Google Analytics 4 Review 2026

What is Google Analytics 4?

Google Analytics 4 (GA4) is the free web and app analytics platform from Google, and since July 2024 it is the only version of Google Analytics you can still log in to. If you are weighing google analytics 4 vs universal analytics for old reports, the short answer is that the older product is gone and its data has been deleted, so the real question now is how to use GA4 well, and when a paid tier or a different tool makes more sense.

What GA4 is and who it suits

GA4 measures websites and mobile apps in one property. Each source of traffic is a data stream: a web stream for your site, plus separate iOS and Android streams if you have apps. Every interaction, from a page view to a purchase, is recorded as an event with parameters attached.

It suits Indian D2C brands on Shopify or WordPress, SaaS startups that run Google Ads, content publishers and agencies that manage many client sites. It is a weaker fit for teams that want simple numbers with no setup, or for privacy-first products that would rather avoid cookies and Google data processing altogether.

Google Analytics 4 vs Universal Analytics: what actually changed

According to Google's help centre, standard Universal Analytics (UA) properties stopped processing hits on 1 July 2023. UA 360 properties with an active 360 order received a one-time extension to 1 July 2024. Access to the UA interface and API then ended around the week of 1 July 2024, and Google states that data not exported before then was deleted and cannot be recovered.

The measurement model is the bigger change. UA had separate hit types (pageview, event, ecommerce, social and so on) and events built on Category, Action and Label. GA4 has no hit types; everything is an event, and the Category/Action/Label structure does not exist. Google's own advice is to redesign your tracking plan rather than copy UA events one to one.

Sessions behave differently too. A new campaign does not start a new session in GA4, and GA4 accepts events that arrive up to 72 hours late, so session counts rarely match old UA numbers. Bounce rate also came back with a new meaning. GA4 counts a session as engaged if it lasts longer than 10 seconds, has a key event, or has two or more page or screen views. Engagement rate is the share of engaged sessions, and bounce rate is simply its inverse.

Views are gone as well. Google suggests comparing an unfiltered UA view with a single GA4 web stream if you ever need to reconcile historical exports.

Key features in practice

• Enhanced measurement: page views, scrolls, outbound clicks, site search and similar actions are collected automatically once the web stream is created.
• Explorations: free-form tables, funnels, path analysis and segment overlap for questions the standard GA4 reports do not answer.
• Audiences: build groups such as "added to cart but did not buy in 7 days" and share them with Google Ads for remarketing.
• Integrations: Google Ads, Search Console and other Google Marketing Platform products link from the Admin screen.
• BigQuery export: raw event-level data sent to your own Google Cloud project for SQL analysis.
• APIs: the Data API is a REST interface (runReport and related methods) with client libraries in Java, Python, Node.js, PHP, C#, Go and Ruby, useful for custom dashboards.

One naming change confuses many teams. In March 2024 Google renamed GA4 "conversions" to key events. The word conversions now refers to Google Ads conversions, which you can create from a key event and which appear in the Advertising section, not in standard reports. Existing conversion history was carried over as key event data, so year-on-year comparisons still work.

Google Analytics 4 events setup: the first week

Google's setup guide walks through an account, a property and a data stream. A small site can get through it in an afternoon:

• Create an account at analytics.google.com, then a property with your reporting time zone (Asia/Kolkata for most Indian businesses) and the reporting currency you sell in.
• Add a web data stream, enter your URL and leave enhanced measurement on.
• Install the Google tag. Paste the G- measurement ID into Shopify, Wix or a WordPress plugin, paste the gtag.js snippet right after the opening head tag, or add the Google Analytics configuration tag in Google Tag Manager.
• Open the Realtime report and visit your own site. Google says collection can take up to 30 minutes to begin.
• Mark the actions that matter (purchase, generate_lead, sign_up) as key events, then link Google Ads and Search Console.

Plan custom GA4 events before you add them. A standard property allows 25 parameters per event, 50 event-scoped custom dimensions and 30 key events, so a messy naming scheme fills the quota quickly. Using Google Tag Manager keeps event logic in one place and makes changes possible without a developer pushing code.

