Google Analytics 4: how to set up the ultimate marketing dashboard

Google Analytics 4 has been the web tracking standard since 2023, yet in 2026 the majority of small businesses use it poorly or not at all. The result: marketing decisions based on gut feeling instead of data, budget spent with no idea what works, monthly reports nobody reads because they're incomprehensible. This guide takes you from a proper GA4 setup to building a marketing dashboard in Looker Studio your team will actually use — including Meta and Google Ads data integration and automated reporting via n8n and AI automation.

#GA4: why everything changed from Universal Analytics

Universal Analytics thought in sessions and pageviews. GA4 thinks in events and users. Every user action — link click, scroll, video view, form submission — is an event with parameters. This paradigm shift means you can track any interaction without custom code, but it also means the default setup isn't enough: you need to configure the events that matter to your business.

The other fundamental change is the attribution model. Universal defaulted to last-click. GA4 uses data-driven attribution, an algorithmic model that distributes conversion credit across all touchpoints in the customer journey. This radically changes how you evaluate channels: the organic Instagram post that generated first contact gets credit, even if the conversion happened through Google Ads 3 weeks later.

#GA4 setup: correct configuration from day zero

A properly done GA4 setup takes 2-4 hours and saves you months of dirty data. Here's the operational checklist we use at +Click for every new web project.

Creating the property and data stream

Go to admin.google.com/analytics → Create property → enter name, timezone, currency. Then create a Web data stream: enter the site URL and enable Enhanced Measurement. This automatically activates: page_view, scroll (to 90%), outbound click, site search, video engagement, file download. Copy the Measurement ID (G-XXXXXXXX): you'll need it for the tag.

Google Tag Manager: why to always use it

Don't insert the GA4 tag directly in the site code. Use Google Tag Manager (GTM). Reasons: you can add and modify tags without touching code, manage consent mode for GDPR, activate custom triggers (CTA clicks, deep scroll, time on page), and centralize all pixels (Meta, LinkedIn, TikTok) in one place.

1. Create a GTM account and install the container snippet on the site (head + body).
2. Create a "GA4 Configuration" tag with your Measurement ID.
3. Set an "All Pages" trigger for the base GA4 tag.
4. Enable Consent Mode v2: consent initialization before the GA4 tag.
5. Test in Preview mode: verify page_view fires on every page.
6. Publish the container only after verifying all tags.

Consent Mode v2 for GDPR

Since March 2024 Google requires Consent Mode v2 for all sites using GA4 in the European Union. Consent Mode communicates user consent to Google: if the user declines cookies, GA4 collects aggregate data (behavioral modeling) without cookies. If they accept, it tracks normally. Implementation requires a CMP (Consent Management Platform) like Cookiebot, iubenda or Osano, configured in GTM.

#Key events to track for marketing

GA4 distinguishes between automatically collected events, enhanced measurement events, and custom events. For marketing, automatic events aren't enough: you must define those that represent value for your business. Here's the list we configure for every client.

Lead generation events

• generate_lead: submission of any contact form or quote request.
• form_start: user begins filling the form (measures abandonment rate).
• cta_click: click on a CTA button (with parameters: CTA text, page position).
• phone_click: click on phone number (triggers call on mobile).
• whatsapp_click: click on WhatsApp link/button.
• scroll_depth_75: user scrolled 75% of the page (interest signal).

E-commerce events

• view_item: product page view.
• add_to_cart: add to cart (with parameters: item_id, item_name, price, quantity).
• begin_checkout: checkout initiation.
• purchase: completed purchase (with value, currency, transaction_id).
• refund: refund (for net ROAS calculation).

Engagement and content events

• blog_read_complete: user read the full article (90% scroll + time > 2 min).
• video_complete: user watched the video to the end.
• share: user shared content.
• newsletter_signup: newsletter subscription.
• download: PDF, catalog, or price list download.

Each custom event is created in Google Tag Manager with specific triggers and parameters, then registered in GA4 (Admin → Events → Create) and, if it represents a business goal, marked as a conversion (Admin → Conversions → New conversion event). For those running Meta Ads campaigns, connecting GA4 conversion events is essential for cross-channel attribution.

#Custom dimensions: tailored data

Custom dimensions are additional parameters you attach to events to segment data in ways GA4 doesn't offer by default. They're GA4's hidden superpower and what separates a basic setup from a professional one.

Useful dimensions for marketing

For every event you can pass custom parameters. The ones we always configure for marketing clients are: content_type (blog, service, landing, product), cta_text (button text clicked), cta_position (header, hero, sidebar, footer), form_name (identifier of the submitted form), lead_source (organic, paid, referral, direct). These parameters become custom dimensions in GA4 and filters in Looker Studio.

To register a custom dimension: GA4 Admin → Custom definitions → Create custom dimension → choose name, scope (event or user), associated event parameter. Free tier limits: 50 event-scoped dimensions, 25 user-scoped. For most SMBs this is more than sufficient.

