Website Visitor Statistics: What They Mean, How to Check
Website visitor statistics are the numbers an analytics tool reports about who visits your site: users (people, roughly), sessions (visits), views (pages loaded), engagement or bounce rate, and traffic sources. To check your own site's statistics, open your analytics tool (in Google Analytics 4, Reports > Acquisition > Traffic acquisition). To check another website's monthly visitors, use an estimator such as Similarweb or Semrush, and treat the result as an estimate, not a count.
This guide explains each web page statistic in plain terms, shows where to find it, how to analyze website traffic once you have it, what "normal" numbers look like, and why two tools never show the same numbers. It's written for beginners, but the definitions come straight from Google's, Plausible's and Similarweb's own documentation, so it doubles as a reference.
The main website statistics, explained
Every analytics report is built from a handful of metrics. Learn these and you can read any dashboard, from GA4 to a hosting control panel.
| Statistic | What it counts | Watch out for |
|---|---|---|
| Users / visitors | Distinct browsers or devices, not people | One person on a phone and a laptop counts twice |
| New vs returning users | First visit vs repeat visit in the date range | Clearing cookies turns a returning user into a new one |
| Sessions / visits | A period of activity; in GA4 it ends after 30 minutes of inactivity | Definitions differ between tools |
| Views / pageviews | Each page or screen loaded | Single-page apps need route-change tracking |
| Views per session | Pageviews divided by sessions | Higher isn't always better: lost visitors click around too |
| Hits | Every request to the server: pages, images, scripts | Old log-file metric, far higher than pageviews |
| Engagement rate | Share of sessions that were engaged | GA4-specific definition, see below |
| Bounce rate | Share of sessions that were not engaged (GA4) or single-page (older tools) | Meaning changed with GA4 |
| Average engagement time | Time the page was in the foreground | Background tabs don't count |
| Traffic sources / channels | Where the visit came from: organic search, direct, referral, social, email, paid, AI assistants | Untagged links often land in "direct" |
| Landing pages | First page of each session | Shows which content brings people in |
| Conversions (key events) | Signups, purchases, downloads you define | Only counted if you set them up |
| Conversion rate | Conversions divided by sessions (or users), times 100 | Say which denominator you used |
Users, active users and total users in GA4
GA4 has several user counts. Total users is the number of unique users who triggered any event. Active users is the number who engaged with the site, and that's the number GA4's reports show by default (Google: user metrics). New users logged a first_visit (or first_open in apps) event in the date range; returning users had at least one earlier session.
In practice, when a GA4 report says "Users", it means active users. If you export to a spreadsheet and see a larger "Total users" number, that's not a bug: it includes people who loaded a page and left without engaging.
"Unique visitors" in other tools is the same idea with a different method. Plausible, for example, doesn't use cookies: it identifies a visitor with a hash of a daily salt, the domain, the IP address and the user agent, and the salt is "rotated and deleted every 24 hours" (Plausible data policy). So the same person counts once per day, but again on a new day or a new device.
Sessions and the 30-minute rule
In GA4 a session starts when someone opens your app or views a page with no active session, and it ends after 30 minutes of inactivity by default (Google: sessions). You can change the timeout under Admin > Data streams > (your stream) > Configure tag settings > Show all > Adjust session timeout. Google notes there's no limit to how long a session can last, as long as the visitor keeps being active.
Plausible uses the same 30-minute window: a visit "ends when no action is taken for 30 minutes" (Plausible metrics). Because the window is the same, sessions are one of the metrics that line up best between tools.
Engagement rate and bounce rate
GA4 counts an engaged session when it lasts longer than 10 seconds, has a key event, or has 2 or more page or screen views. Engagement rate is the percentage of engaged sessions, and bounce rate is the percentage of sessions that were not engaged (Google: engagement rate and bounce rate). So a GA4 bounce rate of 40% means 40% of sessions were short, single-page and had no conversion.
