Analytics Terms
How marketing gets measured. Attribution, reporting, the metrics that matter and the ones that mislead, and reading what the numbers actually say.
129 terms in Analytics
An actionable metric is a number that changes in response to specific decisions your team makes, and tells you clearly what to do next when it moves. It's the opposite of a vanity...
Active users is the count of unique users who completed at least one engaged session on your site or app within a defined window. Google Analytics 4 (GA4) reports three windows: 1-...
AI attribution is the emerging practice of using artificial intelligence, including large language models and machine learning, to work out which marketing efforts deserve credit f...
An annotation is a note added to a specific date or date range in an analytics platform to explain why data changed. When traffic spikes, a campaign launches, tracking breaks, or t...
An assisted conversion is recorded when a marketing channel contributed to a conversion path but was not the final touchpoint before the customer converted. It shows which channels...
Attribution is how you decide which of your marketing efforts gets credit when a customer buys something. Most customers see several ads, get an email, search a couple of times, th...
Attribution models are the rules that decide how much credit each marketing touchpoint receives when a customer converts. Different models assign credit differently: one might give...
An attribution window is the period of time after a user clicks or views an ad during which a conversion is credited to that ad. A 7-day click window means any purchase within 7 da...
An audience trigger is a rule that fires when a user meets a defined condition, such as visiting a specific page, completing a conversion event, or reaching a certain engagement th...
Average engagement time is the mean amount of time a session spent with your site in the foreground of the user's browser. It is recorded in Google Analytics 4 (GA4) and replaces t...
Bayesian and frequentist testing are two statistical frameworks for interpreting an A/B testing result. Frequentist testing asks whether the observed difference could plausibly be...
A benchmark is a reference point used to evaluate performance. In marketing, benchmarks are typically industry averages for metrics like email open rates, conversion rates, CPCs or...
BigQuery Export is a native Google Analytics 4 (GA4) feature that continuously copies your raw event data into Google BigQuery, Google's cloud data warehouse. It gives analysts dir...
Bounce rate is the percentage of sessions where a visitor lands on a page and leaves without visiting any other page on the site. Calculated as single-page sessions divided by tota...
Cardinality refers to the number of unique values a dimension can hold in an analytics report. When that count exceeds the platform's row limit, Google Analytics 4 (GA4) groups the...
Churn probability is a score assigned to each customer expressing how likely they are to stop purchasing or cancel within a defined future window. It is produced by a predictive mo...
Click-through rate (CTR) is the percentage of people who see something and click on it. Calculated as clicks divided by impressions, multiplied by 100. It applies to paid ads, orga...
Cohort analysis groups users by a shared characteristic (usually acquisition date) and tracks their behaviour over time, revealing how retention, engagement or revenue evolves acro...
Cohort table reading is the skill of interpreting a cohort table, a grid that tracks a group of customers acquired in the same period against how many are still active over time. I...
A confidence interval is a range of values, built from sample data, that's likely to contain the true result if you repeated the test many times. A 95 percent confidence interval m...
A confounding variable is a factor outside your test that influences both the thing you changed and the result you're measuring, making it look like your change caused an effect it...
Control and treatment are the two groups compared in an experiment, most commonly A/B testing. The control group experiences the current, unchanged version of something. The treatm...
A conversion event is a specific action you define as meaningful and instruct your tracking tools to record. Examples include a purchase, a form submission, a phone call, a button...
A conversion path is the sequence of touchpoints a customer moves through before completing a defined conversion, such as a purchase or a form submission. It typically includes eve...
Conversion tracking records when a user completes a desired action — a purchase, form submission, phone call or sign-up — and ties that action back to the marketing channel or camp...
Correlation is when two things move together in a pattern. Causation is when one thing actually makes the other happen. Marketing data is full of strong correlations that get repor...
Cost Per Acquisition (CPA) is the average amount you spend on advertising to generate one defined acquisition: a sale, a lead, a signup, or any other conversion you choose to count...
A counter metric, also called a guardrail metric, is a second number you track alongside your main goal to catch damage the main goal doesn't show. It exists to stop you optimising...
Cross-device attribution is the process of connecting a single customer's behaviour across multiple devices, such as a phone, laptop and tablet, so their full journey from first to...
Cross-domain tracking is the configuration that tells your analytics platform to treat a visitor as the same person when they move between two separate domains, for example from yo...
CTR stands for click-through rate. It measures the share of people who saw something and clicked it. Calculated as clicks divided by impressions, expressed as a percentage. CTR app...
A custom dimension is a label you define and attach to analytics data to add context that Google Analytics 4 (GA4) does not collect by default. Examples include a user's membership...
