Apify marketplace (May 2026). Most cover one URL surface only:

Capability	Competitor actors	This actor
Product + reviews + category + compare in one actor	❌ (split across 4+ SKUs)	✅
URL auto-classification (paste any mix)	❌	✅
Pure HTTP, no headless browser	mixed (some use Playwright)	✅
No paid third-party CF-bypass service	mixed	✅
Cookie-warmed sticky session pattern	❌	✅
Per-product alternatives + screenshots in product row	❌ (separate calls)	✅
Flat-per-review billing (1 review = 1 paid row, no nested-array math)	mixed	✅
Thread-level aggregate (rating distribution + sub-rating averages) on every review row	❌ (none surface this at all)	✅
RSC-stream embedded review JSON extraction — 100% fill on reviewId, companySize, recommendationScore, incentivization, etc.	partial (some fields 0%)	✅
Capterra awards & badges (Best Value, Shortlist, Ease-of-Use) per product	❌	✅
Structured integrations per product (name + logo + Capterra URL + review count, not just feature strings)	❌	✅
Capterra-computed review sentiment breakdown (positive / neutral / negative %)	❌	✅
Per-sub-rating averages with reviewer counts (more precise than mean alone)	partial	✅

The top marketplace actor (focused_vanguard/multi-platform-reviews-scraper, 220 users) covers multiple sources (Capterra + G2 + Trustpilot + Gartner + Software Advice + Reddit) but only emits review data. We do everything Capterra exposes in one actor — products, reviews, categories, comparisons — without crossing into other platforms.

---

Notes & limitations

• Capterra routes by numeric product ID — the slug in the URL is decorative. /p/137005/Slack/ redirects to whatever product 137005 actually is (currently SmartPM).
• Cloudflare gating. We rely on residential-proxy IP rotation + cookie reuse to slip past CF's WAF. With Apify residential pool (US country), expect ~95% per-URL success on a normal-cost run.
• Reviews pagination is automatic — the actor follows ?page=N links from each page's own HTML (not constructed URLs, since /x/ placeholder slugs lose query-strings on redirect). Stops when no next-page link found or maxReviewsPerProduct is hit. Each individual review is then emitted as its own flat dataset row with the thread aggregate nested in productAggregate.
• Category flattening. Same pattern as reviews — every product listed in a category becomes its own dataset row with categoryAggregate nested. A category URL with 50 products fills 50 rows of the maxItems budget (capped by maxProductsPerCategory).
• Compare extraction uses per-product fetch — Capterra's compare pages are SPAs, so we read product IDs from the URL and fetch each compared product's full page through the warm session. Result: rich compare rows with real ratings, vendor info, pricing, and deployments per product.
• No mobile app. Capterra is web-only; no API to bypass the WAF via.

---

Support

Issues, feature requests, or custom output shapes? Open a ticket on the actor's Apify page or message the maintainer.