G2 Scraper

Collect structured G2 software reviews, products, analytics, comparisons, and discussions. Filter by region, review source, response type, sentiment themes, and discount, or discover available facet values before filtering.

Why use this scraper?

• Collect all matching public, approved reviews for a product with maxReviewsPerProduct: 0.
• Structured pros/cons/benefits parsed from each review.
• Per-dimension ratings: ease of use, setup, support, security, and more.
• NLP-extracted sentiment themes (what users love/hate).
• Competitive switching intelligence (what products users switched from and why).
• Faceted filtering: narrow reviews by primary region, review source, response type, loved/hated/switching themes, and discount.
• Facet-value discovery: list the valid values for any text facet before filtering.
• Analytics mode with aggregations, percentiles, and significant terms.
• Side-by-side product comparisons in a single request.
• Q&A discussion threads per product.

Use cases

• Compare competitors by sentiment, ratings, and switching patterns.
• Feed structured review data into NLP / AI pipelines.
• Track review trends over time for market research.
• Segment reviews by region, company size, review source, or sentiment theme.
• Find products by category, vendor, or keyword search.
• Search across public, approved reviews for specific topics.

How to use

Scrape reviews for a product

{
  "products": ["https://www.g2.com/products/slack/reviews"]
}

Search for products and scrape their reviews

{
  "search": "project management",
  "maxProducts": 5,
  "maxReviewsPerProduct": 50
}

Browse a category

{
  "category": "CRM",
  "maxProducts": 10,
  "maxReviewsPerProduct": 20
}

Browse by vendor

{
  "vendor": "Salesforce",
  "maxProducts": 5,
  "maxReviewsPerProduct": 10
}

Compare products side-by-side

{
  "compare": ["slack", "clickup", "jira"]
}

Get analytics (no individual reviews)

{
  "products": ["slack"],
  "analytics": true
}

Global review search across all products

{
  "reviewSearch": "terrible customer support",
  "maxReviews": 50
}

Find similar reviews

{
  "similarTo": "The API was difficult to integrate and documentation was poor",
  "maxReviews": 20
}

Filter by date, rating, and company size

{
  "products": ["slack"],
  "dateFrom": "2025-01-01",
  "ratings": ["1", "2"],
  "companySegment": ["enterprise"],
  "maxReviewsPerProduct": 100
}

Filter by facets (region, source, themes)

{
  "products": ["slack"],
  "primaryRegions": ["Americas"],
  "sourceTypes": ["organic"],
  "responseTypes": ["video"],
  "maxReviewsPerProduct": 50
}

Discover valid facet values

Returns the distinct values for each facet to the metadata dataset, so you know what to pass to the facet filters above. No reviews are scraped in this mode.

{
  "enumFields": ["source_type", "response_type"]
}

Search within reviews with highlights

{
  "products": ["slack"],
  "reviewSearch": "customer support",
  "includeHighlights": true,
  "maxReviewsPerProduct": 20
}

Include Q&A discussions

{
  "products": ["slack"],
  "maxReviewsPerProduct": 10,
  "includeDiscussions": true
}

Input

Parameter	Type	Default	Description
products	string[]		Product URLs, slugs, or category URLs
search	string		Find products by keyword
category	string		Browse products by G2 category
vendor	string		Browse products by vendor name
compare	string[]		Compare product slugs side-by-side (analytics)
maxProducts	number	10	Max products in search/category/vendor mode
sortBy	select	DATE_DESC	DATE_DESC, DATE_ASC, RATING_DESC, RATING_ASC, HELPFUL_DESC, G2_DEFAULT; rating sorts use the 1–5 star rating
ratings	string[]	["1","2","3","4","5"]	Filter by star rating: 1, 2, 3, 4, or 5
maxReviewsPerProduct	number	20	Max reviews per product (0 = unlimited)
companySegment	string[]		small_business, mid_market, enterprise
dateFrom	string		Date in YYYY-MM-DD format; reviews from this date
dateTo	string		Date in YYYY-MM-DD format; reviews through this date; cannot be earlier than dateFrom
primaryRegions	select[]		Americas, Asia Pacific, EMEA
sourceTypes	string[]		Filter by how the review was collected; use enumFields to list valid values
responseTypes	string[]		Filter by review medium (e.g. text, video)
loveThemes	string[]		Filter by themes reviewers praised
hateThemes	string[]		Filter by themes reviewers criticized
switchingThemes	string[]		Filter by reasons reviewers switched products
discounts	string[]		Filter by reported discount
reviewSources	string[]		Filter by review source channel
reviewSearch	string		Search within review text. Alone = global search
similarTo	string		Find reviews similar to this text
reviewIds	string[]		Fetch specific reviews by ID
maxReviews	number	100	Max reviews for global/similar search; 0 returns all global matches and uses the 1,000-review Similar cap
enumFields	select[]		Discover source_type, response_type, love_theme_humanized, hate_theme_humanized, switching_theme_humanized, discount, or review_source; writes to the metadata Dataset instead of collecting reviews
analytics	boolean	false	Return aggregation analytics instead of reviews; requires one product selector
collapseByProduct	boolean	false	One review per product in global or Similar search
includeDiscussions	boolean	false	Fetch Q&A discussions in product review mode; cannot be combined with analytics
includeHighlights	boolean	true	Highlight matching text in search results
proxyConfiguration	object	{"useApifyProxy":true}	Configure network routing; Apify Proxy is enabled by default

