Similarweb Traffic Checker (Visits, Sources, AI Referrals)

Website Traffic Checker (Visits, Sources & Chatbot Referrals)

Get monthly visits, traffic sources, top countries and ChatGPT, Claude and Gemini referral share for any list of domains, one row each. Paste this into the JSON tab of the input and click Start:

{"domains": ["airtable.com", "notion.so", "canva.com"], "includeWebsiteChecks": true}

You get 3 rows, usually in under a minute. Bare domains, full URLs and email addresses all work: each becomes its main domain and duplicates are merged. Up to 5,000 domains per run, from Similarweb's public estimates. No login, no API key.

What people use it for:

• Sorting a lead list by monthly visits before the sales team spends time on it.
• Comparing visits, channel mix and top countries for a whole market in one table.
• Seeing which sites get visits from ChatGPT, Gemini, Claude and Perplexity, how many visits each chatbot sends and how each one's share moved over 3 months.
• Filtering an investment or partner list by web presence, growth and category rank.
• Adding traffic, email provider and tech stack columns to every account in a CRM.

What you get

Group	Fields
Traffic	monthlyVisits, monthlyVisitsChange, monthlyVisitsHistory (last 3 months), trafficMonth
Ranks	globalRank, countryRank + countryRankCountry, category, categoryRank
Engagement	bounceRate, pagesPerVisit, avgVisitDurationSeconds
Channels	trafficSources (direct, organic and paid search, referrals, social, paid social, email, display, affiliates, AI chatbots), topTrafficSource, trafficSourceVisits (estimated visits per channel, same keys, e.g. {"direct": 344109372, "organicSearch": 157118626})
Audience	topCountries (top 5 with share), topKeywords (top 5 with search volume and CPC)
AI chatbots	aiChatbotVisits, aiChatbotShare, topAiChatbots, aiChatbotShares (each chatbot's share of the AI chatbot referrals in percent, e.g. {"chatgpt": 61.82, "claude": 12.86, "gemini": 9.45}; Similarweb gives the biggest few only, usually 3, so they need not add up to 100), aiChatbotVisitsByChatbot (estimated visits each chatbot sent, e.g. {"chatgpt": 7422431, "claude": 1544038}), aiChatbotSharesHistory (the per-chatbot shares for each of the last 3 months, oldest first)
Email set-up	emailProvider, emailSecurityGateway, mxRecords, hasSpf, hasDmarc, dmarcPolicy, hasBimi, hasMtaSts
Website	httpStatus, finalUrl, https, cdn, hosting, dnsProvider, cms, ecommerce, analytics, marketing, chatWidgets, payments
SSL and domain	sslIssuer, sslDaysToExpiry, registrar, registeredAt, expiresAt (never owner contact details)
Flags	trafficAvailable, dataFromGoogleAnalytics, trafficError, checkErrors
Company (optional)	company: legalName, registry, id, country, incorporationDate, status, sizeBand, only when companyFacts is on (below)
Competitor watch (optional)	previousMonth, visitsChange, visitsChangePct, globalRankChange, aiShareChange, chatbotShareChanges, changed, watchStatus: only when compareWithPrevious is on

One real row (run on 2026-10-14, month 2026-09, shortened):

