GA4 Setup Checklist 2026 – Google Analytics 4

Google Analytics 4 looks like a five-minute setup. Grab the Measurement ID, paste it into your site, and data starts flowing. But three months later, when you sit down to review the reports, you find nothing useful. Conversions are not defined. Events are incomplete. Audience segments have not been created. E-commerce data is missing. The problem is never GA4 itself. It is the steps that were skipped during setup. A comprehensive GA4 setup checklist is the difference between a system that collects data and a system that actually generates insights you can act on.

Universal Analytics (UA) is now fully retired. Google stopped data collection in UA in mid-2023 and shut off access entirely in 2024. GA4 is built on a completely different architecture: event-based rather than session-based, focused on user behaviour rather than page views. This is not just an interface change. It is a fundamental shift in how data is collected and reported. Trying to carry UA habits into GA4 is a recipe for a broken setup.

Account and Property Creation

Everything in GA4 starts with the account and property structure. Getting this wrong leads to data confusion and access problems down the line. The account structure for a large enterprise needs to differ from that of a single-site SME.

Is the GA4 property under the correct account?

A Google Analytics account can hold multiple properties. For agency clients, keeping each client in a separate account is the cleanest approach. For your own business, a single account can house separate properties for the website and mobile app. Keep account names clear and descriptive: “Company Name – GA4” works well. If you previously used Universal Analytics, you can add a GA4 property under the same account, but creating a fresh account provides a cleaner starting point with no legacy baggage.

Are the property settings correctly configured?

When creating a property, the time zone and currency settings are critical. For a UK-based site, set the time zone to “GMT+0 London” (or “GMT+1” during British Summer Time; Google handles this automatically when you select the London time zone) and the currency to “GBP.” For the US, select the appropriate time zone and “USD.” If the time zone is wrong, daily data in reports will be offset and midnight transitions will be calculated at the wrong time. If the currency is wrong, e-commerce revenue figures will be misleading. These settings can be changed later, but corrections only apply from the date of the change. Historical data stays as it was.

Has the data retention period been adjusted?

GA4’s data retention defaults to two months. This means Exploration reports only have access to the last two months of data. Standard reports are unaffected, but for any deep-dive analysis, two months is far too short. Go to Admin > Data Settings > Data Retention and extend it to 14 months. GA4’s free tier maxes out at 14 months. GA4 360 (the paid version) allows up to 50 months. Change this setting on day one. The data you lose by leaving it on two months cannot be recovered retroactively.

Is Google Signals enabled?

Google Signals lets you track cross-device behaviour for signed-in Google users. If a user visits your site on their phone and then converts on their laptop, Google Signals can link those two sessions. Enable it under Admin > Data Settings > Data Collection. There is a caveat, though: when Google Signals is active, thresholding may be applied, and low-volume data rows can be hidden from reports. For smaller sites, this can cause gaps in your data. Weigh the cross-device tracking benefit against the potential for thresholded reports.

Data Stream Configuration

A data stream is the source that sends data to your GA4 property. Separate data streams are created for websites, iOS apps and Android apps. For most businesses in the UK and US market, a web data stream is all that is needed.

Has a web data stream been created?

After creating the property, add a web data stream. Enter your site URL and give the stream a name. GA4 will provide a Measurement ID in the format “G-XXXXXXXXXX.” This ID is required for adding the tracking code to your site. Note it down somewhere accessible because you will need it during the GTM setup stage.

Have Enhanced Measurement settings been reviewed?

GA4 enables “Enhanced Measurement” by default on web data streams. This feature automatically collects events like page views, scrolls, outbound clicks, site searches, video engagement and file downloads without any additional code. Review each one individually:

• Page views: Should always stay on. This is GA4’s core data source.
• Scrolls: Triggers at 90 per cent scroll depth. Useful for blog sites, but if you need more granular scroll tracking (25 per cent, 50 per cent, 75 per cent), you will need to set up custom events through GTM.
• Outbound clicks: Tracks clicks on links to external sites. Particularly valuable for affiliate sites.
• Site search: Captures search queries from your on-site search box. The default parameter is “q” but if your site uses a different parameter (s, search, query), add it in the settings. WordPress typically uses “s”.
• Video engagement: Tracks embedded YouTube videos. Other video platforms (Vimeo, Wistia) require custom tags in GTM.
• File downloads: Automatically tracks downloads of PDFs, DOCX, XLSX and similar file types.

