Google Analytics is a free web analytics platform that tracks visitor behaviour on websites and apps, providing data on traffic sources, user engagement, conversions, and audience demographics. It powers most data-driven decisions for site owners, marketers, and SEO teams.
Google Analytics works by embedding a small JavaScript snippet—the GA tracking code—into the HTML of every page you want to monitor. When someone loads a page, the script fires and sends an event payload to Google's servers. That payload includes the page URL, timestamp, referrer, user's device type, browser, screen resolution, geographic location derived from IP address, and a randomly generated client ID stored in a cookie to recognize returning visitors.
In GA4, everything is treated as an event. A pageview is an event, a button click is an event, a video play is an event, a form submission is an event. You define what matters, configure enhanced measurement toggles for common interactions, or push custom events via the data layer. The old Universal Analytics model centred on sessions and pageviews; GA4 shifted to user-centric, event-stream architecture better suited to apps and privacy regulations. Data flows into your property in near real-time, typically showing up in reports within seconds to a few minutes.
Google Analytics answers the questions that guide site strategy. Where is traffic coming from—organic search, paid ads, social media, email campaigns, direct visits? Which pages attract the most visitors, and which ones cause people to leave immediately? How long do users stay, how many pages do they view, and what paths do they take before converting or bouncing?
For SEO teams, GA integrates with Search Console to surface the exact queries driving organic traffic, landing pages with the highest impressions, and click-through rates. For paid marketers, it ties Google Ads campaigns directly to on-site behaviour and revenue attribution. Content teams use engagement metrics to identify top performers and content gaps. E-commerce operators track transaction data, product performance, and checkout abandonment. The platform's ubiquity and zero cost make it the foundation layer of almost every digital analytics stack, even when paired with heat-mapping tools or specialized attribution software.
A Google Analytics account can hold multiple properties—one per website or app. In GA4, each property has data streams: web, iOS app, Android app. You create a stream, grab the measurement ID, paste the GA tag into your site's header, and data starts flowing. Many sites use Google Tag Manager to deploy the snippet without touching code directly, which simplifies adding event triggers and managing tags across platforms.
Critical setup steps include enabling enhanced measurement for scroll depth, outbound clicks, site search, video engagement, and file downloads. You must define conversions—newsletter signups, purchases, quote requests—so GA knows what constitutes success. In Universal Analytics, practitioners created separate views with filters to exclude internal traffic or focus on subdomains; GA4 replaced views with data filters and comparisons, requiring you to filter IP ranges early or accept polluted data. Skipping this step means your reports include your own team's browsing, vendor QA testers, and bot traffic, skewing every decision.
Google Analytics offers dozens of pre-built reports: real-time visitors, acquisition breakdowns, engagement summaries, demographic overlays, monetization for e-commerce. The temptation is to celebrate rising session counts or pageview spikes. Those numbers mean little if bounce rate climbs, session duration drops, and conversions stay flat. Focus instead on metrics tied to actual outcomes.
Acquisition reports tell you which channels deliver quality traffic—organic search users might convert at higher rates than social referrals, even if social sends more volume. Engagement metrics reveal whether content holds attention; a blog post with high entrances but immediate exits signals a mismatch between title promise and content delivery. Conversion paths show multi-touch journeys, exposing whether users discover you via organic search, return through direct visits, then convert after a paid retargeting ad. Use audience segments to compare new versus returning visitors, desktop versus mobile behaviour, or geography-based differences. The platform's exploration tool lets you build custom funnels, cohort analyses, and user pathing to answer questions the standard reports cannot.
GA4 emerged partly in response to GDPR, CCPA, and mounting privacy concerns. The platform now anonymizes IP addresses by default, offers granular user-deletion controls, and supports consent-mode integration so tags fire only after visitors accept cookies. Canadian sites must navigate PIPEDA compliance, and Quebec's Law 25 adds stricter consent requirements; failing to obtain clear opt-in before tracking can expose you to regulatory risk.
