Similarweb Website Traffic + AI-Traffic Scraper 📊

The only Similarweb scraper that breaks out AI-assistant traffic — see what % of any domain's visits come from ChatGPT, Claude, Gemini and Perplexity, and exactly which assistants drive them. Plus the full traffic profile: global / country / category rank, estimated monthly visits with a 3-month trend, engagement metrics, the traffic-source split, top countries, and top organic keywords. One clean, flat row per domain. JSON or CSV.

No monthly fee. 🚀 Launch price: $0.99 per 1,000 domains — and only for domains that return data. No API key, no browser, no CAPTCHA. Just paste a list of domains.

Why use this scraper

• AI-traffic intelligence no one else surfaces. See exactly what % of a site's visits come from AI assistants, and which ones (ChatGPT, Gemini, Claude, Perplexity, Grok, Copilot…). This is the fastest-growing channel in web analytics — and the single biggest gap in every other Similarweb scraper on the market.
• No monthly fee. Most Similarweb actors are rentals ($19+/month before you scrape a single domain). This one is pure pay-per-result: 🚀 launch price $0.99 / 1,000 domains, $0 to start, and domains with no data (hasData: false) are free. A one-off check of 500 domains costs about fifty cents.
• Proven reliable. A 199-domain benchmark returned 100% of rows with data, zero failures, in 69 seconds for ~$0.017 of compute. No browser, no CAPTCHA — pure HTTP with rotating residential IPs.
• Bulk, not one-at-a-time. Paste 1 domain or 5,000 — each becomes one self-contained row. Similarweb's own UI makes you check domains one by one.
• Everything the official extension shows — at scale. Reads the same data source the Similarweb browser extension uses, so you get the rich profile for every domain without a paid Similarweb seat.
• Flat rows, ready for a spreadsheet. Ranks, visits, engagement, traffic split, countries and keywords are all denormalized onto one row. Open the CSV and filter.

What you get per domain

Group	Fields
Identity	domain, site name, title, meta description, Similarweb category, profile URL, homepage screenshot
Ranks	global rank, country rank (+ country), category rank (+ category)
Traffic	estimated monthly visits + a 3-month visit-history trend
Engagement	bounce rate, pages per visit, average visit duration
Traffic sources	direct, organic search, paid search, organic social, paid social, referrals, email, display ads, affiliate, AI / GenAI — each as a share of total visits
Audience	top countries with traffic share
SEO	top organic keywords with search volume, estimated value and CPC
AI traffic ★	AI-referral share, total AI-referred visits, and the ranked list of AI assistants sending traffic

Supported input

Paste website addresses in any of these shapes — each is normalized to its root domain automatically:

• Bare domain — stripe.com
• With www. — www.stripe.com
• Full URL — https://stripe.com/pricing?ref=nav
• A Similarweb profile URL — https://www.similarweb.com/website/stripe.com/

Duplicate entries are de-duplicated (looked up once, charged once). Invalid entries are skipped with a clear warning.

Use cases

• Competitive intelligence — benchmark your domain against rivals on rank, visits and engagement.
• SEO & content — pull each competitor's top organic keywords (with volume + CPC) and their organic-search dependence.
• AI-visibility (GEO) audits — measure how much traffic a site already gets from AI assistants and which ones — the new frontier of search.
• Lead qualification — score prospect websites by real traffic before you reach out.
• Market sizing — estimate total addressable traffic across a list of players in a category.
• Investment / M&A diligence — verify a target's traffic claims and trend direction.

How it works

1. Input — your domains are normalized to root domains and de-duplicated.
2. Fetch — for each domain the actor calls Similarweb's public data endpoint (the same JSON the official extension reads). The endpoint rate-limits per IP, so every request runs through a fresh rotating residential proxy; on retry it escalates to a 3-stack HTTP race (impit / gotScraping / curl-impersonate) so a transient block never costs you a row.
3. Emit — the JSON is flattened into one tidy row and pushed to the dataset. No browser, no CAPTCHA — pure HTTP, ~0.8s per domain at volume.

Input configuration

Field	Type	Default	Description
domains	array	—	Domains/URLs to look up (one row each). Required.
maxItems	integer	1000	Max domains per run. Free-tier capped at 100.
maxConcurrency	integer	6	Parallel lookups (5–8 is the sweet spot).
maxRequestRetries	integer	6	Per-domain retry budget; each retry rotates the proxy IP.
proxy	object	Residential	Residential proxies are required (the endpoint blocks single-IP bursts).

Example input

{
  "domains": ["stripe.com", "github.com", "openai.com"],
  "maxItems": 1000,
  "maxConcurrency": 6
}

