Analytics Glossary
Every term you'll encounter in GA4, GTM, and modern web analytics — explained without jargon.
A
ATT (App Tracking Transparency)
Apple's framework introduced in iOS 14.5 that requires apps to request user permission before tracking their activity across other companies' apps and websites. ATT significantly reduced the availability of IDFA for ad targeting and measurement, forcing advertisers to rely on modeled conversions and aggregated data.
Attribution
The process of assigning credit for a conversion to one or more marketing touchpoints. Attribution models range from simple (last-click, first-click) to algorithmic (data-driven). GA4 uses data-driven attribution by default, distributing credit across touchpoints based on their statistical contribution to conversions.
B
Behavioral Modeling
A technique used by GA4 to fill gaps in data caused by users who decline analytics cookies. When consent mode is active, GA4 uses machine learning to model the behavior of non-consenting users based on patterns observed from consenting users, providing more complete reporting without compromising privacy.
BigQuery
Google's serverless, highly scalable data warehouse. GA4 offers a free BigQuery export that streams raw event-level data, enabling advanced analysis, custom attribution models, and data joins that aren't possible within the GA4 interface. Essential for any serious analytics operation.
Bounce Rate
In GA4, the percentage of sessions that were not engaged sessions. This is the inverse of engagement rate. A session bounces if it lasts less than 10 seconds, has no conversion events, and has fewer than 2 pageviews or screenviews. This definition differs significantly from Universal Analytics.
C
CAPI (Conversions API)
A server-side integration that allows advertisers to send conversion events directly from their server to ad platforms (Meta, TikTok, LinkedIn, etc.) without relying on browser-based pixels. CAPI improves data accuracy by bypassing ad blockers, ITP restrictions, and cookie limitations.
Churn
The rate at which customers stop doing business with a company over a given period. In analytics, churn analysis helps identify at-risk users by examining behavioral patterns like decreasing session frequency, reduced engagement, or abandoned features before the user formally cancels.
Cohort
A group of users who share a common characteristic within a defined time period. In GA4, cohort analysis groups users by acquisition date and tracks their behavior over subsequent days, weeks, or months. Useful for measuring retention, lifetime value trends, and the long-term impact of campaigns.
Consent Mode
A Google feature that adjusts how Google tags behave based on the consent status of users. When a user denies consent, tags send cookieless pings instead of full measurement data. Google then uses behavioral modeling to estimate the missing conversions and traffic, providing directionally accurate reporting while respecting privacy.
Container
In Google Tag Manager, a container holds all your tags, triggers, and variables for a specific platform (web, iOS, Android, or server). Each container has a unique ID (e.g., GTM-XXXXX) and can have multiple workspaces for parallel development. Server containers run on your own infrastructure.
Conversion
An event in GA4 that you've marked as representing a valuable user action, such as a purchase, form submission, or sign-up. GA4 counts conversions once per event (not once per session like Universal Analytics), so a single session can generate multiple conversions of the same type.
CPA (Cost Per Acquisition)
The average cost of acquiring one conversion. Calculated by dividing total ad spend by the number of conversions. CPA is a key efficiency metric for performance marketing. Target CPA bidding strategies in Google Ads automatically adjust bids to achieve your desired cost per conversion.
CPL (Cost Per Lead)
The average cost of generating one lead through marketing activities. Similar to CPA but specifically focused on lead generation conversions like form submissions, phone calls, or demo requests. Critical for B2B companies where the sales cycle involves multiple touchpoints after initial lead capture.
Cross-Domain Tracking
The technique of maintaining a single user session across multiple domains or subdomains. GA4 uses a linker parameter (_gl) appended to URLs to pass the client ID between domains. Without proper cross-domain tracking, users appear as new visitors when they navigate between your domains, breaking attribution.
Custom Dimension
A user-defined attribute in GA4 that extends the default data model. Custom dimensions can be scoped to events (describing an action) or users (describing a person). Examples include membership tier, content category, or A/B test variant. GA4 allows up to 50 event-scoped and 25 user-scoped custom dimensions.