{
  "domain": "airtable.com",
  "trafficAvailable": true,
  "trafficMonth": "2026-09",
  "monthlyVisits": 25813226,
  "monthlyVisitsChange": -0.009,
  "monthlyVisitsHistory": [{"month": "2026-07", "visits": 27389611}, {"month": "2026-08", "visits": 26047033}, {"month": "2026-09", "visits": 25813226}],
  "globalRank": 1485,
  "countryRank": 679,
  "countryRankCountry": "US",
  "categoryRank": 23,
  "bounceRate": 0.3458,
  "pagesPerVisit": 7.68,
  "avgVisitDurationSeconds": 575,
  "topTrafficSource": "Direct",
  "trafficSources": {"direct": 0.7579, "organicSearch": 0.045, "paidSearch": 0.0046, "referrals": 0.0746, "socialOrganic": 0.049, "socialPaid": 0.0026, "email": 0.0559, "displayAds": 0.0037, "affiliates": 0.0009, "aiChatbots": 0.0059},
  "trafficSourceVisits": {"direct": 19563111, "organicSearch": 1161782, "paidSearch": 117957, "referrals": 1925377, "socialOrganic": 1265469, "socialPaid": 66386, "email": 1443130, "displayAds": 95414, "affiliates": 22507, "aiChatbots": 152094},
  "topCountries": [{"country": "US", "share": 0.5238}, {"country": "GB", "share": 0.0504}],
  "topKeywords": [{"keyword": "airtable", "searchVolume": 493540, "estimatedValue": 325650, "cpcUsd": 5.16}],
  "topAiChatbots": ["claude.ai", "chatgpt.com", "gemini.google.com", "perplexity.ai"],
  "aiChatbotVisits": 137217,
  "aiChatbotShares": {"claude": 39.88, "chatgpt": 36.98, "gemini": 4.73},
  "aiChatbotVisitsByChatbot": {"claude": 54722, "chatgpt": 50743, "gemini": 6490},
  "aiChatbotSharesHistory": [{"month": "2026-07", "shares": {"claude": 36.38, "chatgpt": 58.96, "gemini": 3.72}}, {"month": "2026-08", "shares": {"claude": 24.2, "chatgpt": 59.26, "gemini": 3.33}}, {"month": "2026-09", "shares": {"claude": 39.88, "chatgpt": 36.98, "gemini": 4.73}}],
  "trafficDataBy": "Similarweb estimates",
  "emailProvider": "Google Workspace",
  "hasSpf": true,
  "hasDmarc": true,
  "dmarcPolicy": "reject",
  "dnsProvider": "AWS Route 53",
  "cdn": "Amazon CloudFront",
  "hosting": "Vercel",
  "sslDaysToExpiry": 104,
  "registrar": "MarkMonitor Inc."
}

The dataset has six views (Overview, Channels and audience, AI chatbots, Email set-up, Website and tech stack, Changes) and exports to JSON, CSV, Excel or HTML. The run's SUMMARY record lists anything that could not be read, with the reason.

Price

You pay per domain returned. There is no charge for Apify platform usage on top.

Row	Free and Starter	Scale	Business
Domain with a traffic estimate	$0.70 per 1,000	$0.525 per 1,000	$0.385 per 1,000
Domain without a traffic estimate (website and email columns only)	$0.35 per 1,000	$0.2625 per 1,000	$0.1925 per 1,000

Plus $0.00005 per run start. Apify's free $5 monthly credit covers about 7,100 domains with traffic estimates. Free-plan runs stop at 100 domains each; any paid Apify plan removes the cap.

What's free: invalid inputs, duplicates, domains skipped by onlyNew, domains left out by the filters, unchanged domains left out by onlyChanged, and domains not reached before the run's time limit. To cap a run's spend, set Maximum cost per run in the run options.

Limits

• Traffic numbers are Similarweb's estimates: the figures its free browser extension shows without an account. They are modelled, not measured, unless dataFromGoogleAnalytics is true. Treat them as a guide to size and trend, especially for sites under about 50,000 visits a month, and credit "Similarweb" if you publish them.
• Estimates cover the latest month Similarweb has published plus the 2 months before it. Older months are not available.
• Small or new sites often have no estimate: trafficAvailable is false and the row is charged at the lower domain-basic price.
• If Similarweb refuses a request on every retry, the row still ships with the website and email columns, trafficAvailable: null and the reason in trafficError, at the lower price. With website checks off, such a domain is skipped and not charged.
• Subdomains are looked up as their main domain (blog.example.com becomes example.com), using the full Public Suffix List.
• The website checks fetch each homepage once and respect its robots.txt. Sites behind bot protection can show no tech stack; checkErrors says why.
• Eight domains are checked at a time, and 1,000 domains take roughly 5 minutes. If the time limit is reached, the run stops cleanly, keeps what it found and lists the rest in SUMMARY.

