The Complete Google Analytics 4 Glossary: Every Metric, Dimension & Report Explained

GA4 vs. Universal Analytics: Why the Vocabulary Changed

Universal Analytics (UA) sunset on July 1, 2024. Google Analytics 4 replaced it with a different data model. UA was built around sessions and hit types, where every interaction fell into a fixed bucket like pageview, event, transaction, or social. GA4 flattened all of that into a single unit called an event. A page view is an event. A scroll is an event. A purchase is an event. A form submission is an event.

The change matters because a lot of the language you still see in tutorials, exam questions, and older Stack Overflow threads comes from UA. Bounce rate got redefined. Goals became conversions and then became key events. Views disappeared. Hit types were retired. If you learned analytics before 2023, expect to unlearn a few things as you read.

For agencies running measurement for clients, this shift also changed what feeds attribution back into SEO strategy and paid media planning. GA4 leans on modeled data and machine learning to fill gaps that older cookie-based tracking used to cover directly.

Users, Sessions & Visitors 10 terms

What is a user in Google Analytics?

A user is a unique visitor to your property. GA4 identifies users through one of three signals: a first-party cookie called the client ID, a User-ID you send when someone logs in, or a Google signal from a Chrome user with ads personalization enabled. The default identity is the client ID, which is browser-specific. Somebody who browses on their phone at lunch and their laptop after dinner without signing in is counted as two users.

GA4 reports three flavors of user: Total Users (everyone GA4 saw in the date range), Active Users (people with at least one engaged session), and New Users (first-time visitors during the range). The Active Users number is what appears on the home card labeled "Users."

New vs. returning users in Google Analytics

A new user is someone whose first session ever with your property falls inside the report's date range. A returning user is someone with at least one prior session before the range began. GA4 uses the New/Established user dimension to slice this. If you compare "new vs returning users google analytics" reports across UA and GA4, expect the numbers to differ because GA4's identity graph now includes signed-in Google signals data.

User vs. new user in Google Analytics

Total Users includes new users. New Users is a subset. If a report shows 10,000 Users and 6,500 New Users, that means 6,500 of the 10,000 were seeing your site for the first time in that date range. The other 3,500 had at least one earlier session on record.

User vs. session in Google Analytics

A user is the person. A session is a single visit. One user can start dozens of sessions over a month. In GA4, a session begins when the session_start event fires and ends after 30 minutes of inactivity by default. Sessions also reset at midnight in the property time zone, and a new campaign source in a fresh URL starts a new session.

The ratio of sessions to users tells you engagement depth. Two sessions per user is typical for a retail site. A B2B blog might see 1.2 sessions per user because most traffic reads once and leaves.

Google Analytics users vs. new users

This is the same distinction as above, phrased the way exam questions ask it. Users is the total. New Users is the first-timer subset. Returning Users is the difference. In GA4's Traffic acquisition report, the "Users" column is Total Users, and "New users" is a separate column right next to it.

Google Analytics session vs. users

Sessions are always higher than or equal to Users, never lower. If sessions equal users, every visitor came exactly once during the range. If sessions run 2x or 3x higher than users, you have strong repeat-visit behavior, a good signal for content sites and a critical one for content marketing programs.

Session

A single visit, bounded by inactivity or campaign change. Session length is the time between the first and last event in the session. Sessions with only a single event that lasts less than 10 seconds and produced no conversion are considered non-engaged.

Engaged session

A session that lasted at least 10 seconds, had at least one conversion event, or produced at least two pageviews or screenviews. Engaged sessions replaced the old "non-bounce" concept from UA. Engagement rate is the percentage of sessions that were engaged.

Bounce rate (GA4 version)

Google reintroduced bounce rate in GA4, but it now means the inverse of engagement rate. If engagement rate is 63%, bounce rate is 37%. It no longer depends on hitting a single page, which is what UA measured. A user who reads one long article for four minutes and leaves is engaged in GA4 and bounced in UA.

Client ID

The anonymous first-party cookie value GA4 uses to identify browsers. It expires after two years of inactivity by default. Clearing cookies, switching browsers, or using private mode all break client ID continuity, which is one of the main reasons cross-device attribution requires a User-ID or Google signals.