Data retention settings determine how long user-level and event-level data persist. GA4 defaults to two months for user data, though you can extend to fourteen months. After that window, aggregated reports remain, but exploration tools lose granular event details. Many practitioners discover this too late, losing the ability to analyze historical user paths or build retroactive cohorts. Universal Analytics stopped processing data in July 2023, and Google will delete that historical data entirely by mid-2024 unless exported. If you haven't migrated and archived old UA data, it's gone—another reason setup decisions and migration timing matter more than most site owners realize.
The most frequent mistake is deploying the tracking code incorrectly—double-tagging a page because the snippet exists in both the theme template and a plugin, inflating pageview counts. Another is failing to exclude internal traffic; without IP filters, your own team's daily site checks appear as engaged sessions, distorting engagement and conversion metrics. Not setting up cross-domain tracking when you have subdomains or separate checkout domains breaks user journeys into artificial sessions, ruining attribution.
Ignoring bot filtering—enabled by default but sometimes toggled off—lets referral spam and crawler hits pollute reports. Skipping UTM parameters on campaign links means you can't distinguish email blast traffic from organic or attribute conversions to specific promotions. Not configuring e-commerce tracking when you run a store leaves revenue invisible, forcing guesswork on product performance. And many sites never define events or conversions, treating GA as a passive visitor counter rather than a decision engine. Each oversight costs clarity, and stacked together they render the platform nearly useless despite its power.
Google Analytics 4 is the current version, launched in 2020 and mandatory since July 2023 when Universal Analytics stopped processing data. GA4 uses an event-based model instead of session-based tracking, offers cross-device and app-plus-web measurement in a single property, anonymizes IP addresses by default, and integrates machine-learning predictions. Universal Analytics relied on pageviews and sessions as core metrics; GA4 treats everything as events, making it more flexible but requiring a learning curve for those accustomed to the old interface and reporting structure.
Google Analytics is free for most users, with no charge for standard web and app tracking. The free tier supports up to ten million hits per month per property, which covers the vast majority of small to mid-sized sites. High-traffic enterprises that exceed limits or need advanced features like unsampled reporting, service-level agreements, and dedicated support can pay for Analytics 360, the premium version. For typical use cases—tracking a business site, blog, or e-commerce store—the free version provides all core functionality without cost.
Create a Google Analytics account, set up a GA4 property, add a data stream for your website, and copy the measurement ID. Paste the GA tracking snippet into the header section of every page, or use Google Tag Manager to deploy it without editing code. Most CMS platforms—WordPress, Shopify, Wix—offer plugins or built-in fields to insert the tracking ID. After installation, verify data appears in the real-time report within a few minutes, then configure conversions, enable enhanced measurement, and filter internal traffic to ensure accurate reporting.
No. Google's terms of service explicitly prohibit sending personally identifiable information—names, email addresses, phone numbers—into Analytics. The platform assigns anonymous client IDs stored in cookies to recognize returning visitors across sessions, but those IDs are random strings with no personal data. If you send PII through custom dimensions or event parameters, you violate the terms and risk account suspension. For user-level insights, integrate a separate CRM or marketing automation tool that complies with privacy laws, and use hashed or anonymized identifiers if cross-referencing data.
Discrepancies arise from different tracking methodologies, attribution models, and data-processing rules. Google Ads counts clicks; Analytics counts sessions, and not every click results in a session if the page fails to load or the user blocks JavaScript. Social platforms report link clicks; Analytics records only those clicks that successfully fire the tracking code. Ad blockers, browser privacy settings, and users who decline cookie consent prevent GA from collecting data, undercounting traffic. Sampling in large datasets, time-zone mismatches, and bot-filtering differences also cause variance. Expect rough alignment, not exact parity.
GA4 cannot import historical Universal Analytics data, and Google will delete UA properties entirely by mid-2024. If you rely on year-over-year comparisons, trend analysis, or need to reference pre-2023 performance, export your UA data now—use the API, BigQuery integration, or third-party backup tools. Once deleted, that history is irretrievable. GA4 offers superior privacy controls and cross-platform tracking going forward, but losing years of baseline data hampers strategic planning, so archiving UA data is critical before the cutoff.