Only new since last run, and schedules

Similarweb publishes one new month at a time. Turn on onlyNew and schedule the Actor monthly: each run skips domains already delivered for the current month, and skipped domains cost nothing. The memory is shared by every run of this Actor on your account. Set up a schedule under Schedules in Apify Console.

Filters: only the domains you want

Paste a long list and keep only the domains that match. Domains left out are not returned and not charged, and their website and email checks are skipped, so a filtered run is also faster. All filters are optional and off by default.

Filter	Keeps a domain when
minMonthlyVisits, maxMonthlyVisits	visits in the latest month are at least / at most this
maxGlobalRank	its global rank is this or better (e.g. 10000 = the top 10,000)
topCountry	its biggest source of visits is this country (two-letter code: US, GB, DE)
topCountry + minTopCountryShare	that country brings at least this percent of visits, wherever it ranks
minTopCountryShare alone	its biggest country brings at least this percent (a mostly-local audience)
minAiChatbotShare	at least this percent of visits came from AI chatbots (aiChatbotShare)
categoriesInclude / categoriesExclude	its Similarweb category contains / doesn't contain any of these words (not case-sensitive)

{
  "domains": ["shopify.com", "etsy.com", "bbc.co.uk", "github.com"],
  "minMonthlyVisits": 1000000,
  "topCountry": "US",
  "minTopCountryShare": 30,
  "categoriesExclude": ["news"]
}

A domain without the number a filter needs (too small for an estimate, or traffic not read this run) is left out by that filter, because there's no way to confirm it matches. A domain with no category passes categoriesExclude but not categoriesInclude. The run's message and SUMMARY (filteredOut) count how many domains each filter left out. Rows that pass are exactly the rows you'd get without filters, at the same price.

Company facts: who runs the site

Turn on companyFacts and each row gets a company object: the organisation behind the domain, from the Open Company Graph, our index of official company registers (UK Companies House, France, the Nordics and Baltics, the US SEC and several US states, Singapore, Japan, Australia, Canada and the global LEI register), matched to company websites. The shape (illustrative values):

"company": {
  "legalName": "Acme Widgets Ltd",
  "registry": "Companies House",
  "id": "GB-CH-01234567",
  "country": "GB",
  "incorporationDate": "2015-03-02",
  "status": "Active",
  "sizeBand": "10-49"
}

• A key appears only when the register has it. sizeBand (1-9, 10-49, 50-249, 250-999, 1000+) comes from a staff count the register or a filing states, so many companies have none.
• Coverage is still growing: about 210,000 company websites are linked to a register entry today, mostly where the site prints its own company number (strongest for the UK, France, the Nordics and the Baltics). Many sites, including some big brands, have no link yet, so expect company: null for a good share of a typical list.
• company: null means no company is linked to the domain yet, or the lookup could not be read this time; the run's message counts the second case. Rows are delivered either way.
• Same price: no extra charge per row. Organisations only, never people.

Competitor watch (schedule it weekly)

Turn on compareWithPrevious and the Actor remembers each domain's visits, global rank and AI chatbot shares, then tells you what moved since the last run of the same list. Each row gains previousMonth, visitsChange, visitsChangePct, globalRankChange (positive = the site climbed), aiShareChange (percentage points), chatbotShareChanges (points per chatbot, e.g. {"chatgpt": -3.2, "claude": 2.9}) and changed. The Changes dataset view shows just these columns.

{
  "domains": ["hubspot.com", "salesforce.com", "pipedrive.com", "attio.com"],
  "compareWithPrevious": true,
  "watchName": "crm-competitors",
  "onlyChanged": true
}

• First run: each domain is saved as the baseline and its change columns are empty. With onlyChanged on, that first run returns nothing and costs nothing; leave it off once if you want the starting numbers too.
• changed is true when Similarweb has published a newer month for the domain, or when, within the same month, visits moved by at least minVisitsChangePct (default 10%) or the AI chatbot share, or one chatbot's share, moved by at least minAiShareChangePts (default 1 point).
• onlyChanged returns and charges only rows where changed is true. Unchanged domains are read and remembered but not returned and not charged.
• Expect quiet weeks. Similarweb publishes one month at a time, so a weekly schedule mostly reports "no change" until the new month lands, then every domain shows the month-on-month move. With onlyChanged on, those quiet weeks cost nothing.
• Several lists on one account: give each its own watchName. The memory lives in a key-value store in your own Apify account, and a failed lookup never overwrites it.