Dimensions & Metrics 7 terms

What is a dimension in Google Analytics?

A dimension describes an attribute of your data. It answers "what" or "where" or "who." Country is a dimension. Page path is a dimension. Device category, Browser, Session source, City, Landing page, and Event name are all dimensions. Dimensions are always strings or categorical values, never numbers you would add up.

What is dimension in Google Analytics (short version)

Same term, same answer. A dimension is the descriptive column in any report. If you can group your data by it, it is a dimension.

What is a metric in Google Analytics?

A metric measures your data quantitatively. Users, Sessions, Views, Event count, Conversions, Revenue, Average engagement time, and Bounce rate are all metrics. Metrics are numbers. You add them, average them, and compare them.

Every GA4 report pairs at least one dimension with at least one metric. The Traffic acquisition report, for example, uses Session default channel group as the dimension and Users, Sessions, and Engagement rate as the metrics.

What is a secondary dimension in Google Analytics?

A secondary dimension adds a second grouping to an existing report. If your Traffic acquisition report shows Organic Search as a row, adding Landing page as a secondary dimension expands that row into every organic landing page. You add secondary dimensions using the blue plus button next to the primary dimension name in any GA4 Explore or standard report table.

Custom dimension

A dimension you define yourself and send with an event as a parameter. Common examples: logged_in_status, membership_tier, article_author, product_category. Custom dimensions must be registered in GA4's Admin panel before they can appear in reports, and they only start collecting data from the moment you register them.

Custom metric

A numeric parameter you register as its own metric. Examples: quiz_score, video_percent_watched, engagement_points. Same registration flow as custom dimensions, same forward-only collection rule.

What is conversion rate in Google Analytics?

GA4 offers three conversion-rate metrics. User conversion rate is unique users who triggered a key event, divided by total users. Session conversion rate is sessions that included a key event, divided by total sessions. First-time purchaser conversion is a specific ecommerce version. If someone asks "what is conversion rate in google analytics," they usually mean session conversion rate for ecommerce or user conversion rate for lead-gen. Both live under the Metrics library and both need at least one event marked as a Key Event to calculate.

Events, Conversions & Hits 6 terms

What is an event in Google Analytics?

An event is any user interaction GA4 captures. Every page load, click, scroll, form submit, and video play is an event. GA4 divides events into four buckets: automatically collected events (first_visit, session_start), enhanced measurement events (scroll, click, file_download, video_start), recommended events that Google names but you must implement (login, sign_up, purchase), and custom events that you name yourself for anything unique to your business.

Every event carries parameters. A purchase event carries transaction_id, value, currency, and items. A file_download event carries file_name and file_extension. Parameters are what let you slice events into meaningful reports.

Key event (the new "conversion")

In late 2024 Google renamed GA4 "conversions" to key events to separate GA4 conversion tracking from Google Ads conversion tracking. Any event you flag as a key event becomes eligible for conversion-rate reporting. Purchase, sign_up, and lead_form_submit are common flags. Google Ads still calls its own version conversions, and imports from GA4 happen at the key event level.

Conversion (Google Ads sense)

If your team says "conversion" and means the number in Google Ads, they mean a Google Ads conversion action, which often comes from GA4 key events but lives in a separate system with its own attribution model. Reconciling the two is the single most common analytics support ticket we see for paid media clients.

What kinds of hits does Google Analytics track?

Older certification exams still ask this, so here is the honest answer. In Universal Analytics, GA tracked pageview hits, event hits, ecommerce (transaction and item) hits, and social interaction hits. Some course materials also list screenview hits (mobile apps), timing hits, and exception hits. GA4 retired the hit model entirely and replaced it with events. If you get this question on a legacy exam, answer with the UA hit types. If you get it in a real interview, mention both.

Assigning a value to a Google Analytics goal

In UA, assigning a monetary value to a goal let you compare non-transaction conversions like newsletter signups against ecommerce revenue in the same reports. GA4 preserves this: when you mark an event as a key event, you can send a value parameter with it. Assigning a value lets you calculate the total revenue impact of non-purchase events and feed sensible bid strategies back to Google Ads.

Parameter

A piece of contextual data attached to an event. Every event has at least three automatic parameters (page_location, page_referrer, page_title for a page_view) and can carry up to 25 custom parameters. Parameters become dimensions or metrics only after you register them in Admin.