That's a different number from the old Universal Analytics bounce rate (single-page sessions), and from Similarweb's or Plausible's definitions. Don't compare bounce rates across tools.
| Tool | What counts as a bounce | Effect on the number |
|---|---|---|
| GA4 | Session under 10 s, one page view, no key event (Google) | Lower: reading a page for 11 seconds is not a bounce |
| Universal Analytics (retired) | Single-page session, regardless of time | Higher: a 5-minute read on one page was a bounce |
| Plausible | Landed and left without viewing another page or triggering a custom event (Plausible) | Close to the old UA meaning |
| Similarweb | Estimated from its panel, single-page visits | Modeled, not measured |
Bounce rate also swings with the type of site. Here is Similarweb's estimate for some of the biggest sites in the world, all measured the same way:
Worldwide, all traffic. Same tool and same definition for every site, so the comparison is fair; don't compare these with your GA4 bounce rate. Source: Similarweb Top Websites Ranking
WhatsApp's site is mostly a download and login page, so most visitors leave after one page, which is exactly what it's for. A high bounce rate on a page that answers a question or sends people to an app is not a problem.
Pages per session and engagement time
Views per session (pages per visit) and average engagement time tell you how deep a visit goes. Similarweb's free page for github.com, for example, shows 5.69 pages per visit and an average visit duration of 6 minutes 27 seconds for September 2026 (Similarweb: github.com). A documentation site or a store should see several pages per visit; a single blog post often sees close to one.
GA4 only counts engagement time while the page is in the foreground, so a tab left open in the background doesn't inflate it. Plausible counts single-page visits as 0 seconds and keeps them in the average, which pulls its visit duration down compared with GA4.
Key events and conversion rate
A key event (Google renamed "conversions" to key events in GA4) is any event you mark as important: a signup, a purchase, a demo request, a download. You set them in Admin > Data display > Events by toggling "Mark as key event" on an existing event.
Conversion rate is simple arithmetic:
conversion rate = conversions / sessions x 100
Example: 1,200 sessions, 30 signups
30 / 1,200 x 100 = 2.5%
Some teams divide by users instead of sessions. Both are fine, as long as you always use the same one and say which in your report.
Hits vs visits vs visitors
"Web page hit statistics" come from the early web, when server logs were the only source. A hit is any file request, so one page with 30 images and scripts produces 30-plus hits. Old-school hit counters showed how many visitors to website pages had loaded them, but they really counted page loads, including reloads and bots. Modern tools report views and sessions instead. If a hosting control panel shows "hits", divide mentally by the number of assets per page before comparing it to anything else.
That's also why "how many hits does a website get" is the wrong question today. A site with 10,000 visits a month can easily log over a million hits, because every image, font, stylesheet and API call is one.
Traffic sources and channels explained
Knowing how many people came is half the story; knowing where they came from tells you what's working. GA4 sorts every session into a default channel group using the source and medium of the visit (Google: default channel group):
| Channel | Rule in GA4 (simplified) | Typical examples |
|---|---|---|
| Direct | Source is (direct) and medium is (none) or (not set) |
Typed URL, bookmark, untagged link in an app or PDF |
| Organic Search | Source is a known search engine, or medium is organic |
Google, Bing, DuckDuckGo results; Google's AI Overviews and AI Mode count here |
| Paid Search | Search engine source with a paid medium such as cpc or ppc |
Google Ads, Microsoft Ads |
| Referral | Medium is referral, app or link |
Links from other websites, directories, forums |
| Organic Social | Known social site, or medium such as social or sm |
X, LinkedIn, Reddit, Facebook posts |
Source or medium is email (or e-mail, e_mail) |
Newsletters, tagged email links | |
| AI Assistant | Referrer is on Google's list of AI assistants, medium set to ai-assistant |
ChatGPT, Gemini, Copilot, Grok, DeepSeek |
| Unassigned | No rule matched | Usually a typo in your UTM tags |
Google's help page lists ChatGPT, Gemini, Deepseek, Copilot and Grok as sources for the AI Assistant channel, and explicitly excludes Google's own AI Overviews and AI Mode, which stay in Organic Search. If you want to know how much traffic AI tools send you, this is the row to watch.
Tag your links with UTM parameters
Anything you share yourself (newsletters, social posts, launch announcements) should carry UTM parameters, or GA4 has to guess. A tagged link looks like this:
https://example.com/pricing?utm_source=newsletter&utm_medium=email&utm_campaign=nov-launch
Because the medium is email, GA4 puts that session in the Email channel, and the campaign name shows up in Reports > Acquisition > Traffic acquisition when you switch the dimension to "Session campaign". Keep values lowercase and consistent: Email and email look the same in the channel rules (matching is case-insensitive), but they show up as two rows in source/medium reports. To check what a long tracking link actually contains, paste it into the URL parser.