D
Data Layer
A JavaScript object (window.dataLayer) that acts as a structured intermediary between your website and Google Tag Manager. It stores information about the page, user, and events in a standardized format that GTM can read and use in tags, triggers, and variables. A well-structured data layer is the foundation of reliable tracking.
Data Stream
The connection point between your app or website and a GA4 property. Each data stream has a unique measurement ID (e.g., G-XXXXXXX) and can be configured with enhanced measurement settings. A single GA4 property can have multiple data streams for web, iOS, and Android.
Dimension
A descriptive attribute of your data in GA4. Dimensions describe what happened: which page was viewed, which campaign drove the traffic, which country the user is from. Dimensions are typically text values and are used to segment and filter your data in reports. Examples: page_title, source, medium, country.
E
Engaged Session
A session in GA4 that lasted longer than 10 seconds, had at least one conversion event, or had two or more pageviews/screenviews. The engagement rate (engaged sessions divided by total sessions) replaced bounce rate as the primary engagement metric in GA4.
Enhanced Conversions
A Google Ads feature that supplements your existing conversion tags by sending hashed first-party conversion data (email, phone, address) from your website to Google. This data is matched against signed-in Google accounts to improve conversion measurement accuracy, especially in environments where cookies are limited.
Event
The fundamental unit of data collection in GA4. Everything is an event: pageviews, clicks, purchases, form submissions. Each event has a name and can carry up to 25 parameters with additional context. GA4's event-based model replaced Universal Analytics' session-based hit model.
F
First-Party Data
Data collected directly by a company from its own customers and users through owned channels (website, app, CRM, email, point of sale). First-party data is increasingly valuable as third-party cookies are deprecated. It includes purchase history, on-site behavior, email engagement, and declared preferences.
Funnel
A visualization of the sequential steps users take toward a conversion goal. GA4's funnel exploration report shows how many users enter each step and where they drop off. Funnels can be open (users can enter at any step) or closed (users must start at step one). Critical for identifying conversion bottlenecks.
G
GA4 (Google Analytics 4)
The current generation of Google Analytics, built on an event-based data model rather than the session-based model of Universal Analytics. GA4 features cross-platform tracking, machine learning insights, BigQuery integration, and privacy-centric design with consent mode and data deletion capabilities.
GTM (Google Tag Manager)
A tag management system that allows you to deploy and manage marketing and analytics tags on your website or app without modifying code directly. GTM uses a container-based architecture with tags (what to do), triggers (when to do it), and variables (dynamic values). Available in web and server-side variants.
I
ITP (Intelligent Tracking Prevention)
Apple Safari's privacy feature that limits the lifespan and capabilities of cookies used for cross-site tracking. ITP caps first-party cookies set via JavaScript to 7 days (or 24 hours if the referring domain is classified as a tracker), significantly impacting GA4's ability to recognize returning users on Safari.
L
LTV (Lifetime Value)
The total revenue a business can expect from a single customer account throughout their entire relationship. In GA4, the predictive LTV metric estimates future revenue from users over the next 28 days. Accurate LTV calculation requires proper revenue tracking and sufficient historical data.
M
Measurement ID
A unique identifier (format: G-XXXXXXXXXX) assigned to each web data stream in a GA4 property. The measurement ID is used in the gtag.js configuration or GTM's GA4 Configuration tag to specify which GA4 property should receive the data. Not to be confused with the GTM container ID (GTM-XXXXX).
Metric
A quantitative measurement in GA4. Metrics answer how much or how many: total users, session count, revenue, conversion rate, engagement time. Metrics are always numerical values that can be summed, averaged, or otherwise aggregated. They pair with dimensions to create meaningful reports.
P
Pageview
An automatically collected event in GA4 (page_view) that fires when a page loads or the browser history state changes. Pageviews capture the page URL, title, and referrer. In single-page applications, GA4's enhanced measurement can track history-based pageviews automatically, though custom implementation often provides better accuracy.