Pages, Landing Pages & Properties 5 terms

What is a landing page in Google Analytics?

A landing page is the first page a user sees during a session. If someone arrives from an organic Google search on /pricing/, then clicks through to /features/, /pricing/ is the landing page for that session. GA4 exposes this as the Landing page dimension and pairs it with Sessions, Users, Engagement rate, and Key event count in the Landing page report under Life cycle > Engagement.

Landing page analysis is one of the highest-value GA4 workflows for SEO, because every strong organic landing page is a candidate for expansion, internal linking, and CRO. This is exactly where our revenue marketing and CRO playbooks start.

Entrances

"What are entrances in google analytics" is a UA term. In UA, Entrances counted the number of sessions that started on a given page, effectively "how many times was this page the landing page?" GA4 does not have a metric literally called Entrances. The equivalent view is Sessions in the Landing page report, since every session has exactly one landing page.

What is property name in Google Analytics?

A property is the data container that holds all of your GA4 measurement for a given website or app. The property name is the human-readable label you give it, usually something like "eMac Media (Website)" or "Client Name (GA4)." Each property has a Measurement ID (starts with G-) that you install on the site. Under a property sit data streams (Web, iOS, Android). Above the property sits the account, which can hold up to 2,000 properties on standard accounts.

Pageview

The automatically collected page_view event. Fires every time a user loads a page (and again on virtual page changes in single-page apps if you configure enhanced measurement correctly). Pageviews are the raw currency of content analytics and feed into the Views metric on every landing-page and page-path report.

Views

The metric formerly known as Pageviews. In GA4's UI, "Views" counts both web page_view events and app screen_view events. If a session includes 5 pageviews and 2 screenviews across web and app, Views for that session equals 7.

Sources, Channels & Referral Traffic 9 terms

Google Analytics direct vs. organic

Direct traffic is any session where GA4 cannot identify a referring source. Someone typing your URL, clicking a bookmark, or arriving via an app with no referrer set all count as direct. Organic traffic (specifically Organic Search) is any session that came from an unpaid search-engine result. Google, Bing, DuckDuckGo, and Yahoo referrers with no gclid or paid identifier land in Organic Search.

Direct traffic often gets inflated by broken UTM tagging on email links, mis-configured social sharing, and https-to-http protocol drops. If your direct traffic looks unnaturally high, audit your campaign tracking before drawing conclusions.

What is referral in Google Analytics?

A referral is a session that came from clicking a link on another website. The referring domain is captured as the Session source. If someone reads a Forbes article about your product and clicks through, Forbes.com is the referral source and the Referral channel picks it up.

What is referral traffic in Google Analytics?

Same concept, just the aggregated view. Referral traffic is the sum of all sessions from external websites that are not search engines, social networks, or paid ads. It is a core signal for measuring the ROI of link building and digital PR: quality referring domains often deliver both authority and direct clicks.

What sources are available in Google Analytics?

GA4 recognizes any string in a session's referrer or utm_source parameter as a source. The most common values you will see in reports: google, bing, yahoo, duckduckgo (search engines), facebook, l.facebook.com, m.facebook.com, instagram, linkedin, twitter, x.com, tiktok, pinterest, reddit, youtube (social), newsletter, email, mailchimp, klaviyo (email, if tagged), direct (no referrer), and any other domain that linked to you. You can also send custom utm_source values on any inbound URL.

What is not considered a source in Google Analytics by default?

Internal traffic from your own domain is not considered a source. GA4 automatically excludes same-domain referrers so you do not credit yourself for every internal click. Cross-domain traffic between properties you own (like site.com and shop.site.com) also needs cross-domain measurement configured, or the second domain will show up as a referral to the first. Offline actions, phone calls, and in-store visits are also not sources by default. They only become measurable if you push them into GA4 with the Measurement Protocol.

What is l.facebook in Google Analytics?

l.facebook.com is Facebook's link-shim service. When someone clicks a link inside Facebook's desktop feed, Facebook routes the click through l.facebook.com before sending them to the destination. GA4 sees l.facebook.com as the referrer, which is why it appears alongside facebook.com in your Referral or Social channel reports.

