Gartner Peer Insights Scraper

Gartner Peer Insights Scraper — reviews API for enterprise software

Pulls verified user reviews from Gartner Peer Insights for any product on the site. You give it a product URL; you get one row per review with the review body, the headline, the 1–5 star rating, the date, and who wrote it — job title, function, industry, company size — plus the product, vendor and Gartner market it belongs to.

No Gartner account, no cookie, no token. Everything here is what a logged-out visitor can read.

What a row looks like

Real output, Zoom Meetings, unedited:

{
  "reviewId": "6786406",
  "reviewUrl": "https://www.gartner.com/reviews/market/meeting-solutions/vendor/zoom/product/zoom-meetings/review/view/6786406",
  "headline": "Difficulty Managing Participant Controls Raises Concerns About Zoom Usability",
  "reviewText": "I signed up to manage the Zoom presence on the National Day of Prayer. However, when I started the meeting I was notified someone was 'waiting' to be admitted...",
  "rating": 3,
  "reviewDate": "May 7, 2026",
  "reviewerJobTitle": "Software Developer",
  "reviewerFunction": "Software Development",
  "reviewerIndustry": "Education",
  "companySize": "Gov't/PS/ED <5,000 Employees",
  "isPartnerReview": false,
  "reviewSourceCode": 4,
  "reviewIncentiveCode": 2,
  "upVotes": 0,
  "pros": null,
  "cons": null,
  "productName": "Zoom Meetings",
  "vendorName": "Zoom",
  "marketName": "Meeting Solutions",
  "productUrl": "https://www.gartner.com/reviews/product/zoom-meetings"
}

Every run also returns one free summary row per product: average rating, the full 1–5 rating distribution, Gartner's own review count, the markets the product sits in, the product description and its pricing model. Summary, sample and diagnostic rows carry _charged: false and cost nothing.

Input

Paste the URL from your address bar. All of these work:

{
  "productUrls": [
    "https://www.gartner.com/reviews/product/azure-data-factory",
    "https://www.gartner.com/reviews/market/meeting-solutions/vendor/zoom/product/zoom-meetings/reviews",
    "snowflake"
  ],
  "maxReviewsPerProduct": 200,
  "recentMonths": 12
}

Speed and scale

Gartner hands back one review per request, so throughput is simply how many requests you can make. No proxy is needed for the usual path. Measured on the Apify platform, wall-clock, container start included:

Zoom Meetings has 6,322 reviews on Gartner, so a 250-review pull is a small slice of one product.

What this does not do

Being straight about the login wall, because it is the whole story on this site.

• Per-review pros/cons are login-walled. Gartner publishes verbatim "what I liked / what I

disliked" text for exactly two reviews per market — the one favourable and one critical review it spotlights. Those two rows get real pros and cons. Every other row leaves both fields null. Nothing here fakes them.

• Per-review sub-ratings are login-walled too. There are no per-review evaluation, deployment,

support or "likelihood to recommend" scores in the public data, so this actor does not emit any. Product-level aggregates (average rating, 1–5 distribution) are public and are in the summary row.

• There is no date sort. Gartner's endpoint honours its own helpfulness ranking and silently

ignores every date-sort value it is given. Use recentMonths to get recent reviews — it is a real server-side filter and it is cheaper, because you are not charged for the rows it removes.

• Star filtering is done here, not by Gartner. Gartner has no rating filter on this endpoint, so

the actor reads reviews and drops the ones you did not ask for. You are only charged for rows you keep, but a rare rating costs reading time; scanning stops after about 12 reads per requested review.

• Review counts differ between screens on Gartner's own site, because ratings and written

reviews are counted differently and a product's reviews are split across its markets. The summary row reports both reviewsOnGartner (what the review endpoint says) and ratingsCount (what the ratings endpoint says) rather than pretending they agree.

• Duplicates exist in Gartner's index. One 132-review market answered 132 index positions with

126 distinct review ids. Rows are deduplicated by reviewId, so you are charged for 126, not 132.

• A few markets are slower than the rest. Gartner's own edge blocks a handful of market slugs on

the fast route — ai-platforms-for-data-science-and-machine-learning is one — and those fall back to a slower, rate-limited path. Other markets of the same product are unaffected, and a market that cannot be read at all is reported as a free diagnostic row rather than silently skipped.

Pricing

Pay per review returned, plus a small per-run start fee. You are only charged for genuine review rows — summary, sample and diagnostic rows are free. A run with no input, or with input that is not a Gartner product reference at all, is not charged anything, not even the start fee.

FAQ

Does this need a Gartner login or a Peer Insights account? No. It reads the same public endpoints the logged-out product page reads. There is no account, no password and no session cookie anywhere in this actor.

Can I get the full text of every review? Yes. reviewText is the reviewer's own written review, in full, on every row — not a truncated teaser. What is *not* public is the separate pros/cons breakdown and per-review sub-scores.

How do I scrape Gartner Peer Insights reviews for a specific product? Copy the product URL from Gartner (for example https://www.gartner.com/reviews/product/azure-data-factory), put it in productUrls, set maxReviewsPerProduct, and run. That is the whole setup.

How do I get only recent reviews? Set recentMonths. A product with 6,322 lifetime reviews returns 48 at recentMonths: 12 and 3 at recentMonths: 1.

Why do some rows have pros and cons and most do not? Because Gartner only publishes that breakdown for the two reviews it spotlights per market. Filling the other rows would mean inventing text, so those stay null.

Can I monitor a product for new reviews? Yes — schedule the actor with recentMonths: 1 and a small maxReviewsPerProduct. Reviews are keyed by reviewId, so deduplicating across runs is a one-line join.

What are reviewSourceCode and reviewIncentiveCode? Gartner's own internal codes for how a review was solicited and whether it was incentivised. Gartner does not publish a legend for them, so they are passed through as raw numbers rather than guessed at.

Does it work for products in more than one Gartner market? Yes. Gartner splits a product's reviews by market. If your URL does not name a market, the actor walks every market the product appears in and deduplicates the results.

What about Capterra, G2 or Software Advice? Different sites, different scrapers. This one covers Gartner Peer Insights only.