Have unwanted referrals been listed?

Payment gateways (Stripe, PayPal, Worldpay, SagePay), login pages and third-party redirect services can generate false source/medium data in GA4. When a user goes from your site to a payment page and returns, GA4 may record this as a new session with the source “stripe.com / referral.” This corrupts your conversion tracking data. Go to Admin > Data Streams > More Tagging Settings > List Unwanted Referrals and add these domains. For UK e-commerce sites, common additions include stripe.com, paypal.com, worldpay.com and the domains of any 3D Secure verification providers you use.

GTM Integration

Google Tag Manager (GTM) is essential for a proper GA4 setup. While you can add GA4 directly to your site’s code, every new event, parameter or change will require a developer. GTM allows the marketing team to manage tags without writing code, which dramatically speeds up the iteration cycle.

Is the GTM container code correctly installed on the site?

GTM has two code snippets. The first goes inside the head tags, the second immediately after the opening body tag. Both must be present on every page. A surprisingly common mistake is adding only the head snippet and forgetting the body snippet. In that scenario, GTM works but data is lost in browsers with JavaScript disabled. If you use WordPress, the GTM4WP plugin or Google Site Kit handles automatic code placement.

Has a GA4 Configuration tag been created?

Inside GTM, create a “Google Tag” (previously known as the GA4 Configuration Tag). Enter your Measurement ID (G-XXXXXXXXXX) and set the trigger to “All Pages.” This tag sends baseline page view data to GA4. Test in Preview mode before publishing. In Preview mode, the tag should show as “Fired.” If it shows as “Not Fired,” there is a trigger configuration error that needs fixing.

Has Debug View been used for verification?

GA4’s Debug View feature shows incoming events in real time. With GTM’s Preview mode open, browse your site and check GA4 > Admin > Debug View for a live feed of events. You should see page_view, scroll, click and other events appearing. You can also inspect event parameters in this view to confirm they are being passed correctly. Debug View is the most reliable way to confirm your setup is working before you rely on the data for decisions.

Events

GA4 operates on a fully event-based data model, unlike Universal Analytics. Every interaction is an event. A page view is an event. A button click is an event. A form submission is an event. This model is more flexible, but it requires careful configuration to produce meaningful data.

Have automatically collected events been reviewed?

GA4 automatically collects certain events without any configuration: first_visit, session_start, page_view and user_engagement. You cannot turn these off, and you should not want to. On top of these, Enhanced Measurement adds events like scroll, click and file_download. Check the Realtime report to see which events are coming through. Are the events you expect actually appearing in the list?

Have custom events been defined?

Automatic events cover the basics, but tracking interactions specific to your business requires custom events. Contact form submissions, quote requests, phone button taps, WhatsApp button clicks, visits to specific pages, video completion events. All of these should be set up as custom events.

To create custom events in GTM: build a new GA4 Event tag, set the event name (form_submit, whatsapp_click, etc.), add the necessary parameters and attach the appropriate trigger. Use snake_case for event names (underscores, no spaces, no special characters). GA4 is case-sensitive for event names: “Form_Submit” and “form_submit” are recorded as two separate events. Pick a convention and stick to it across the entire implementation.

Are event parameters correctly defined?

Each event’s parameters carry the details of that interaction. For example, a “form_submit” event could include form_name (which form?) and form_location (where on the page?) as parameters. GA4 allows up to 25 parameters per event. Crucially, you must register parameters under “Custom Definitions” in the GA4 interface, otherwise they will not appear in reports. Unregistered parameter data is lost within 72 hours. Do not skip this step.

Key Events (Conversions)

Conversions in GA4 are now called “Key Events.” Conversion tracking is the most critical component of any Analytics setup. Setting up GA4 without defining conversions is like installing an electricity meter that does not record usage: something is happening, but you have no idea how much.

Have Key Events been marked?

Marking an event as a Key Event in GA4 is straightforward: go to Admin > Events, find the relevant event and toggle “Mark as key event.” But choosing which events to mark is a strategic decision. If you mark every click as a conversion, your conversion data becomes meaningless. Select only the actions that genuinely create value for your business:

• E-commerce: purchase, add_to_cart, begin_checkout
• Lead generation: form_submit, phone_call, whatsapp_click
• Content sites: newsletter_signup, pdf_download

Has the attribution window been configured?