What is m.facebook.com in Google Analytics?

m.facebook.com is Facebook's mobile web version. Clicks originating from the Facebook mobile browser (not the native app) come through as m.facebook.com referrers. You can group l.facebook.com, m.facebook.com, and facebook.com under one channel using a custom channel group.

Session source

The specific referrer domain or utm_source value for a session. Example: google, newsletter, facebook.com. Session source is a dimension, not a metric.

Session medium

The category of traffic: organic, cpc, referral, email, social, none (for direct). Combined with Session source, it forms the classic source/medium pair (google / organic, newsletter / email, google / cpc).

Paid Search, CPC, Display & Impressions 5 terms

What is paid search in Google Analytics?

Paid Search is the default channel that groups all traffic from paid ads on search engines. Sessions land here when the utm_medium is cpc, ppc, paidsearch, or when Google Ads auto-tagging sends a gclid parameter. Google Ads campaigns linked to GA4 flow into Paid Search automatically. Bing Ads, when tagged correctly, also lands here.

What is CPC in Google Analytics?

CPC stands for cost per click. As a medium in GA4, "cpc" is the standard utm_medium value for paid search ads. As a metric, CPC is the average amount you paid per click and only shows up in reports if you have Google Ads linked to your GA4 property. In the Google Ads report under Acquisition, GA4 shows Impressions, Clicks, Cost, CPC, CTR, and Conversions side by side.

What are impressions in Google Analytics?

Impressions are the number of times your ad or organic listing was shown, whether or not it was clicked. Ad impressions come from Google Ads via the Ads linking. Organic search impressions come from Search Console via the Search Console linking. GA4 itself does not measure page-level impressions natively. For that you need Search Console data or a scroll-in-view event.

What is display in Google Analytics?

Display is the default channel for banner and image ads shown on the Google Display Network, YouTube, and other display partners. Sessions land in Display when utm_medium is display, banner, or expandable, or when Google Ads reports the campaign type as Display. Display sessions typically show higher session counts and lower engagement than Paid Search because they attract browsers rather than active searchers.

CTR (click-through rate)

Clicks divided by impressions, expressed as a percentage. Appears in the Google Ads and Search Console sections of GA4. A 2% CTR on a paid search ad and a 2% CTR on a Search Console query mean very different things because the intent behind an ad impression differs from an organic result.

Remarketing & Audiences 4 terms

What is remarketing in Google Analytics?

Remarketing (also called retargeting) is the practice of showing ads to people who already visited your site or used your app. GA4 does not run the ads itself. It builds and shares audiences (lists of users who match specific behavior) with Google Ads and Display & Video 360. When someone in the audience browses the web, Google Ads can serve them a targeted ad.

Which remarketing audiences can be defined in Google Analytics?

GA4 lets you build audiences from almost any combination of dimensions, metrics, and events. Common examples: users who viewed a product but did not purchase, users who added to cart in the last 7 days, users who completed a specific key event, users from a specific country, users on a specific device category, users with more than three sessions, users who watched at least 50% of a video. You can also build predictive audiences based on Google's machine-learning models: likely 7-day purchasers, predicted churners, and predicted 28-day top spenders.

What is not a benefit of Google Analytics remarketing?

Common exam trap. Remarketing lets you re-engage past visitors, tailor messaging to their prior behavior, exclude converters to protect ad budget, and build lookalikes for prospecting. What remarketing does not do: it does not increase the size of your top-of-funnel audience, and it does not give you new users who never touched your site. Anyone answering that "growing new-user acquisition" is a benefit of remarketing has the wrong end of the stick. That is a prospecting or awareness benefit, not a remarketing one.

Audience trigger

A rule that fires an event when a user joins a specific audience. Useful for measuring how often users hit qualifying behavior thresholds (like "engaged 3+ times in a week"). Audience triggers show up as regular events in your reports and can themselves be flagged as key events.

Data GA4 Cannot (or Will Not) Track 5 terms

What data is Google Analytics unable to track?

GA4 cannot reliably track: users who reject the analytics cookie, sessions in private or incognito browsing without special config, traffic behind aggressive ad blockers or privacy browsers like Brave with strict shields, offline events that never touch a webhook or Measurement Protocol call, and cross-device behavior when the user is not signed in and Google signals is off. It also cannot distinguish humans from most bot traffic beyond the known-bot list Google maintains.