A tracking event you define and implement yourself, beyond GA4's automatically collected events. Custom events let you track actions specific to your business — video watch milesto...
A custom metric is a measurement you define yourself inside an analytics platform, beyond the standard ones the platform provides by default. In Google Analytics 4 (GA4), custom me...
Customer Acquisition Cost (CAC) is the total amount a business spends to acquire one new customer, covering every dollar that touched the acquisition: ad spend, sales salaries, too...
Dark social refers to content sharing that happens through private channels — WhatsApp messages, Slack channels, direct emails, SMS — where no referrer data is passed to analytics....
A visual interface consolidating key marketing metrics into a single view, updated on a regular cadence. Dashboards replace ad hoc reporting with consistent, comparable data that s...
Data retention settings control how long a platform stores individual user and event-level data before deleting it automatically. In Google Analytics 4 (GA4), the default is two mo...
Data storytelling is the practice of presenting analytics in a way that leads someone to a decision, rather than just showing them numbers. It combines a chart, a clear insight and...
A data stream is a source of behavioural data flowing into a Google Analytics 4 (GA4) property. Each property can have multiple streams, one per platform: web, iOS app, Android app...
Data thresholding is a privacy mechanism in Google Analytics 4 (GA4) that removes rows from reports when the underlying user counts are too small to be anonymised reliably. When th...
The practice of representing data graphically — charts, graphs, maps, heatmaps — to make patterns, trends and outliers visible that would be invisible or slow to interpret in tabul...
Data-driven attribution is an attribution model that uses machine learning to assign conversion credit across all touchpoints in a customer's journey. Instead of applying a fixed r...
DAU (daily active users) and MAU (monthly active users) measure how many unique users engage with a product on a given day or month. The ratio of DAU divided by MAU, often called t...
Website traffic that arrives without a referrer URL that GA4 can identify. Direct traffic can be from users typing your URL directly, but a significant portion is 'dark traffic' —...
The buying and selling of goods and services online. In a marketing context, ecommerce typically refers to the channel performance of an online store — tracking purchase funnel met...
An engaged session is a Google Analytics 4 (GA4) metric that counts any session where a visitor stayed for more than 10 seconds, triggered a conversion event, or viewed at least tw...
Any interaction an audience has with your content — likes, comments, shares, saves, clicks, watch time, replies. Engagement indicates that content resonated and prompted a response...
Enhanced Measurement is a Google Analytics 4 (GA4) feature that automatically tracks common user interactions without requiring custom code. It captures scroll depth, outbound link...
In GA4, an event is any interaction a user has with your website or app that you want to track. Page views, button clicks, form submissions, video plays and purchases are all event...
An event parameter is a piece of extra information attached to a tracking event. When Google Analytics 4 (GA4) records a button click or page view, event parameters describe the co...
Event-driven analytics is a way of measuring behaviour where the basic unit is an event, a specific action a user takes, rather than a pageview. Every click, scroll, form submit, v...
The percentage of pageviews on a specific page that resulted in a visitor leaving the site — regardless of how many pages they viewed before it.
Explorations is the custom analysis workspace inside Google Analytics 4 (GA4). It lets you build reports that go beyond the standard dashboards, including funnel analysis, path ana...
First-click attribution is an attribution model that assigns 100% of the credit for a conversion to the first marketing touchpoint a customer interacted with. If a customer clicked...
First-party data is information you collect directly from people who interact with your business: customers, subscribers, site visitors and app users. It includes email addresses,...
Funnel drop-off is the percentage of visitors who leave your funnel at a specific stage without proceeding to the next one. It is calculated for each transition in the funnel and s...
Funnel Exploration is a report type inside Google Analytics 4 (GA4) that shows how users move through a defined sequence of steps, and where they drop off. Each step is an event or...
A specific action you want users to take on your website, configured in analytics to measure how often that action occurs and which traffic sources drive it.
Google Analytics is a free web and app analytics platform that tracks how people find and use your website. It reports on traffic sources, user behaviour, conversions and audience...
A heat map is a visual overlay on a webpage that shows where users click, how far they scroll, and where their cursor moves. Colour intensity indicates frequency: hot colours like...
A visual representation of where users click, scroll or move their mouse on a webpage, using colour intensity to show areas of high and low engagement.
A holdout group is a portion of your audience deliberately excluded from a marketing campaign so you can measure what happens without it. Comparing the holdout's behaviour to the g...
The process of connecting multiple data points — browsing sessions, devices, email addresses, purchase records — to build a unified profile of an individual customer across touchpo...