Choose one input mode: one of products, search, category, or vendor; global reviewSearch; similarTo; reviewIds; compare; or enumFields. reviewSearch can also be combined with a product selector to search within those products. Set analytics: true alongside one product selector. Repeated selectors and products found through overlapping inputs are emitted once.

Review filters apply to product reviews, global search, Similar results, analytics, and comparisons. A direct reviewIds lookup returns only the requested records, and enumFields returns facet values, so review filters do not apply to those two modes.

All review and analytics modes use public, approved reviews. A requested review ID is omitted when it does not identify a public, approved review.

When you switch away from the prefilled product example, clear products before setting another mode. Conflicting modes, malformed dates, inverted date ranges, and invalid enum values fail with an input error before collection starts. An unknown product or comparison target is skipped; if none of the requested products resolves, the run succeeds with empty datasets. A valid search with no matches also succeeds with an empty reviews dataset.

sortBy orders product and global-search reviews. Similar results remain in relevance order, while direct-ID result order is not guaranteed. Highlights are returned only when reviewSearch is present and includeHighlights is enabled.

Some valid older or rating-only reviews have an empty title or rawText, and some discussions contain only a title. These fields remain empty strings rather than being filled with invented text.

Output

Datasets

Dataset	When	Content
reviews (default)	Most modes	Individual review records
products	Product-based modes	Product profiles
analytics	analytics or compare mode	Aggregation statistics
discussions	includeDiscussions: true	Q&A threads
metadata	enumFields mode	Distinct facet values

Review fields

The selected fields below came from one live Dataset record. Values are unchanged. Reviewer identity and other personal fields are omitted.

{
  "id": "13038740",
  "productId": 3437,
  "productName": "Slack",
  "reviewUrl": "https://www.g2.com/products/slack/reviews/13038740",
  "title": "One-Click Integrations That Just Work for slack",
  "pros": "Integration with other apps has to be 1 click",
  "cons": "UI when you send the messages. The typing window should be adjustable",
  "benefits": "connect and chatting with the team One-Click Integrations That Just Work for slack",
  "rawText": "Integration with other apps has to be 1 click\nUI when you send the messages. The typing window should be adjustable\nconnect and chatting with the team One-Click Integrations That Just Work for slack",
  "responseType": "text",
  "nps": 8,
  "npsScore": 4,
  "easeOfUse": 5,
  "easeOfSetup": 5,
  "easeOfAdmin": 5,
  "easeOfDoingBusinessWith": 4,
  "meetsRequirements": 5,
  "qualityOfSupport": 3,
  "securityConfidence": 3,
  "securityImportance": 7,
  "direction": null,
  "loveTheme": null,
  "loveThemeHumanized": null,
  "loveThemeParentHumanized": null,
  "hateTheme": null,
  "hateThemeHumanized": null,
  "hateThemeParentHumanized": null,
  "switchedFromOtherProduct": "unknown",
  "switchedFromProductIds": null,
  "switchedReason": null,
  "switchingThemes": null,
  "switchingThemesNested": null,
  "submittedAt": "2026-06-26T17:10:48.781-05:00",
  "publishedAt": "2026-06-26T19:30:00.900-05:00",
  "reviewSource": "user_profile",
  "sourceType": "user_profile",
  "helpful": 0,
  "englishAvailable": true,
  "contractLength": null,
  "discount": null,
  "price": 5,
  "status": "approved",
  "isPublic": true,
  "published": true,
  "productPaidProfile": false,
  "productIds": [3437],
  "categoryIds": [2174, 2360, 2363],
  "defaultSort": 8.525176115868225,
  "templateKeyword": "react",
  "videoReviewSubmittedAt": null,
  "editionIds": [],
  "expertiseIds": [],
  "serviceLocation": null,
  "contractLengthWeight": null,
  "discountWeight": null,
  "everApproved": true,
  "isInLimbo": false,
  "read": false,
  "reportable": true,
  "userUpdatedAt": "2026-06-26T17:10:48.965-05:00",
  "highlightedSnippets": null,
  "scrapedAt": "2026-06-28T06:00:34.982Z"
}