What data does Google Analytics prohibit collecting?

Google's terms of service prohibit sending any personally identifiable information (PII) to GA4. That means no email addresses, no phone numbers, no full names, no physical mailing addresses, no credit card numbers, no government IDs, and no precise geolocation (like lat/lng from GPS) tied to a specific individual. This applies to page URLs, event parameters, and user properties. If a URL contains "?email=jane@example.com," you must strip it before it reaches GA4 or you risk having your account terminated.

What data is Google Analytics goals unable to track?

Legacy UA phrasing. In UA, goals could not track: revenue after the moment of conversion (like refunds), offline events unless imported, engagement below the goal-triggering action, and any interaction that happened before the tracking snippet loaded. GA4 inherits most of the same limits under its key-event system. Refunds require a separate refund event. Offline conversions require Measurement Protocol or Google Ads offline conversion imports.

What is "not set" in Google Analytics?

(not set) is GA4's placeholder for a dimension value it does not have. You see it a lot in Landing page, Session source, and Session medium reports. Common causes: the session's first event was not a page_view (so no landing page was recorded), a page loaded before GA4 finished initializing, a UTM parameter was missing, or a bot fired an event with no context. High "(not set)" percentages usually mean a tracking implementation gap. Fixing them is one of the fastest ways to improve report quality.

Data thresholding

When a report row would reveal an individual user because the audience is too small, GA4 hides it and shows "(other)" or omits the row entirely. This kicks in most often on Google signals data with fewer than 50 users. If you keep seeing "(other)" in reports, either widen the date range or turn off Google signals if the privacy trade-off is worth it for you.

Filters, Reports & System Integrations 6 terms

What are the options for filtering data in Google Analytics?

GA4 offers three main data-filtering places. First, data filters in the Admin panel, which include Developer traffic filter and Internal traffic filter. These are set to Testing by default so you can preview before applying. Second, report filters, which live at the top of standard reports and let you scope any report to a subset (like device category = mobile). Third, Explore segment and filter controls, which live inside the Explorations builder and let you slice free-form reports with much finer granularity. Comparisons work like non-destructive filters that show multiple slices side by side.

Google Analytics can report on data from which systems?

GA4 accepts data from any system that can send an HTTP request. Native integrations: Google Ads, Google Search Console, Google BigQuery (export), Google Merchant Center, Google Play (via Firebase), Salesforce Marketing Cloud, and any tag-management system like Google Tag Manager. Beyond those, the Measurement Protocol lets you push server-side events from CRMs, ERPs, offline point-of-sale systems, IoT devices, and physical stores. Anything that can make a POST request to Google's measurement endpoint can feed GA4.

Using tracking code, Google Analytics can report on...

The classic quiz answer: with just the on-page tracking snippet, GA4 can report on website usage: pageviews, sessions, users, events, ecommerce, and traffic sources. It cannot report on data that requires other integrations (Google Ads cost, Search Console impressions, offline sales) unless you link those systems separately.

Data stream

The channel that feeds data into a GA4 property. There are three types: Web (for websites), iOS (for Apple apps via Firebase), and Android (for Android apps via Firebase). One property can hold multiple data streams, which is how GA4 supports cross-platform measurement in a single property. Each stream has its own Measurement ID.

Data retention

How long GA4 keeps event-level data available for exploration. The default is 2 months. You can extend it to 14 months in Admin. After that, event-level data purges from Explore and audience-building surfaces, though aggregated standard reports remain available. If you need multi-year raw data, connect the free BigQuery export before the retention window closes.

Explorations

The GA4 tool for building custom reports. Free-form, Funnel exploration, Path exploration, Segment overlap, User explorer, Cohort exploration, and User lifetime are the seven templates. Explorations are where analysts do their real work. Standard reports are what leadership sees.

Author

Michael Timi

Partner & Marketing Manager, eMac Media

Analytics implementation, GA4 migrations, and performance measurement across 200+ industries. Focused on tying data back to revenue.

Editor

Princess Pitts

Director of Communications Strategy, eMac Media

Specializes in editorial strategy, content governance, and brand communications at scale.