To run it weekly: save the input as a task, open Schedules in Apify Console, create a schedule (for example every Monday at 08:00) and add the task. To get the changes in Slack or Google Sheets, add an integration to the task, or start from this free n8n template: Track AI chatbot referral traffic for competitor sites.

Integrations: API, n8n, Make and AI agents

API. Run it and get the rows back in one call:

curl -X POST "https://api.apify.com/v2/acts/locaihost~website-traffic/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"domains": ["airtable.com", "notion.so"]}'

Apify's API clients for Python and JavaScript work the same way.

n8n, Make, Zapier, Google Sheets, Slack, webhooks. Apify's built-in integrations send each run's dataset wherever you need it.

AI agents (MCP). Add this Actor to Claude, Cursor or any MCP client with https://mcp.apify.com/?tools=locaihost/website-traffic, then ask something like "Which of these 20 competitors grew traffic last month, and where do their visitors come from?"

FAQ

How much does it cost to check website traffic?

$0.70 per 1,000 domains with a traffic estimate on the Free and Starter plans, less on Scale and Business, and half that for domains without an estimate. Apify's free $5 monthly credit covers about 7,100 domains.

Can I use it through the API or from Python?

Yes. Use the run-sync-get-dataset-items call above, or Apify's Python and JavaScript clients. Every run's data is also available as JSON, CSV or Excel from its dataset.

Can I use it from an AI agent?

Yes. Through the Apify MCP server (https://mcp.apify.com/?tools=locaihost/website-traffic) any MCP client can call it with a list of domains and read the rows back.

Is it legal to check website traffic this way?

The Actor reads publicly available data: Similarweb's public estimates, public DNS records, public web pages and public domain registries. It never logs in and never collects personal contact details (email addresses typed as input are reduced to their domain, and DMARC report addresses are masked). You are responsible for how you use the results, including Similarweb's terms and any data-protection and marketing rules that apply to you.

How many visits does ChatGPT (or Claude, Gemini, Perplexity) send to a site?

aiChatbotVisitsByChatbot gives the estimate per chatbot for the latest month, for example {"chatgpt": 7422431, "claude": 1544038, "gemini": 1134616} for github.com in September 2026. It is Similarweb's AI chatbot visit count (aiChatbotVisits) multiplied by each chatbot's share. aiChatbotSharesHistory shows how those shares moved over the last 3 months, so you can see whether Claude or Gemini is gaining on ChatGPT for a site.

How many visits come from Google search, paid ads or social?

trafficSourceVisits turns the channel shares into visit counts: monthly visits multiplied by each channel's share. The channels add up to monthlyVisits, give or take rounding. trafficSourceVisits.aiChatbots can differ from aiChatbotVisits, most on smaller sites, because Similarweb counts AI chatbot visits separately from its channel split.

Can I get the company behind each website?

Yes: turn on companyFacts. Each row gets the legal name, register, company id, country, incorporation date, status and size band where the official register has them, at no extra cost. See "Company facts" above.

Why are some fields empty?

Small and new sites often have no Similarweb estimate (trafficAvailable: false), and some sites block the homepage check (checkErrors says why). Keyword and country lists only exist where Similarweb publishes them.

More from locaihost data

• Tech Stack & Domain Lookup: the email, DNS and tech stack checks on their own, for any list of domains.
• Company Enrichment by Domain: company name, description, logo, socials and role contact addresses from each site.
• Google Maps Leads Scraper: find local businesses, then check their traffic here.
• SEO Site Audit: page-level SEO checks for the sites that matter.

Did it help?

If this saved you time, a quick rating on the Store helps other people find it.

Support

Something not working, or a field you need? Open an issue on the Issues tab: we answer issues within 24 hours.