If reports look empty after setup, check four things: the tag is firing (use Tag Assistant or Realtime), you are not filtering out your own traffic by mistake, consent banners are not blocking every tag, and enough time has passed. Google's data freshness page says processing can take 24 to 48 hours, with daily data for standard properties typically taking around 12 hours.

GA4 free limits and GA4 360

Google's Analytics terms say the service is free for GA4 properties, and no monthly event cap is stated for the free tier. The limits sit elsewhere, as the table shows (figures from Google's 360 help article).

Retention needs one clarification. The setting only affects explorations and funnel reports, while standard aggregated reports keep their history. Many properties sit on the 2-month option without anyone noticing, so check the setting on day one and pick 14 months. Large and XL properties are held to 2 months, and age, gender and interest data is always kept for 2 months.

On export, streaming has no event limit for standard properties, while the daily batch file stops if you stay over 1 million events a day. You can export to the BigQuery sandbox at no charge, though its free storage is small (10 GB), so heavier users pay Google Cloud for storage and queries.

Analytics 360 has no public price. Google's 360 page asks buyers to talk to sales, and it highlights service level agreements, faster data freshness (intraday data in about an hour versus 2 to 6 hours on the free tier) and unsampled results. Expect it to make sense only for large enterprises with very high traffic.

Privacy, consent and data rules

GA4 does not log or store IP addresses, according to Google, so the old IP masking setting is no longer needed. For users in the EU, Switzerland and the UK, data is collected through servers in those regions and IP addresses are dropped before logging. You can also switch off Google signals and granular location and device data per region.

Consent mode passes a visitor's cookie choice to Google tags; it does not give you a banner. In the basic setup, tags wait for consent, and nothing is collected from people who decline. In the advanced setup, tags load first and send cookieless pings on refusal, which lets GA4 apply behavioural modelling to fill the gaps. You still need a consent management platform to show the banner.

Is GA4 compliant with GDPR or India's DPDP Act by default? Google does not make that claim for you. Compliance depends on your consent flow, your privacy policy and what you send. The terms bar sending anything Google could recognise as personally identifiable information (a narrow exception covers properly hashed data in features built for it), so never put raw emails or phone numbers in event parameters.

Expect gaps in small reports. Data thresholds can hide rows that include demographics or search query data when user counts are low. You cannot switch thresholds off, but a wider date range often helps.

GA4 vs Matomo

Matomo is open source. Its On-Premise Community edition is free to download with unlimited users and hits, and Matomo Cloud is a paid hosted plan with a free trial and data stored in Frankfurt, Germany. Choose Matomo when you must host analytics yourself or keep data out of Google's systems. Choose GA4 when Google Ads, Search Console and audience sharing matter more than full data ownership, and when no one on your team wants to maintain a server.

GA4 vs Plausible

Plausible takes the opposite approach to GA4. It uses no cookies, publishes its code and shows one simple dashboard. Its Starter plan starts at $9 a month for up to 10,000 monthly pageviews, with a 30-day free trial and no card required. For a blog or a small SaaS marketing site that only needs traffic sources and top pages, Plausible is easier to read. For ecommerce funnels, app tracking or raw data in BigQuery, GA4 does far more for no licence fee.

Against Adobe Analytics, the gap is mainly price and buyer. Adobe sells Select, Prime and Ultimate packages on custom quotes, which suits enterprises with dedicated analysts.

Limitations to know before you commit

• The interface is harder to learn than UA was, and many useful views live in explorations rather than standard reports.
• Free users mostly rely on the help centre and community forums; SLAs are a 360 feature.
• Numbers will not match ad platforms, CRMs or older UA exports exactly, because of modelling, thresholds and session rules.
• Leaving exploration retention on 2 months catches people out a year later.

Verdict for Indian teams

For most Indian startups, agencies and online stores, GA4 is the sensible default: it costs nothing, it links with Google Ads, and BigQuery export gives you raw data without an enterprise contract. Budget a few days for a proper event plan and a consent setup, and keep one alternative such as Plausible or Matomo in mind for privacy-led projects.