Key Terminology

A few of these words have already gone past: hits, visits, visitors. Before they get defined, pause for a second and try it yourself, because they are simpler than they sound. A visitor is a person browsing your site or app. A visit is one trip they make to it. And a hit is a single action inside that trip: a page load, a download, any one thing you are tracking. Those three are your baseline traffic metrics, and almost every report stands on them.

Those everyday definitions are close enough to start with, but Adobe means something exact by each, and the precision is not pedantry, it is the difference between a report you can defend and one that quietly misleads. So make them sharp. Picture a customer in a physical store. A single action, picking up one item, glancing at one shelf, is a hit: one data collection call to Adobe, a page view or a click. A whole trip through the store, from walking in to walking out, is a visit: a sequence of hits that ends after 30 minutes of inactivity. And the shopper themselves, recognized by their loyalty card across every trip they ever make, is the visitor: a unique cookie or device ID, not necessarily a unique human (more on that trap below).

This nesting is not just trivia; it's the spine of segmentation. Every segment you build later asks the question at one of these three levels: "show me the hits where X," "the visits where X," or "the visitors who ever did X." Same data, three very different answers.

A visit ends four ways, and only one of them is famous

The 30-minute rule is the one everybody learns, and it is the one that fires almost every time, which is why it is fair to define a visit by it. It is not the only way a visit ends. Adobe closes a session under four conditions, and the other three are worth knowing because they are the ones that produce numbers nobody can explain.

The middle two are the ones that catch experienced people, and they catch them quietly. A single-page application that sends a tracking call on every route change, filter, and lazily loaded panel accumulates hits at a rate a traditional page-view site never approaches, and a long media session is instrumented far more densely than the page it sits on. Nothing errors, nothing is flagged, and one person's single sitting is now two visits in the report.

One more piece of visit behavior surprises people, and it is worth carrying because it ends an argument rather than starting one. A visit does not end at midnight. A session running from 11:45 PM to 12:10 AM is one visit, not two. In a daily trend it appears on both days, because it genuinely happened on both, but the total for the period counts it once. That is why the daily visit numbers in a trend will not add up to the month's total, and the gap is not a defect. It is the same deduplication that makes unique visitors behave the way they do, applied to sessions.

What you actually want to know

Step back from the vocabulary for a moment. When you say you want to track what customers do on your site, what are you really after? The people, their visits, and their actions, yes, that is the baseline. But you want far more than counts. You want to know the journey they took through the site. You want to know which groups they fall into. You want to know their technical setup, the browser, the device, the operating system. If you sell things, you want to know which products they look at and buy. And above all, you want to know where they came from, and what brought them to you in the first place.

That last one quietly drives the whole business. Picture a fashion site that keeps promoting jeans while its visitors keep reaching for t-shirts. Miss that signal and you lose sales you never knew were on the table. Knowing what your visitors actually want, and where they arrive from, is exactly what lets you shape the site, the merchandising, and the marketing spend around them.

Here is the useful part: every one of those things you want to know has a name in Adobe Analytics. The descriptors, the journey, the group, the browser, the product, the traffic source, are dimensions. The counts and sums, how many visits, how much revenue, are metrics. Which is exactly the distinction that follows.

Dimensions and metrics: the two axes of every report

Open any table in Analysis Workspace and you're looking at two kinds of thing. Dimensions are the descriptive rows, the "by what": page name, campaign, browser, country. Metrics are the numbers filling the table, the "how much": page views, visits, revenue, orders. "Revenue by campaign" is simply one metric broken down by one dimension. Almost everything you build is some set of dimensions crossed with some set of metrics. Hold that and Workspace stops feeling like a maze. The variables module goes deep on both.

eVar vs prop: the distinction people get wrong

This one quietly causes more bad reports than any other. A prop (traffic variable) is written in pencil: it captures what was true on a single hit, then it's erased. Great for pathing and traffic patterns, but it does not follow the visitor and gets credit for nothing that happens later. An eVar (conversion variable) is a sticky label the visitor keeps wearing: set it once, say campaign = summer-sale, and it persists for however long you configure it, so when that visitor converts three pages or three days later, the eVar is still there to receive the credit. That persistence is exactly what makes attribution possible.

The practical rule of thumb: if you want to know how someone moved through the site, that's prop territory; if you want to credit an outcome back to something earlier, that's an eVar. The variables module goes all the way down on scope, expiration, and allocation.

Quick reference

With the vocabulary in hand, a different question tends to arrive next, and it usually arrives from somebody holding a budget. Why this tool, and not the one already in the building? Adobe Analytics vs Google Analytics takes that question seriously rather than defensively.