Where data on website traffic comes from
All data on website traffic comes from one of three places, and each one sees a different slice of reality:
- A tag in the page (GA4, Plausible, Matomo, Cloudflare Web Analytics, Clarity). Accurate for visitors who load the script; misses people who block it.
- Server or CDN logs (Nginx, Apache, Cloudflare, Vercel). See every request, including bots and visitors who block scripts.
- Panels and models (Similarweb, Semrush, Ahrefs). Estimate traffic for any domain from browsing panels, search rankings and partner data. The only way to see other people's sites, and the least precise.
How to check your own website visitor statistics
To check site statistics for your own domain, you need access to at least one first-party source: an analytics tool, Search Console or your logs. Here's where each number lives.
In Google Analytics 4
Google Analytics web traffic reports all live under Reports in the left menu:
- Visitors over time: Reports > Reports snapshot, or Reports > Engagement > Overview.
- Where visitors come from: Reports > Acquisition > Traffic acquisition (sessions by channel) and User acquisition (first-visit channel).
- Most viewed pages: Reports > Engagement > Pages and screens.
- Landing pages: Reports > Engagement > Landing page.
- Who's on the site now: Reports > Realtime shows how many visitors on a website are active in the last 30 minutes. This is the fastest way to check site visitors right now, for example while a launch post is live.
Set the date range in the top-right corner, and switch on Compare to put this period next to the previous one. To check site visitors per day, look at the line chart on any report: it plots daily values by default for ranges of a few weeks. Google says standard properties typically finish processing daily data in about 12 hours, and intraday data in 2 to 6 hours, so yesterday's numbers can still move (Google: data freshness).
You can also check site stats on your phone: the Google Analytics app shows the same Home, Realtime and Reports screens.
Pull GA4 numbers with code
If you'd rather not click through menus every week, the GA4 Data API returns the same numbers as JSON. A minimal Node.js script with the official @google-analytics/data package:
// npm install @google-analytics/data
// Auth: a service account added as a Viewer on the GA4 property,
// with GOOGLE_APPLICATION_CREDENTIALS pointing at its JSON key.
const { BetaAnalyticsDataClient } = require('@google-analytics/data');
const client = new BetaAnalyticsDataClient();
async function main() {
const [res] = await client.runReport({
property: 'properties/123456789', // your GA4 property ID
dateRanges: [{ startDate: '28daysAgo', endDate: 'yesterday' }],
dimensions: [{ name: 'sessionDefaultChannelGroup' }],
metrics: [{ name: 'sessions' }, { name: 'activeUsers' }, { name: 'engagementRate' }],
});
for (const row of res.rows) {
console.log(row.dimensionValues[0].value, row.metricValues.map(m => m.value).join(' | '));
}
}
main();
That prints one line per channel with sessions, active users and engagement rate for the last 28 days, which is most of a weekly traffic report.
In a simpler tool
Plausible, Umami, Fathom and Cloudflare Web Analytics put visitors, pageviews, sources, pages and countries on a single dashboard, with no menus to learn. Hosted visitor-counter services such as Web-Stat, with plugins for WordPress, Wix and Shopify, work the same way. Website builders also have their own stats screens, so you may already have the numbers without adding any tag. See the best website traffic analysis tools for setup and pricing, this comparison of web analytics services for feature-by-feature differences, or browse analytics tools on DevHunt.
From your server logs
Your web server's access log has every request, including visitors who block analytics scripts. A log analyzer such as GoAccess turns it into a report:
# HTML report from an Nginx or Apache log in the standard "combined" format
goaccess /var/log/nginx/access.log -o report.html --log-format=COMBINED
# Rough count of distinct IPs in today's log, skipping obvious bots
grep -viE 'bot|crawler|spider' /var/log/nginx/access.log | awk '{print $1}' | sort -u | wc -l
Logs also include bots, so filter them before trusting the totals. Cloudflare's 2025 data shows how big that slice is: as of December 2, 2025, humans generated 47% of HTML requests and non-AI bots 44%, with AI bots averaging 4.2% over the year (Cloudflare Radar 2025). A distinct-IP count is also only a rough proxy for visitors: an office or mobile network can put many people behind one IP. If you need to tell a real browser from a scripted client, check its user agent with the user agent parser.
SEO and web traffic metrics in Search Console
For search traffic, Google Search Console's Performance report shows clicks (the number of times a user clicked your site from Google Search results), impressions (how many times your site appeared in results), CTR (clicks divided by impressions) and average position (the average position of your topmost result) (Google: Performance report). It's the most reliable view of organic search, because it's Google's own data rather than a script on your page.
