GA4 Dashboards Are Now Available: Build a One-Page Marketing Report in 2026

Most marketing reports fail for the same reason: they show everything, so they settle nothing. Google Analytics finally has a native surface for fixing that, and the temptation is to fill it with all fifteen cards on day one.

Google Analytics introduced native dashboards on 9 September 2026. They let teams arrange scorecards, tables, line charts, bar charts, donut charts and funnels on one drag-and-drop canvas inside GA4 — no export, no second tool, no connector to maintain.

The best first dashboard is not a collection of every available metric. It is a compact decision page showing reach, traffic quality, meaningful outcomes, and the channels or landing pages responsible for them.

The direct answer: begin with 8–10 cards tied to business questions. Show users and sessions for scale, engagement rate for visit quality, key events and session key event rate for action, revenue or qualified leads for value, and channel and landing-page views for diagnosis. Compare against a relevant earlier period. If a card cannot change a decision, leave it out.

What Google launched in September 2026

Google's 9 September release note describes dashboards as a flexible, single-page way to view key performance indicators. The documentation confirms you can position and resize visualizations on a grid, drag dimensions and metrics onto cards, save the page, and publish it into the Reports navigation.

An Editor or Administrator can create and publish a dashboard. Anyone with access to the property can view a published one. That sharing model matters: it is convenient for an internal reporting page, but a dashboard should not carry a label or breakdown that another property user should not see.

The current limits shape what you can build. A standard property allows up to 15 cards; a premium property allows up to 30. Google's documentation lists API support, segments and card-level comparisons as unsupported at launch. Published dashboards are shared with the property rather than privately with selected viewers.

A note on terminology: If you are following an older tutorial, it will say "conversions" where GA4 now says "key events." Google split the two: key events live in Analytics, conversions live in Google Ads. Everything below uses the current naming.

Decide what the dashboard must help you do

A useful marketing dashboard answers four questions in under a minute:

• Are enough relevant people reaching the website or app?
• Are those visits showing meaningful engagement?
• Are visitors completing actions that matter to the business?
• Which channels and entry pages deserve investigation or investment?

Write these questions before you build a single card. Otherwise the easy-to-find metrics fill the page while the decision-making information stays buried three clicks away.

For a lead-generation business, the outcome might be a submitted enquiry, a booked consultation or a qualified call. For ecommerce, it is purchases, revenue and checkout completion. A publisher may care about engaged sessions, returning users and newsletter sign-ups. The dashboard should reflect the business model, not a generic template.

Ten cards for a practical one-page marketing report

This structure stays below the 15-card standard-property limit and balances volume, quality and outcomes. The table maps each business question to the cards that answer it; the detail follows below.

Use only verified metrics, and let the layout progress from scale to quality to outcomes to diagnosis.

1. Active users

A scorecard for distinct active users in the selected period. It measures audience scale but not commercial value, so never let it stand alone at the top of the page.

2. Sessions

Sessions show visits rather than unique people. Compare their trend against users and outcomes to understand repeat activity. Google's session documentation notes the default inactivity timeout is 30 minutes.

3. Engagement rate

A fast quality check, but a contact page, an article and a checkout step have different expected behaviours. Compare periods first, then investigate any change by channel or landing page.

4. Key events

Add only events representing meaningful actions. Google defines key events as actions important to the business. A purchase or qualified enquiry belongs here; every scroll or page view usually does not. Audit naming and counting before you trust the total.

5. Session key event rate

This asks what share of sessions included at least one selected key event. It ties traffic to action more clearly than raw event counts. Keep it beside sessions, because a rate can rise while total opportunities fall.

6. Revenue or a qualified-lead proxy

Ecommerce can use total revenue. Lead-generation businesses need a carefully defined key event or an imported qualified-lead measure. Validate purchase values, currency and duplicate events before trusting the number.

7. Sessions by default channel group

A bar or donut chart comparing recorded channels. Since Google added the dedicated AI Assistant channel on 13 May 2026, properties receiving recognized referrals will see it here alongside organic search and direct recorded with the medium ai-assistant and the campaign.