GA4’s default attribution window is 30 days. If a user visits your site today and converts 29 days later, the conversion is attributed to the source of the original visit. If you operate in B2B where decision cycles are long, extending the window to 90 days makes sense. Adjust this under Admin > Attribution Settings. For e-commerce, 7 to 30 days is usually appropriate. Getting this wrong means crediting the wrong channels for your conversions, which distorts your budget allocation decisions.

Has the Google Ads connection been established?

Importing GA4 conversion data into Google Ads directly improves campaign optimisation. Go to Admin > Product Links > Google Ads and link your Ads account. Once connected, GA4 Key Events become available as conversion actions in Google Ads. Using the same conversion definitions in both platforms ensures consistent performance measurement. This is especially important for Google Ads campaigns using automated bidding strategies like Target CPA or Target ROAS, which rely on accurate conversion data to optimise effectively.

Audience Definitions

Audience definitions in GA4 let you segment user groups based on specific characteristics or behaviours. Well-defined audiences are critical for both reporting and remarketing through Google Ads.

Have core audiences been created?

At minimum, define the following audiences in GA4:

• Converted users: Users who have completed any Key Event. Needed for building “similar audiences” in Google Ads and for excluding existing customers from acquisition campaigns.
• Cart abandoners (e-commerce): Users who triggered add_to_cart but did not complete purchase. The most valuable audience for remarketing campaigns.
• High-value users: Users with more than three sessions in the last 30 days, or users who have exceeded a specific spending threshold.
• Blog readers: Users who visit blog pages but have not viewed service pages. Useful for measuring the effectiveness of content marketing efforts.

Are audience triggers active?

When defining an audience in GA4, the “audience trigger” feature lets you automatically fire an event when a user joins that audience. This event can also be marked as a Key Event. Having an automatic event fire when a user enters your “high-value” audience definition is extremely useful for CRM integrations and marketing automation workflows. It bridges the gap between analytics and action.

Have audience segments been exported to Google Ads?

Audiences defined in GA4 can be exported to Google Ads for use in remarketing campaigns. This requires the GA4 and Google Ads accounts to be linked (covered in the previous section). Audiences become available in Google Ads automatically after the link is established. There are minimum size requirements: Display campaigns need at least 100 users, and Search campaigns need at least 1,000 users in the audience list before they can be used for targeting.

Report Customisation

GA4’s default reports are insufficient for most businesses. Customising reports to match your business saves analysis time and leads to better decisions.

Have default reports been reviewed?

In the Reports section, GA4 provides pre-built reports under Lifecycle (Acquisition, Engagement, Monetisation, Retention) and User categories. Open each one and check whether meaningful data is flowing. In the Acquisition report, are traffic sources accurate? If “direct / none” accounts for more than 30 per cent of traffic, there is likely a UTM parameter problem that needs investigating.

Have custom reports and explorations been created?

The Explorations section goes beyond standard reports, enabling deep analysis. Free Form, Funnel Exploration, Path Exploration and Cohort Analysis templates are all available. At a minimum, create one conversion funnel report: visit > product view > add to cart > checkout > purchase. This funnel reveals exactly where you are losing users in the conversion process, which is one of the most valuable pieces of information GA4 can provide.

Have UTM parameters been standardised?

Accurate traffic source reporting in GA4 depends on UTM parameters. If your team uses UTM parameters in email campaigns, social media posts and ad campaigns, everyone must follow the same naming convention. “facebook” vs “Facebook” vs “fb” are treated as three separate sources in GA4. Create a UTM naming guide and share it with the entire team. Google’s Campaign URL Builder tool helps you generate parameters consistently. This is one of those small operational details that has an outsized impact on data quality.

E-commerce Tracking

If you run an e-commerce site, GA4’s e-commerce measurement capabilities are essential. Standard event tracking does not capture product views, add-to-cart actions and purchase details automatically. These need dedicated configuration.

Are e-commerce events correctly configured?