Search Console and GA4 won't match exactly. Search Console counts clicks, GA4 counts sessions that loaded the tag, and Google says the Performance report "doesn't show all data" for privacy reasons (rare queries are hidden) (Google: about Search Console data). Use Search Console for "which queries and pages bring searchers" and GA4 for "what they did next".
How to analyze website traffic: a 20-minute weekly routine
Most people open their analytics, look at the big number, and close it. To actually analyze website traffic, ask the same five questions every week, in the same order. Checking website traffic once a week is enough for most small sites; daily checks mostly show noise.
- Is the trend up or down? Compare the last 28 days with the previous 28 days (not last week with this week, because weekdays and weekends differ). A change under about 10% on a small site is usually noise.
- Which channel moved? In Traffic acquisition, sort by the change in sessions. If total traffic dropped 20% and Organic Search dropped 40% while everything else held, the problem is search, not the site.
- Which landing pages moved? Open Landing page, compare periods, and find the two or three pages that explain most of the change. Traffic changes are almost always concentrated in a few pages.
- Did engagement change? If sessions went up but engagement rate fell sharply, check where the new traffic came from: a viral post or a bot spike brings a lot of quick, one-page visits.
- Did key events follow? Traffic that doesn't turn into signups or sales isn't the goal. Check conversion rate by channel: email and direct often convert far better than social.
Write one sentence per question in a running doc. After a couple of months you'll know your site's normal range, which is more useful than any industry average.
How to read a traffic report: a worked example
Say your GA4 Traffic acquisition report for October shows:
| Channel | Sessions | Change vs September | Engagement rate | Key events |
|---|---|---|---|---|
| Organic Search | 6,200 | +4% | 58% | 93 |
| Direct | 2,100 | +35% | 41% | 40 |
| Referral | 900 | -2% | 63% | 27 |
| Organic Social | 600 | +80% | 22% | 2 |
| AI Assistant | 150 | +50% | 71% | 6 |
(Illustrative numbers.) Reading it: search is stable and converts at 1.5% (93 / 6,200). The jump in Direct together with Organic Social suggests a post that spread through chat apps, and its low engagement says those visitors skimmed and left. Referral and AI Assistant traffic is small but highly engaged, so it's worth finding which pages those visitors landed on and doing more of that. Nothing here needs fixing; one thing (the referral and AI sources) needs more of the same.
Why did my traffic drop? A checklist
| Symptom | Likely cause | Where to check |
|---|---|---|
| All channels down to almost zero overnight | Tag removed or broken in a deploy, or consent banner change | GA4 Realtime while you browse the site; page source for the tag |
| Only Organic Search down | Ranking loss, Google update, page removed or noindexed | Search Console Performance and Pages reports |
| Only one page down | That page lost a ranking or a link | Search Console, filtered by page |
| Direct spikes, engagement near 0% | Bot traffic | Realtime by country and city; server logs |
| Sudden "Unassigned" rows | Misspelled or missing UTM medium | Source/medium dimension |
| Small dip every year in the same month | Seasonality | Compare with the same month last year |
How many visitors does a website get?
There's no typical number, and the honest answer is "far fewer than you'd think". In Ahrefs' study of about 14 billion pages, 96.55% got zero traffic from Google and 1.94% got between one and ten visits a month (Ahrefs). At the other end, Similarweb's free page for github.com shows 645.5M "Total Visits Last 3 Months" as of September 2026, roughly 215 million visits a month (Similarweb). Traffic follows a power law: a tiny number of sites get most of the visits, so any "average website traffic" figure is dragged up by giants and means nothing for a normal site.
How many visitors website owners should expect depends on what the site is and how it gets found:
- A new site or a personal blog: a few visitors a day for the first months is normal, mostly from you, friends and the occasional search. Search traffic usually takes months to build.
- A small business site: what matters is how many site visits turn into calls, bookings or orders, not the raw count. A local plumber with 300 visits a month and 15 calls is doing fine.
- A SaaS or developer tool: traffic comes in spikes from launches (a launch day on DevHunt, a Hacker News post) on top of a slowly growing search baseline. A jump in daily visitors on website launch days, followed by a drop back to a slightly higher baseline, is the normal shape.
- Content sites: traffic tracks the number of pages that rank. Here, average page views for a website matter less than how many pages earn any search traffic at all.