User properties: segmenting by user type

User properties are user-level dimensions, not event-level. Useful for: client_type (new, returning, vip), membership_status (free, premium), industry (if the site serves multiple sectors). These properties are set via GTM or JavaScript and persist for 14 months. They let you filter every report by user type — "what's the conversion rate for VIP users vs. new ones?" becomes a question answerable in 2 clicks.

#Looker Studio: building the ultimate marketing dashboard

Looker Studio (formerly Google Data Studio) is Google's free tool for creating interactive dashboards. It connects natively to GA4, Google Ads, Google Sheets, BigQuery and via connectors to Meta Ads, LinkedIn Ads, HubSpot and hundreds of other sources. It's the tool we use to build reports for every +Click client.

Essential marketing dashboard structure

An effective marketing dashboard has 4 sections. Page 1 — Overview: main KPIs (users, sessions, conversion rate, revenue/leads) with previous period comparison, weekly trend chart, breakdown by channel. Page 2 — Acquisition: detail by source/medium, campaign performance, new vs returning. Page 3 — Engagement: top pages, average time, key events, conversion funnel. Page 4 — Conversions: goal completions, attribution, value per channel, ROAS.

1. Create a Looker Studio report and add a GA4 data source (native connector).
2. Overview page: KPI scorecards + 30-day line chart + channel table.
3. Acquisition page: source/medium table with engagement and conversion metrics.
4. Engagement page: top pages, custom events, funnel visualization.
5. Conversions page: conversion table by channel, data-driven attribution.
6. Add global filters: date range, channel, device, country.
7. Share with the team: view link or scheduled PDF.

Charts and metrics that actually matter

The temptation is to fill the dashboard with charts. Resist. The metrics your team needs to see every week are: active users (how many arrive), engagement rate (how many interact), conversion rate (how many buy/contact), cost per conversion (how much you spend per result), ROAS (how much you earn per dollar spent). Everything else is on-demand detail, not the main view.

The +Click case study portfolio shows how correct tracking and a marketing dashboard enable budget optimization: if you don't measure, you don't improve. And if you measure the wrong things, you get worse while believing you're getting better.

#Connecting Meta and Google Ads data

A marketing dashboard showing only GA4 tells half the story. For the full picture — and to calculate real ROAS — you need to integrate spend and performance data from Meta Ads and Google Ads.

Google Ads: native connector

Google Ads connects to Looker Studio in one click: add data source → Google Ads → select account. Available metrics: impressions, clicks, cost, conversions, conversion value. The real power comes when you combine these metrics with GA4 data in the same chart: you can see Google Ads campaign cost next to revenue tracked in GA4, for a true ROAS.

Meta Ads: connector via Supermetrics or n8n

Meta Ads doesn't have a native Looker Studio connector. Options are: Supermetrics (paid connector, around $99/month for the base plan, most widely used), Funnel.io (enterprise alternative), or n8n free: extract data via Meta Marketing API, write to Google Sheets, connect the Sheet to Looker Studio. At +Click we use n8n for most clients: zero license cost and full customization of extracted data.

The n8n workflow is: scheduled trigger (every night at 3 AM) → Meta Marketing API node (extracts campaigns, ad sets, previous day's metrics) → Google Sheets node (writes data to a dedicated sheet) → optional AI node (generates a text comment on significant variations). In Looker Studio you connect the Google Sheet as a data source and have Meta data refreshed every morning.

#Automated reporting with n8n

The monthly report nobody reads is a non-automated report. If the team has to "remember to check the dashboard," they won't. The solution is bringing data where the team already works: Slack, email, WhatsApp. With n8n we build reporting systems that don't wait for someone to look.

Automatic alerts on anomalies

The most useful workflow: every morning n8n compares the last 24 hours of metrics with the 7-day rolling average. If anything deviates by more than 20% (positive or negative), it sends a Slack alert with detail: "Organic Google traffic yesterday: +42% vs 7-day average. Probable cause: blog article indexed yesterday." AI analyzes the delta and generates the explanation. The team reads 3 lines and knows what's happening.

Weekly AI-commented report

Every Monday at 8 AM the team gets a Slack summary of the previous week: main KPIs with % change, top 3 pages by traffic, ad campaigns with ROAS, conversions by channel. Below each data point, a 1-2 line AI comment that contextualizes: not "traffic +15%" but "traffic +15%, driven by the Reel published Wednesday that generated 4,200 clicks to site."

This type of reporting eliminates the "weekly meeting to look at numbers": the team arrives already informed and discusses what to do, not what happened. For those wanting a deeper look at how AI integrates into marketing workflows, our AI marketing automation article covers the full stack.

Data doesn't serve the person who collects it. It serves the person who reads it at the right time to make the right decision. If the report arrives late or not at all, it's the same as not having one.

— Niccolò Giuseppetti, founder +Click

#FAQ GA4 and marketing dashboards