GA4’s e-commerce measurement requires specific event names and parameter structures. You must use the standard event names defined by Google:

Each event’s “items” array must contain product details (item_id, item_name, price, quantity, item_category). Without these parameters, GA4’s e-commerce reports remain empty. WooCommerce users can rely on the GTM4WP plugin to push these events into the dataLayer automatically. Shopify requires custom integration or third-party tools like Elevar. Headless commerce setups and custom platforms generally need bespoke development work.

Is the transaction_id in purchase events unique?

Every purchase event must include a unique transaction_id parameter. This prevents duplicate transactions from being recorded when a user refreshes the order confirmation page or navigates back to it. GA4 does not automatically deduplicate based on transaction_id, so you need to implement control at the tag level or prevent the event from firing on repeat visits to the confirmation page. Duplicate transactions inflate revenue data and distort return on ad spend calculations.

Has product data been validated?

Open GA4’s Monetisation > E-commerce Purchases report. Are product names correct? Are categories populated? Are prices logical? A dataLayer error can sometimes show all product prices as zero or in the wrong currency. After configuring e-commerce tracking, place two or three test orders and verify the data in Debug View. In the UK, make sure currency is set to GBP and prices include VAT where appropriate. In the US, verify USD currency and tax handling.

Data Quality and Privacy

Collecting data is not enough. The data must be accurate, and it must be collected in compliance with legal requirements. In 2026, GDPR compliance in the UK and EU, and state-level privacy laws in the US , cannot be ignored in analytics configurations.

Has internal traffic been filtered?

Visits from your own team pollute the data. To filter internal traffic in GA4, go to Admin > Data Streams > More Tagging Settings > Define Internal Traffic and add your office IP address. Then go to Admin > Data Settings > Data Filters and activate the filter. GA4 leaves the filter in “Testing” mode by default. Do not forget to switch it to “Active.” For remote workers whose IP addresses change, a GTM-based solution using cookies or custom dimensions may be necessary.

Is bot and spam traffic being blocked?

GA4 automatically filters known bot traffic, but it cannot catch everything. Referral spam (fake referral traffic designed to inflate numbers or promote spammy sites) is still a problem. If your reports show traffic from domains you do not recognise, add those domains to the Unwanted Referrals list under Admin > Data Streams. Server-level blocking through . htaccess rules or Cloudflare firewall rules provides a more comprehensive defence.

Is the cookie consent mechanism compatible with GA4?

Under GDPR (and the UK’s Data Protection Act 2018), collecting analytics data without user consent creates legal exposure. GA4’s “Consent Mode” feature sends anonymised pings before a user grants consent. Consent Mode v2 has been required by Google since 2024. Integrate Consent Mode with GTM using tools like CookieYes, Complianz, Cookiebot or OneTrust, all of which have GTM templates available. In the US, requirements vary by state, but implementing Consent Mode is increasingly considered best practice nationwide. The cost of non-compliance is significant: GDPR fines can reach up to 4 per cent of annual global turnover or EUR 20 million, whichever is higher.

Is there a data backup and export plan?

GA4 data lives on Google’s servers and is subject to Google’s policy changes. Exporting critical data regularly is sound practice. The BigQuery integration (available even in GA4’s free tier) automatically sends raw data to Google BigQuery. In BigQuery, you can run SQL queries, build custom reports and store data for the long term beyond GA4’s 14-month retention limit. BigQuery itself is free for the first 1 TB of queries and 10 GB of storage per month, which most sites will not exceed.

Setup Prioritisation

Completing every item on this list in a single day may not be realistic. Prioritise so that the most critical steps happen first.

Day one essentials: Account and property creation, data stream configuration, GTM installation, GA4 Configuration tag and Debug View verification. Without these, no data is collected at all, and every day of lost data is irrecoverable.

First week priorities: Custom events, Key Events (conversions), audience definitions, internal traffic filtering and unwanted referral listing. These determine data quality and reporting capability.

First month objectives: E-commerce tracking (if applicable), report customisation, BigQuery integration and Consent Mode configuration. These are advanced steps, but they make a significant difference over the long term.

Setting up GA4 correctly during the website build or immediately after launch is far easier than retrofitting later. Corrections made after the fact only produce clean data from the date of the fix. Everything before that date stays as it was. That is why completing this GA4 setup checklist before or immediately after your site goes live ensures your data-driven decision-making rests on solid foundations from the very beginning.

Frequently Asked Questions