Impressions count how many times a piece of content or an ad was displayed on a screen. One person seeing the same ad five times is five impressions. Impressions measure exposure,...
An index is a number rescaled against a baseline so you can compare change over time or across groups, while an absolute value is the raw figure itself, like sales, clicks or dolla...
Internal traffic filtering is the process of excluding visits from your own team, office or agency from your analytics data. Without it, your employees browsing the site, testing p...
A key event is what Google Analytics 4 (GA4) calls a conversion event from March 2024 onward. The underlying mechanics are the same: you mark an existing GA4 event as important, an...
KPI stands for key performance indicator. A KPI is a specific, measurable value that tells you whether a business, team or campaign is making progress toward a defined goal. The wo...
Last-click attribution is a method of assigning conversion credit that gives 100% of the credit to the final touchpoint a customer interacted with before converting. If someone cli...
A leading indicator is a metric that moves before a result and can predict it, like traffic or pipeline volume. A lagging indicator is a metric that confirms a result after it's ha...
Lift and attribution answer different questions about marketing effectiveness. Attribution decides which channels get credit for conversions that happened. Lift (also called increm...
Linear attribution is a model that divides conversion credit equally across every marketing touchpoint in the customer journey. If a customer saw four ads before buying, each ad re...
Google's free data visualisation and reporting tool. Connect multiple data sources, build live dashboards and share reports with clients or stakeholders without exporting spreadshe...
Marketing automation is software that executes marketing tasks automatically based on predefined rules or user behaviour. It handles repetitive work — sending triggered emails, upd...
Marketing reporting is the process of collecting, organising and presenting marketing performance data to inform decisions. Good reporting shows what is working, what is not and wh...
Measurement Protocol is a Google Analytics 4 (GA4) feature that lets you send event data directly to GA4 from a server, without any browser or app code running. You make an HTTP re...
A metric tree is a diagram that breaks one top-level business outcome, like revenue, into the smaller metrics that drive it. Each branch shows how a lever, such as conversion rate...
Minimum Detectable Effect (MDE) is the smallest change in a metric, such as conversion rate, that an experiment is sized to reliably detect. Set before an A/B test runs, it tells y...
Multi-touch attribution (MTA) is a method of assigning conversion credit across multiple marketing touchpoints in a customer journey, rather than giving all credit to one channel....
The ratio of first-time visitors to users who have visited previously. Returning visitors typically convert at higher rates, have longer sessions and cost less to acquire than new...
A North Star Metric (NSM) is the single measure a business uses to represent the core value it delivers to customers, chosen because moving it reliably predicts long-term revenue a...
Novelty effect is the temporary boost in engagement or conversion a change gets simply because it's new and different, not because it's actually better. It shows up commonly in A/B...
A p-value is a number between 0 and 1 that tells you how likely you would see your test results, or something more extreme, if there was no real difference between the two things y...
The total number of times a page has been loaded or reloaded in a browser. One of the oldest web analytics metrics. Often misinterpreted because it counts every load including refr...
Path exploration is a report type in Google Analytics 4 (GA4) that shows the sequence of pages or events users move through on your site or app. It maps what happens before and aft...
Path length is the number of distinct marketing touchpoints a customer encounters before converting. A customer who clicks a paid search ad, then opens an email, then converts on a...
The peeking problem is what happens when you check an A/B test's results early and stop it as soon as one variant looks like it's winning. Doing this repeatedly inflates the chance...
Position-based attribution, also called U-shaped attribution, assigns the most credit to the first and last touchpoints in a customer journey, with the remainder shared equally acr...
Predictive analytics uses statistical models and machine learning to forecast future outcomes based on historical data. In marketing, it is used to predict which leads will convert...
Predictive audiences are segments built by Google Analytics 4 (GA4) using machine learning to identify users likely to take a specific action, such as making a purchase or churning...
Purchase probability is a machine-learning score that predicts how likely a specific user is to complete a purchase within the next seven days. Google Analytics 4 (GA4) generates t...
Reach is the number of unique individuals who saw your content or ad at least once, within a given time period. Unlike impressions, which count every view including multiple views...
Real-time data is information that is collected, processed and made available for use immediately as events happen, with minimal latency. In marketing, it enables live campaign mon...
A referral exclusion tells your analytics platform to ignore a specific domain when it appears as a traffic source. Instead of starting a new session and crediting that domain with...
Referral traffic is the visitors who arrive at your website by clicking a link on another website that is not a search engine. If someone clicks a link to your site in a news artic...
Regression to the mean is the statistical tendency for unusually high or unusually low results to move back toward average on the next measurement, even with no change in strategy....
