Unique visitors vs. visits vs. pageviews
One person, two coffee breaks and seven pages is 1 visitor, 2 visits and 7 pageviews. Which number to quote, when, and why no two tools ever agree.
Three numbers, one morning of traffic
Someone finds your site over morning coffee, reads the homepage and two blog posts, and closes the tab. After lunch they come back and read four more pages. That's 1 unique visitor, 2 visits, 7 pageviews — three different numbers describing one person's day, and all three appear on every analytics dashboard ever built. Most confusion about traffic data is just these three being used interchangeably.
The hierarchy: visitors are people (roughly), visits are sittings, pageviews are page loads. Each layer counts the one below it multiple times, which is why pageviews is always the biggest, most flattering number — and the one quoted in press releases.
Unique visitors — people, approximately
A unique visitor is one distinct person in the period — the honest measure of reach. The catch is in how tools recognize 'the same person again'. Cookie-based tools set an identifier in the browser; cookieless tools like Bigdelta derive a short-lived anonymous identifier instead, no consent banner required. Both approximate: the same human on a laptop and a phone counts twice, and a cleared browser becomes a stranger.
So visitor counts are a slight overcount of actual humans, and — importantly — 'unique' is period-relative: a loyal daily reader is one weekly unique but seven daily uniques, which is why summing a week of daily uniques produces a number bigger than the week's own figure. That's not a bug; it's the definition.
Visits — sittings with a timeout
A visit (or session) is one continuous stretch of activity. The industry convention: 30 minutes of inactivity ends it, and the next pageview starts a new one. The morning-coffee reader above generated two visits because lunch outlasted the timeout. The 30 minutes is pure convention — someone reading one long essay for 40 minutes without clicking may register as two visits, or one, depending on the tool's heartbeat tracking.
Visits mostly earn their place in ratios: visits per visitor is rough loyalty, pageviews per visit is rough depth, and bounce rate is the share of visits that ended after one page. A jump in visits with flat visitors means the same people returning more often — a very different story than new reach, and one the visitor count alone would hide.
Pageviews — loads, not people
A pageview is one page render. It measures activity volume, not audience: one obsessive fan refreshing a leaderboard can outscore a hundred calm readers. Its modern complication is the single-page app, where navigation happens in the browser without a traditional page load — a visitor can browse six screens and register one 'pageview' unless the analytics tool counts client-side route changes, which good ones do and server logs don't. This is a big part of why platform request counters and real analytics disagree so sharply.
Pageviews still have honest jobs: spotting which content actually gets read, and capacity planning. Just never let the biggest number stand in for 'audience' — 50,000 pageviews can be 40,000 people or 900.
Why your tools disagree
Install two analytics tools on the same site and they will report different numbers — routinely by 10–30%. The gap is structural, not a bug: tools differ on bot filtering, adblockers block some scripts and not others, consent banners suppress cookie-based counting for whoever declines, session timeouts vary, and SPA handling varies more. Server-side request counting sits furthest from the truth about humans, inflated by every crawler and uptime checker on the internet.
The practical stance: pick one tool as the source of truth and read trends, not absolutes. Whether last week was 4,100 or 4,700 visitors matters less than whether the direction held after the launch — and a trend is trustworthy within one tool even when the level is debatable across tools.
The practical takeaway
Quote unique visitors when the question is reach, visits when it's engagement frequency, pageviews when it's content or load — and keep the labels straight, because a '50,000 hits' claim means almost nothing until it's unpacked into these three. The ratios between them are where the actual insight lives: rising pages-per-visit says the site got stickier; rising visits-per-visitor says it became a habit. One person reading seven pages and coming back after lunch is a better day than seven strangers bouncing — and only the full trio of numbers can tell those two days apart.