Predictive Audience
An audience in GA4 built using machine learning predictions about future user behavior. GA4 can predict purchase probability, churn probability, and predicted revenue for users over the next 28 days. These audiences can be exported to Google Ads for targeted remarketing campaigns.
Privacy Sandbox
Google's initiative to develop privacy-preserving alternatives to third-party cookies for the open web. Key APIs include Topics (interest-based targeting), Protected Audience (remarketing without cross-site tracking), and Attribution Reporting (conversion measurement without individual tracking). These APIs aim to balance user privacy with advertiser needs.
R
Remarketing
The practice of showing targeted ads to users who have previously interacted with your website or app. In GA4, you create remarketing audiences based on user behavior (e.g., visited product page but didn't purchase) and export them to Google Ads. Effective remarketing requires accurate user tracking and proper audience segmentation.
ROAS (Return on Ad Spend)
The revenue generated for every unit of currency spent on advertising. Calculated as conversion value divided by ad cost. A ROAS of 5:1 means you earned €5 for every €1 spent. ROAS is the primary efficiency metric for e-commerce advertising and is heavily dependent on accurate conversion tracking.
S
Session
A group of user interactions with your website that take place within a given time frame. In GA4, a session starts with the session_start event and times out after 30 minutes of inactivity (configurable). Unlike Universal Analytics, a new campaign click does not start a new session in GA4.
sGTM (Server-Side Google Tag Manager)
A server-side variant of Google Tag Manager that runs in a cloud environment (typically Google Cloud Run or App Engine) on your own first-party domain. sGTM receives data from the client, processes it server-side, and forwards it to analytics and advertising platforms. It improves data accuracy, extends cookie lifetimes, and reduces client-side JavaScript.
SKAN (SKAdNetwork)
Apple's privacy-preserving framework for measuring iOS app install campaign effectiveness without revealing user-level data. SKAN provides aggregated, delayed conversion data with limited granularity. Advertisers receive a conversion value (0-63) assigned by the app, but cannot see which specific user converted.
Server-Side Tagging
The practice of moving tag execution from the user's browser to a server you control. Instead of loading multiple third-party scripts on the client, a single first-party request is sent to your tagging server, which then distributes data to various endpoints. Benefits include improved page performance, better data quality, extended cookie lifetimes, and enhanced privacy control.
T
Tag
In Google Tag Manager, a tag is a snippet of code that executes when a trigger condition is met. Tags send data to third-party tools like Google Analytics, Google Ads, or Facebook. Common tag types include GA4 event tags, Google Ads conversion tags, and custom HTML tags. Each tag requires at least one trigger.
Trigger
In Google Tag Manager, a trigger defines when a tag should fire. Triggers listen for specific events such as page loads, clicks, form submissions, scroll depth, or custom data layer events. Each trigger has conditions that must be met (e.g., fire only on pages matching a URL pattern). A tag can have multiple triggers.
U
User
In GA4, a user is identified by a unique identifier stored in a first-party cookie (client ID) or a user-provided identifier (user ID). GA4 reports on total users, new users, and active users. User identification across devices requires either User-ID implementation or Google Signals, both of which depend on user consent.
UTM Parameters
URL parameters (utm_source, utm_medium, utm_campaign, utm_term, utm_content) used to track the effectiveness of marketing campaigns. When a user clicks a URL with UTM parameters, GA4 captures them as session-level dimensions for attribution. Consistent UTM naming conventions are critical for clean reporting.
V
Variable
In Google Tag Manager, a variable is a named placeholder for a value that is populated at runtime. Built-in variables include Page URL, Click URL, and Form ID. User-defined variables can extract values from the data layer, cookies, DOM elements, or JavaScript. Variables are used in both tags (as dynamic values) and triggers (as conditions).
W
Workspace
In Google Tag Manager, a workspace is an isolated environment for making changes to a container without affecting the live version. Multiple team members can work in separate workspaces simultaneously. Changes are published from a workspace to the container, creating a new version. The free GTM tier allows up to 3 simultaneous workspaces.
Need help making sense of your analytics?
Free 30-minute discovery call. I'll look at your setup and tell you what needs fixing.
Book a Free Audit