Looker Studio Alerts and Scheduled Reports: Automating GA4 Monitoring (2026)

How do I set up automated GA4 alerts and reports in Looker Studio?

Looker Studio has native scheduled email delivery for reports (daily/weekly/monthly PDF or link) but no native conditional alerting — it cannot send a notification when sessions drop 30% or conversions spike. For conditional monitoring, you need either the GA4 Data API + Google Apps Script (free, flexible, works for most use cases) or a third-party tool like Supermetrics, Funnel.io, or a custom Cloud Function (paid, lower maintenance).

The native scheduled report feature handles "send this dashboard to stakeholders every Monday at 9am" reliably. For "alert me when traffic drops more than 20% week-on-week," build the Apps Script solution — it takes about 90 minutes and runs indefinitely for free.

Native Looker Studio scheduled reports

Setting up scheduled delivery

In any Looker Studio report you own or can edit:

1. Open the report
2. Click Share → Schedule delivery
3. Configure:

• Recipients: individual emails or Google Groups
• Frequency: daily, weekly, monthly
• Time: specific hour and timezone
• Format: PDF attachment or link-only email

What gets sent: a PDF snapshot of the current report state (with whatever date range and filters are currently set) or a link that opens the live report.

The PDF limitation: PDFs are static snapshots. Interactive controls (date pickers, dropdown filters) don't work in the PDF. Design scheduled-delivery dashboards with fixed date ranges (rolling 7-day or rolling 30-day) rather than date pickers — the PDF will always show the correct window.

Setting rolling date ranges for scheduled reports:

In each chart, set the date range to Auto and then configure the data source with a default date range of Last 7 days or Last 30 days. The scheduled PDF will always reflect the trailing window from the send date.

Who can receive scheduled reports

Recipients must have at least Viewer access to the report. If you're sending to stakeholders without Google accounts, either:

• Share the report publicly (View access to anyone with the link) — then recipients can receive the link email
• Use a third-party delivery tool that generates a PDF independently

Limitations of native scheduling

• No conditional sending (always sends on schedule, regardless of whether data changed)
• No threshold alerts ("only send if conversions dropped more than 10%")
• PDF rendering can fail on complex blended reports or very large date ranges — the report times out before the PDF renders
• No Slack, Teams, or non-Gmail delivery natively

Conditional alerting with Apps Script

For genuine anomaly detection, use the GA4 Data API via Google Apps Script. This runs on Google's servers on a cron schedule, costs nothing, and can send emails or Slack messages.

The weekly anomaly check template

This script compares this week's sessions and conversions to last week and sends an email if either metric deviates by more than 20%:

Setup steps:

1. Open script.google.com, create a new project
2. Paste the script, replace YOUR_GA4_PROPERTY_ID and email address
3. Run checkGA4Anomalies once manually to authorise GA4 Data API access
4. Add a time-driven trigger: Triggers → Add trigger → checkGA4Anomalies → Time-driven → Weekly → Monday 8am

Required OAuth scope: The script uses ScriptApp.getOAuthToken() which accesses whatever scopes you've authorised. Add https://www.googleapis.com/auth/analytics.readonly to the Apps Script manifest (appsscript.json) if not already present.

Adding Slack notifications

Replace the GmailApp.sendEmail call with:

Get the webhookUrl from your Slack workspace: Apps → Incoming Webhooks → Add to Slack → select channel → copy webhook URL. Store the URL in Script Properties (not hardcoded) for security.

The five metrics worth automated monitoring

Not all GA4 metrics benefit from automated alerts. Alert fatigue destroys the value of any monitoring system.

Monitor these:

1. Sessions week-on-week — fundamental traffic health. A >20% drop in 7 days is almost always a tracking issue or a real traffic problem worth investigating.

1. Key events (conversions) week-on-week — conversion tracking is fragile. A sudden conversion drop is often a broken tag, not a business problem. Catch it fast.

1. Purchase revenue vs previous period — if you have e-commerce, revenue variance is your highest-stakes signal. Even 10% drops are worth investigating.

1. Bot/spam traffic spikes — sessions from unexpected countries or with impossible engagement metrics (0ms session duration, 100% of sessions from a single IP range). Automate this check against your normal geographic distribution.

1. Tag coverage % (events with valid session IDs) — events arriving in GA4 without a valid ga_session_id parameter indicate a dataLayer or tag configuration issue. Monitor the ratio of events with missing session IDs as a proxy for tracking health.

Don't automate alerts on these:

• Day-over-day metrics for properties with weekly seasonality (weekday vs weekend traffic patterns generate constant false positives)
• Bounce rate / engagement rate (fluctuates with content mix, not just tracking health)
• Average session duration (affected by definition changes, consent mode shifts)
• New vs returning user ratio (reporting identity changes affect this more than genuine behaviour shifts)
• Any metric during your own marketing campaigns (traffic spikes are expected and intentional)

Scheduled report design for stakeholders

For stakeholder-facing scheduled reports, format the dashboard for the PDF delivery format:

Layout principles for scheduled PDFs:

• Use A4/Letter landscape orientation in Looker Studio report settings
• One page of key scorecards at the top (sessions, conversions, revenue, engagement rate vs prior period)
• One page of time-series trend (rolling 30 days with comparison period overlay)
• One page of channel breakdown table
• Three pages maximum — PDFs beyond this rarely get read in full

Date range for scheduled reports:

Configure each chart with Last 28 days as the default date range (28 days rather than 30 for clean week alignment). Add a comparison to Previous period for the scorecard row.