Docs

GA4 Dashboards: What the New Canvas Does and Doesn't Show

GA4's new Dashboards report gives you a 15-card drag-and-drop canvas — but it can quietly hide the event-level detail your team still needs.

Analytics

Google Analytics 4 shipped a genuinely new report type this month: Dashboards, a grid-based canvas where you drag dimensions and metrics straight onto the page instead of building each chart as a separate Exploration. Six chart types are available at launch — scorecards, tables, line charts, bar charts, donut charts, and funnel charts — and a finished dashboard can be published directly into the Reports navigation, so anyone on the team can open it without knowing how it was built. For teams tired of rebuilding the same weekly view in Looker Studio, that’s a real time save.

The risk is what a scorecard-and-funnel canvas leaves out. Dashboards are capped at 15 cards on standard properties (30 on GA4 360), which forces teams to pick a handful of top-line numbers and quietly drop the segment-level breakdowns that explain why those numbers moved. A funnel card shows the drop-off between steps, but not which acquisition channel or device category is driving it — that still requires jumping back into Explorations. Teams that treat the new dashboard as their only view of the data will see the shape of a problem days before they can see the cause, because nobody kept the underlying breakdown anywhere durable.

Data Points to Track

  • Card composition per published dashboard — which dimensions and metrics were chosen for the 15 available slots, logged outside GA4 so you have a record when someone edits or replaces a card without telling the team
  • Funnel card step-to-step drop-off, paired with a saved Exploration that segments the same funnel by channel, device, and landing page — the detail the dashboard card itself won’t show
  • Scorecard metric definitions, since a scorecard can silently be pointed at a different date range or comparison period than the rest of the canvas, producing numbers that look inconsistent at a glance
  • Dashboard view and edit events, so you know who is actually relying on a given dashboard before you redesign or retire it
  • Cross-reference against raw event data for any metric a dashboard card summarises, to catch cases where a filter applied to one card silently excludes data another card includes

Setup Steps

  1. Decide the dashboard’s one job before building it — a 15-card limit punishes “everything” dashboards, so scope each one to a single audience (execs, growth, support) rather than trying to cover every metric anyone might ask about.
  2. Pair every funnel or scorecard on the canvas with a linked Exploration that holds the segment breakdown, and note that link somewhere the whole team can find it, since the dashboard card itself has no room to show it.
  3. Standardise the date range and comparison settings across all cards before publishing, and re-check them after any edit — a mismatched card is the most common source of “the numbers don’t add up” questions.
  4. Track who published and last edited each dashboard using GA4’s own admin activity where available, or a simple changelog, so a metric definition change doesn’t get discovered three weeks later during a leadership review.
  5. Export a monthly snapshot of each dashboard’s underlying data rather than relying on the live canvas as your only record — GA4’s data thresholds and sampling can shift a metric retroactively, and a live dashboard won’t tell you it happened.

Actionable Insights

A dashboard card that suddenly shows a flat or dropping trend is a starting point, not an answer — the fifteen-card ceiling means the explanation almost always lives in an Exploration you haven’t opened yet. When a scorecard and a related funnel card disagree about the direction of a metric, check date range and comparison settings first; that mismatch is far more common than an actual data anomaly. And because dashboards can be published and shared widely with no built-in audit trail beyond GA4’s admin logs, a metric that starts getting quoted in exec meetings is a good trigger to double-check its definition matches what the team building it originally intended.

Expert help

Need help tracking this in your app?

Our team sets up analytics pipelines for mobile and web teams every day. Talk to us and get your first events flowing in under an hour.

Talk to an expert