So how many visitors to a website is "good"? Enough to hit your goal at your conversion rate. If you need 20 signups a month and 2% of sessions sign up, you need about 1,000 sessions a month. Work backwards from that instead of from benchmarks.
Compare your site with similar sites in GA4 benchmarking
If you want a fair comparison, GA4 has one built in. Benchmarking compares your property with a peer group of similar businesses for metrics such as new user rate, engagement rate, average session duration, views per session and bounce rate (Google: benchmarking). It shows the peer median and a shaded range from the 25th to the 75th percentile on the Home page cards.
To turn it on, enable Modeling contributions & business insights under Admin > Account settings. Benchmarking data starts from May 30, 2024, is refreshed every 24 hours, and only appears once your peer group has enough properties and your property has enough users. You can change the peer group (industry category and subcategory) from the benchmarking badge on a card.
This is the closest thing to "average website traffic by industry" that's based on real analytics data rather than a survey or a guess, and it's free.
- July 1, 2023Standard Universal Analytics properties stop processing data. GA4 and its engagement-based bounce rate become the default (Google).
- May 30, 2024Earliest date covered by GA4 benchmarking data (Google).
- July 1, 2024One-time processing extension for Universal Analytics 360 properties ends.
- Week of July 1, 2024Universal Analytics interface and API shut down; unexported UA data is no longer accessible (Google).
How to check monthly visitors of another website
You can't get real numbers for a site you don't own, unless the owner publishes them (some do, in media kits, open startup dashboards or investor updates). Everything else is an estimate:
- Similarweb. Search for the domain at similarweb.com/website. No account is needed, and the free checker allows 3 searches a day. The free page shows "Total Visits Last 3 Months", the change from last month, bounce rate, pages per visit, visit duration, top countries and the top traffic channels. Divide the three-month total by three for a rough monthly figure.
- Semrush Website Traffic Checker. The free checker shows total visits, unique visitors, pages per visit, average visit duration and bounce rate, and runs once a day without an account. A free Semrush account includes "10 reports per day across all vital tools" (Semrush plans).
- Ahrefs traffic checker. Free, but it gives organic search traffic estimates only, not total visits. Paid traffic is in the paid Site Explorer.
| Estimator | Free limit (as of November 2026) | Estimates | Best for |
|---|---|---|---|
| Similarweb | 3 searches a day, no account | Total visits (all channels), engagement, countries, top channels | Total visits for medium and large sites |
| Semrush | 1 check a day without an account; 10 reports a day with a free account | Total and unique visitors, engagement, organic and paid search | A second opinion on total visits |
| Ahrefs | Free checker | Organic search traffic and top keywords | Search traffic only |
Estimating web traffic this way works best when you compare several sites in the same tool: "site A gets roughly twice site B's traffic" is far more reliable than "site A gets 412,000 visits".
How accurate are traffic estimates?
None of these vendors publishes an error rate. Similarweb says it combines analytics data shared by site owners, a panel of contributing devices, public data and partner data, and models the rest (Similarweb data sources). Semrush describes its numbers as modeled from browsing patterns, search signals and partner data, and says they "should be viewed as directional rather than exact" (Semrush). Ahrefs' organic estimates come from rankings and keyword volumes. In practice:
- Estimated site traffic is closer for large sites and unreliable for small ones (a few thousand visits a month may show no data, or a number that's far off).
- Two tools often disagree by a large factor on the same site.
- Trends and comparisons between similar sites are more trustworthy than absolute numbers.
If the number matters (buying a site, pricing a sponsorship), ask the owner for read access to their analytics or a screenshot with a date range.
Why your stats don't match between tools
- Blocked scripts. Ad blockers and privacy browsers block many trackers, and declined consent banners stop GA4 from setting cookies. Developer audiences block more than most.
- Different definitions. GA4 counts active users and 30-minute sessions; Plausible counts unique visitors per day with a rotating hash (Plausible data policy); server logs count requests.
- Bots. Logs include crawlers; most analytics tools filter known bots. On the web as a whole, bots made about as many HTML requests as humans in 2025 (Cloudflare Radar).
- Time zones and processing delays. Check that both tools use the same reporting time zone, and don't compare today's numbers while GA4 is still processing.
- Search Console vs analytics. A click in Search Console isn't always a session in GA4: the visitor may leave before the tag loads, or block it.