Output sample

{
  "domain": "openai.com",
  "siteName": "openai.com",
  "title": "OpenAI",
  "category": "ai_chatbots_and_tools",
  "similarwebUrl": "https://www.similarweb.com/website/openai.com/",
  "hasData": true,
  "globalRank": 201,
  "countryRank": 270,
  "countryRankCountryCode": "US",
  "categoryRank": 6,
  "categoryRankCategory": "AI_Chatbots_and_Tools",
  "monthlyVisits": 203086642,
  "bounceRate": 0.5819,
  "pagesPerVisit": 2.81,
  "avgVisitDurationSec": 144.8,
  "engagementMonth": "2026-05",
  "monthlyVisitsHistory": [
    { "date": "2026-03-01", "visits": 203740797 },
    { "date": "2026-04-01", "visits": 195737812 },
    { "date": "2026-05-01", "visits": 203086642 }
  ],
  "trafficDirect": 0.4367,
  "trafficSearchOrganic": 0.1888,
  "trafficReferrals": 0.0933,
  "trafficGenAi": 0.2358,
  "topCountries": [
    { "countryCode": "US", "share": 0.2216 },
    { "countryCode": "IN", "share": 0.0708 }
  ],
  "topKeywords": [
    { "keyword": "chatgpt", "volume": 176305070, "estimatedValue": 21579710, "cpc": 0.15 }
  ],
  "aiTrafficTotalVisits": 46810867,
  "aiReferralTraffic": 0.2329,
  "aiTrafficTopSources": ["chatgpt.com", "gemini.google.com", "claude.ai", "perplexity.ai", "grok.com"],
  "screenshotUrl": "https://site-images.similarcdn.com/image?url=openai.com&t=1&s=1",
  "scrapedAt": "2026-06-10T13:35:39.892Z"
}

Key output fields

Field	Meaning
globalRank	Worldwide traffic rank (1 = most-visited site on earth).
monthlyVisits	Estimated visits in the latest snapshot month.
monthlyVisitsHistory	Last 3 months of estimated visits — the trend direction.
bounceRate / pagesPerVisit / avgVisitDurationSec	Core engagement quality metrics.
traffic*	Share (0–1) of visits from each channel: direct, search, social, referral, mail, display, affiliate, GenAi.
trafficGenAi	Share of visits from AI assistants — the standout, fastest-growing channel.
aiTrafficTopSources	Which AI assistants send traffic, ranked.
topKeywords	Top organic keywords with volume, estimated value and CPC.
topCountries	Top countries by traffic share.
hasData	false when Similarweb has no data for the domain — flagged, not charged.

Pricing

🚀 Launch price: $0.99 per 1,000 domains (pay-per-result). You're charged only for domains that return data — empty/unknown domains (hasData: false) are free. No subscription, no Similarweb seat required. This is an introductory rate while the actor builds its track record — lock it in now.

FAQ

Do I need a Similarweb account or API key? No. The actor reads Similarweb's public data endpoint directly.

Is there a monthly fee? No. Many Similarweb actors charge a monthly rental ($19+/month) on top of usage. This one is pure pay-per-result: $0.99 per 1,000 domains (launch price), nothing upfront, and you only pay for domains that return data. If you scrape 200 domains once, you pay for 200 domains — not a month.

Does it really include AI-assistant traffic? Yes — it's the headline feature. Every row carries trafficGenAi (the AI share of total visits) and aiTrafficTopSources (the ranked list of assistants — ChatGPT, Claude, Gemini, Perplexity, Grok, Copilot…). This is the new frontier of search analytics and most scrapers don't expose it.

How fresh is the data? It's Similarweb's monthly snapshot — the snapshotDate and engagementMonth fields tell you exactly which month each row describes. This matches what Similarweb shows the public and its browser extension.

Why is topCompetitors empty / null? Similarweb gates the "similar sites" competitor list behind its paid product; the public data surface usually returns it empty. The field is included so it auto-populates if Similarweb ever exposes it, but treat it as not available today. Everything else (ranks, visits, traffic sources, keywords, AI traffic) is fully populated.

Are the numbers exact? They're Similarweb's estimates, same as you'd see on similarweb.com. For small/low-traffic sites the isSmall flag is set and figures are coarser; very obscure domains return hasData: false.

Can I pass full URLs instead of bare domains? Yes — full URLs, www. prefixes, and Similarweb profile URLs all get normalized to the root domain automatically.

How fast is it? About 0.8s per domain at the default concurrency — a 1,000-domain run finishes in a few minutes.

Support

Found a bug or need a field added? Open an issue on the actor's Issues tab and we'll take a look.

Explore more scrapers

Building a competitive-intelligence or market-research stack? Pair this with:

• G2 Reviews Scraper — software reviews, ratings & buyer sentiment
• Capterra Scraper — software listings + reviews
• Trustpilot Scraper — brand reviews & ratings at scale
• Crunchbase Scraper — company funding rounds & profiles
• Glassdoor Scraper — employer reviews, salaries & interviews
• Product Hunt Scraper — launches, makers & contact emails

See the full memo23 actor catalog →

⚠️ Disclaimer

This actor collects only publicly available website-traffic estimates exposed by Similarweb's public data endpoint. It does not log in, bypass authentication, or access any private or personal data. Traffic figures are third-party estimates provided by Similarweb, not the scraper, and are intended for research, competitive analysis and market sizing. You are responsible for using the data in compliance with Similarweb's terms and all applicable laws.

SEO Keywords

similarweb ai traffic, ai traffic scraper, chatgpt referral traffic, claude traffic scraper, genai traffic share, ai assistant traffic, GEO audit, generative engine optimization, similarweb scraper, website traffic scraper, similarweb api, web traffic data, domain traffic estimate, monthly visits scraper, website rank checker, global rank scraper, traffic source analysis, organic keywords scraper, competitor traffic analysis, SEO competitive intelligence, bulk domain traffic, website analytics scraper, similarweb alternative, similarweb no rental, traffic intelligence API