Google Analytics has redesigned Dashboards inside GA4: any report can now become a dashboard in one click, cards can be dragged and resized on a grid canvas, and funnel visualisations can sit directly next to score cards and line charts. Once published, a dashboard goes straight into the left-hand navigation, skipping the old Library step that made custom views hard to find. For product and marketing teams, that’s a real unlock — a stakeholder-facing KPI view that used to take a data analyst an afternoon to wire up in Looker Studio can now be built in minutes, by anyone with edit access.
The problem is that “anyone can build one” quickly becomes “everyone has a slightly different one.” A dashboard that mixes an event-scoped metric with a session-scoped dimension will silently double-count or misattribute rows, and nothing in the drag-and-drop builder warns you when that happens. Because dashboards are now cheap to publish, they proliferate — and without a way to see who built what, from which underlying report, teams end up debating whose number is right in a meeting instead of trusting any of them. The fix isn’t to stop people building dashboards; it’s to track how they’re built and used so drift gets caught before it reaches a leadership deck.
Data Points to Track
- Dashboard creation events — who published a new dashboard, from which source report, and how many cards it contains
- Scope mismatches — cards combining event-scoped and session-scoped dimensions on the same canvas, which GA4 allows but silently skews totals
- Card-level metric definitions, captured at publish time, so a later change to an underlying custom metric or audience doesn’t retroactively change historical numbers without anyone noticing
- View counts per dashboard to identify which ones stakeholders actually rely on versus ones nobody opens after week one
- Duplicate dashboard detection — near-identical dashboards built by different people covering the same KPIs, a strong signal of tooling fragmentation
- Edit history and last-modified-by, since dashboards published straight to navigation have no built-in approval step
Setup Steps
- Audit existing dashboards across every GA4 property to find out how many already exist and who owns each one.
- Standardise a small set of canonical dashboards per property (acquisition, engagement, revenue) and document which report each card sources from.
- Restrict dashboard-publish permissions to editor-or-above roles if your GA4 property currently allows broader access, to reduce uncontrolled sprawl.
- Log dashboard metadata externally — name, owner, card count, source reports — on a schedule, since GA4 doesn’t expose a native audit trail for dashboard changes.
- Set a quarterly review to archive dashboards with near-zero views and merge duplicates before they cause a numbers dispute.
Actionable Insights
Once you’re tracking dashboard sprawl, the ratio of active dashboards (viewed weekly) to total published dashboards tells you whether the feature is being used as intended or turning into shadow reporting. A property with three canonical dashboards and high weekly views across all of them is in good shape; one with thirty near-duplicate dashboards and a handful of weekly viewers means the underlying report definitions have likely drifted apart, and it’s only a matter of time before two people bring conflicting numbers to the same meeting. Catching that drift in the audit log is far cheaper than reconciling it after a decision has already been made on the wrong chart.
Related Resources
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