Analytics output

Aggregated over all matching public, approved reviews for a product. nps/easeOfUse/etc. each carry avg, stdDev, min, max, and count; the full record also includes easeOfSetup, easeOfAdmin, qualityOfSupport, meetsRequirements, easeOfDoingBusinessWith, securityConfidence, securityImportance, direction, topHateThemes, uniquelyHated, and bySourceType. The selected fields below came from one live analytics record and keep their values unchanged. This example omits reviewer identity from nested samples.

{
  "productId": 3437,
  "productName": "Slack",
  "productSlug": "slack",
  "totalReviews": 20201,
  "totalHelpfulVotes": 69,
  "nps": { "avg": 9.211870699470323, "stdDev": 1.1668658162744323, "min": 0, "max": 10, "count": 20201 },
  "easeOfUse": { "avg": 6.549680122289532, "stdDev": 0.7953808393082141, "min": 1, "max": 7, "count": 17663 },
  "ratingDistribution": { "1": 40, "2": 93, "3": 468, "4": 3646, "5": 15954 },
  "npsPercentiles": { "p10": 8, "p25": 9, "p50": 10, "p75": 10, "p90": 10 },
  "uniqueCountries": 140,
  "uniqueIndustries": 145,
  "topLoveThemes": [
    { "theme": "Chat", "count": 3436, "sampleReviews": [{ "title": "Slack Review", "npsScore": 4 }] },
    { "theme": "Collaboration", "count": 2254, "sampleReviews": [{ "title": "\"Ping\" From Your Creative Side", "npsScore": 5 }] }
  ],
  "uniquelyLoved": [
    { "theme": "Chat", "score": 3.472215696777018, "count": 3436 },
    { "theme": "Collaboration", "score": 1.2096931571663356, "count": 2254 }
  ],
  "switchingReasons": [
    { "reason": "Ease of use", "count": 126 },
    { "reason": "Communication Features", "count": 108 }
  ],
  "bySegment": [
    { "code": 179, "label": "Small Business (50 or fewer emp.)", "count": 8191 },
    { "code": 180, "label": "Mid-Market (51-1000 emp.)", "count": 8141 },
    { "code": 181, "label": "Enterprise (>1000 emp.)", "count": 3516 }
  ],
  "byRegion": [
    { "region": "Americas", "count": 12760 },
    { "region": "Asia Pacific", "count": 4821 },
    { "region": "EMEA", "count": 2526 }
  ],
  "reviewsOverTime": [
    { "month": "2014-02", "count": 1 },
    { "month": "2026-06", "count": 73 }
  ]
}

Facet values output (metadata dataset)

{
  "field": "source_type",
  "type": "enum_values",
  "values": [
    "ai_assisted_review", "api_collect", "g2_incentivized", "g2_nonincentivized",
    "g2gives", "influitive", "organic", "pop_up", "referral", "review_referral",
    "review_widget", "stale_reviews_callout", "thank_you_page", "un_survey",
    "unincentivized", "user_profile", "vendor", "vendor_nonincentivized",
    "vendor_self", "vendor_self_nonincentivized", "vendor_self_sendoso"
  ],
  "count": 21
}

Limitations

• Reviewer job titles and avatar URLs are not available.
• industry and role are returned as numeric category codes with a human-readable label for common values only; they are not available as filters.
• country and region are not available as filters because their values are not normalized across languages. Use primaryRegion (Americas / Asia Pacific / EMEA) for consistent geography filtering.
• similarTo returns at most 1,000 reviews per run (a single relevance-ranked fetch; it does not paginate).
• A product's reviewCount reflects G2's total across all reviews; review scraping and analytics return only public, approved reviews, so their counts are typically lower.
• Some reviews have unstructured text (no clear pros/cons/benefits separation).