That channel is worth watching on its own card if assistant referrals are growing for you, but read it carefully: only recognized, trackable referrals are attributed, so the channel understates the real influence of AI tools on demand. Our guide to how much traffic AI tools are sending your website covers what GA4 captures, what lands in direct instead, and how to sanity-check the numbers.

8. Key events or revenue by channel

Traffic volume and business value are different questions. Put an outcome metric beside the channel breakdown so high-volume sources are not automatically treated as the best sources. Keep scope consistent — session-scoped channel dimensions are the clearest starting point for analysing individual visits.

9. Landing pages with sessions and outcome rate

A table of top landing pages connects acquisition to the first page of the visit. Include sessions and a relevant outcome rate. A page with moderate traffic and strong qualified action often deserves more attention than your highest-traffic article.

10. Weekly outcome trend

Finish with a weekly line chart for key events, qualified leads or revenue. Record campaigns, site releases, outages and tracking changes against it — GA4's annotations feature lets you mark these directly on the chart, so the team does not mistake an implementation change for customer behaviour.

How to build the dashboard in GA4

• Open the correct Google Analytics property and select Reports.
• Select Create, then choose Dashboard.
• Drag a scorecard, table, line chart, bar chart, donut chart or funnel onto the canvas.
• Add compatible dimensions and metrics to each visualization.
• Resize and arrange cards so the page reads from scale to quality to outcome to diagnosis.
• Apply a useful date range and check whether comparisons appear where supported.
• Save the dashboard, then publish it to the Reports navigation once it is ready for the wider property audience.

Some dimensions and metrics are incompatible and will appear unavailable. Google's dimensions and metrics guidance explains that (not set) appears when Analytics did not receive a dimension value. Resist the urge to hide data-quality problems for a cleaner-looking page — a blank you understand is worth more than a number you do not.

Native dashboard, Exploration or external report?

The new dashboard is useful, but it does not replace every reporting surface. Choose the lightest one that answers the question accurately.

Google warns that reports and Explorations can differ because supported fields, filtering, segments, retention and thresholds are not identical. Its reporting-surfaces comparison is worth reading before you treat two mismatched views as an implementation error.

Eight mistakes that make dashboards misleading

• 1. Filling every available card: Fifteen mediocre cards are not better than eight decisive ones. Keep the first screen focused and push investigation to standard reports or Explorations.
• 2. Reporting traffic without outcomes: Users and sessions show scale, not success. Pair acquisition with qualified actions, revenue or another defensible business result.
• 3. Mixing scopes without understanding them: First-user, session and event dimensions answer different questions. A dashboard can be technically valid and still conceptually confusing if these are mixed casually.
• 4. Treating every key event as equally valuable: A newsletter sign-up, an enquiry and a purchase may all matter, but they do not carry the same commercial meaning. Separate or label them rather than presenting one inflated total.
• 5. Ignoring tracking changes: A new tag, a consent configuration or an event rule can move a chart without any real change in demand. Keep a dated implementation log and annotate the trend.
• 6. Using totals without comparisons: A number needs context. Compare against a meaningful earlier period and account for campaign timing, weekends, festivals and seasonal demand.
• 7. Forgetting data thresholding: When Google Signals is enabled and a row represents a small number of users, GA4 suppresses that row rather than showing the figure. Stakeholders see gaps and assume tracking is broken. If a reporting property exists mainly to produce dashboards, check whether Signals needs to be on at all.
• 8. Publishing before checking access: Published dashboards are visible to everyone who can access the property. Review card labels, custom dimensions and table breakdowns before you publish.

Open the dashboard on the same weekday with a consistent comparison period. Identify the largest material change, locate its channel or page, rule out tracking changes, then assign one action and one owner. That action might be repairing an event, reviewing an underperforming landing page, expanding a high-quality traffic source, or waiting for more data. A dashboard earns its place when it shortens the path from observation to a responsible decision.

Frequently asked questions