Rules-based attribution assigns credit to marketing touchpoints using fixed, human-defined logic, such as last-click or first-click. Algorithmic attribution uses statistical models...
Sample size is the number of observations, visitors or responses needed in a test or survey before the results can be trusted. In marketing, it determines how long an A/B test must...
Sampling in analytics is when a platform calculates your report using a subset of your data rather than all of it. The result is an estimate, not an exact count. Sampling happens w...
Segment overlap is the portion of users or customers who qualify for two or more audience segments at the same time. When a person appears in multiple segments, they can be counted...
Segmentation in reporting is the practice of breaking a top-line metric down into smaller groups, such as channel, device, region or customer type, rather than reporting a single b...
Self-serve analytics is a setup where marketers, product managers or founders can pull their own reports and answer their own questions in a dashboard, without asking an analyst to...
Sentiment analysis is the use of natural language processing to classify text about your brand, product or category as positive, negative or neutral. It is used to monitor brand he...
Sequential testing is a method for analysing A/B tests that lets you check results continuously as data comes in, and stop the test early once a statistically valid conclusion is r...
Server-side tracking sends user and conversion data to analytics and advertising platforms from your own server rather than from the user's browser. Because the data never touches...
A session is a single continuous visit to a website by a user. It begins when someone first arrives and ends when they leave or when the session timeout threshold is reached. Sessi...
Session duration measures how long a visitor spends on your website during a single visit. In GA4, this is reported as average engagement time, which measures time the browser is a...
Sessions per user is the average number of times a user visits your site or app within a defined period. It is calculated by dividing total sessions by total users. In Google Analy...
Simpson's Paradox is a statistical pattern where a trend appears in several separate groups of data but disappears or reverses when the groups are combined. It happens because the...
A single source of truth (SSOT) is the one system or dataset an organisation agrees is correct when numbers disagree elsewhere. It's not a report or a dashboard. It's the underlyin...
Split testing is a method of comparing two or more versions of a webpage, email, ad or other marketing element by showing each version to a different group of real users. The versi...
Standard deviation is a measure of how spread out a set of numbers is around its average. A low standard deviation means most results sit close to the mean. A high one means result...
Statistical power is the probability that a test correctly detects a real effect when one genuinely exists. Low power means a test can run cleanly and still miss a true winner, rep...
A result is statistically significant when the probability it happened by chance is low enough to trust — typically below 5%. Without it, your A/B test winner might be noise.
Stickiness is the ratio of Daily Active Users (DAU) to Monthly Active Users (MAU), expressed as a percentage. It measures how often your monthly users return on a given day. A high...
Survivorship bias is the mistake of drawing conclusions only from the people, campaigns or businesses that succeeded, while ignoring the ones that tried the same thing and failed....
Third-party data is audience information collected by external companies and sold or licensed to advertisers. It includes demographics, interests and browsing behaviour gathered fr...
The Time Lag report shows how many days pass between a customer's first marketing touchpoint and their eventual conversion. It tells you whether your customers tend to convert quic...
Time-decay attribution is a multi-touch attribution model that gives more credit to the marketing touchpoints closest in time to a conversion. The further back a touchpoint sits in...
A traffic source is the origin of a visitor to your website — where they came from before arriving. Google Analytics 4 groups these into default channels: Organic Search, Paid Sear...
Triangulation in marketing measurement means cross-checking three independent sources of evidence: Marketing Mix Modelling (MMM), multi-touch attribution, and incrementality experi...
Unified measurement is an approach to marketing analytics that combines multiple measurement methods, typically multi-touch attribution (MTA), marketing mix modelling (MMM) and inc...
In analytics, a user is an individual visitor to your website or app. In GA4, the default 'Users' metric specifically means Active Users — people who had at least one engaged sessi...
UTM parameters are small pieces of text added to the end of a URL that tell your analytics platform where a visitor came from. Urchin Tracking Module (UTM) is the original name, fr...
A vanity metric is a number that looks impressive in a report but doesn't tell you whether the business is actually improving. Followers, impressions and total pageviews are classi...
Variance is a statistical measure of how spread out a set of results is around the average. High variance means results bounce around a lot from day to day or test to test. Low var...
W-shaped attribution is a multi-touch attribution model that assigns most credit to three specific touchpoints in the customer journey: the first touch, the touch that creates a le...
Walled garden attribution is when each major ad platform, such as Meta, Google and TikTok, measures and reports conversions using its own data, its own attribution model, and its o...
Other categories
Want the full picture?
Hub gives you persistent connections, monthly score tracking and a marketing plan that updates itself.
Get started with Hub