Pick one tool as the source of truth for each metric and track its trend over time. If you're choosing between the two most common options, see Google Analytics vs Plausible. For the bigger picture of what to track and how to set up goals, read Web analytics for beginners, and if you build software, web analytics tools for developers covers APIs, self-hosting and event tracking.
A few web analytics statistics worth knowing
From W3Techs' traffic analysis survey, as of 4 November 2026:
- Google Analytics is used on 46.8% of all websites, an 81.7% share of sites that use any traffic analysis tool.
- Meta Pixel is second at 8.4%, followed by Microsoft Clarity (4.2%) and Cloudflare Web Analytics (4.0%).
- Matomo, the most common open-source option in the survey, is on 1.4%.
Sites can run several tools, so the shares add up to more than 100%.
Share of all websites W3Techs surveys. A site can use several tools. Source: W3Techs
Two more numbers that shape how you should read your own reports:
- Mobile is the majority. StatCounter puts worldwide web traffic at 59.9% mobile, 38.51% desktop and 1.59% tablet for October 2026 (StatCounter). Developer tools and B2B sites usually skew desktop, so check your own split under Reports > Tech before you decide what to optimize.
- Google is most of search. Google held 91.05% of worldwide search engine market share in October 2026, Bing 4.8% (StatCounter). For most sites, Organic Search in GA4 is mainly Google, which is why Search Console is worth setting up first.
FAQ
How do I check the monthly visitors of a website?
For your own site, open your analytics tool and set the date range to the last month (in GA4, Reports > Acquisition > Traffic acquisition). For someone else's site, look it up on Similarweb or Semrush and treat the result as an estimate. Similarweb shows a three-month total, so divide by three.
What's the difference between a hit, a visit and a visitor?
A hit is any file request to the server (pages, images, scripts). A visit, or session, is one period of activity on the site. A visitor, or user, is the browser or device that made the visits. One visitor can have many visits, and one visit can produce dozens of hits.
What is a good bounce rate?
There's no universal benchmark, because it depends on the page and the tool's definition. A blog post that answers a question in one page will have a higher GA4 bounce rate than a pricing page. Compare a page with itself over time, or turn on GA4 benchmarking to see the bounce rate range for similar businesses.
Why does Google Analytics show fewer visitors than my server logs?
Server logs count every request, including bots and visitors who block scripts. Google Analytics only counts browsers that load its tag, and in many regions only those that accept cookies.
What are the most important web page statistics to track?
Users, sessions by source, top landing pages and conversions. Those four tell you whether people find the site, where they come from, what brings them in and whether they do what you want. For a fuller list and how to report on it, read Web analytics for beginners.
Is it possible to see how many visitors a website gets?
Only the owner sees real numbers. Anyone else can get an estimate from Similarweb, Semrush or Ahrefs (organic search only). Estimates are reasonable for big sites and often far off for small ones, so use them to compare sites rather than as exact counts.
How many visitors to a website is good?
Enough to reach your goal at your conversion rate. If 2% of sessions convert and you want 20 customers a month, you need around 1,000 sessions a month. Since 96.55% of pages get no Google traffic at all (Ahrefs), a site with steady, growing traffic is already ahead of most of the web.
What is the average website traffic?
There isn't a meaningful one. Traffic is so concentrated that a handful of huge sites drag any average far above what a typical site gets. Compare your site with its own history, or use GA4's benchmarking feature to see medians for similar businesses.
How can I see how many visitors to my website come each day?
In GA4, open Reports > Reports snapshot and look at the users chart, which plots daily values; use Realtime for the last 30 minutes. Plausible, Fathom and Cloudflare Web Analytics show daily visitors on their main dashboard.
Which Google tool is used to track website traffic?
Google Analytics 4 tracks all traffic to your site (every channel, plus what visitors do), and Google Search Console tracks traffic from Google Search (clicks, impressions, queries). They're both free, and most sites should use both.
How do I find website traffic statistics for a competitor?
Enter the competitor's domain in Similarweb's free checker for total visits, channels and countries, and in Ahrefs' free checker for search traffic and top keywords. Compare a few competitors in the same tool and look at relative size and trend, not the exact figure.
Does traffic from ChatGPT show up in Google Analytics?
Yes, when the visitor clicks a link and the referrer is passed. GA4 groups visits from ChatGPT, Gemini, Copilot, Grok and DeepSeek in the AI Assistant channel, while clicks from Google's AI Overviews and AI Mode stay in Organic Search (Google).