How to Build a GA4 Dashboard in Looker Studio That People Actually Use

Most Looker Studio dashboards fail the same way: too many charts, no clear hierarchy of information, and metrics that require context the viewer doesn’t have. The result is a dashboard that gets opened, not understood, and eventually ignored.

This guide is about building dashboards that get used — by clients who need simple answers and by analysts who need detail without noise. We’ll cover the GA4 data connector, blending multiple sources, useful chart types, and the design decisions that separate dashboards people trust from ones they dismiss.

---

Connecting GA4 to Looker Studio

Option 1: Native GA4 Connector

Looker Studio has a built-in GA4 connector that works without any setup beyond authentication.

1. Looker Studio → Create → Report
2. Add data → Google Analytics
3. Select your GA4 account and property
4. Looker Studio loads your available dimensions and metrics

What you get: Access to all GA4 standard dimensions and metrics. Segment filtering. Date range controls. No coding required.

Limitations: Processed data only — no raw event-level access, no custom SQL. Sampling may apply for complex queries on large properties.

Option 2: BigQuery Connector (Recommended for Accuracy)

If your GA4 property exports to BigQuery, connect Looker Studio directly to BigQuery instead. This gives you:

• Unsampled data
• Custom SQL for metrics GA4’s UI doesn’t offer (e.g., sessions with custom calculations)
• Ability to join GA4 data with Shopify, CRM, or advertising data

1. Add data → BigQuery
2. Select your project, dataset, and table (e.g., events_* with a date filter)
3. Use Custom Query for computed fields:

SELECT
  event_date,
  COUNT(DISTINCT CONCAT(user_pseudo_id, CAST(
    (SELECT value.int_value FROM UNNEST(event_params) WHERE key = 'ga_session_id') AS STRING
  ))) AS sessions,
  COUNTIF(event_name = 'purchase') AS purchases,
  SUM(IF(event_name = 'purchase',
    (SELECT value.double_value FROM UNNEST(event_params) WHERE key = 'value'), 0
  )) AS revenue
FROM `your-project.analytics_XXXXXXXXXX.events_*`
WHERE _TABLE_SUFFIX >= FORMAT_DATE('%Y%m%d', DATE_SUB(CURRENT_DATE(), INTERVAL 90 DAY))
GROUP BY 1
ORDER BY 1

This query runs in BigQuery and streams results to Looker Studio — no sampling, full accuracy.

Official reference: Looker Studio GA4 connector

---

Dashboard Architecture: What Goes Where

Before adding a single chart, decide on the hierarchy of your report. A good structure:

Page 1: Executive Overview (30-Second Read)

Audience: Business owner, CMO, client Goal: Answer “is the business growing or declining?” in one look

What belongs here:

• 3–4 KPI scorecards (Sessions, Revenue, Conversion Rate, AOV) with period comparison
• One trend line chart showing sessions and revenue over the last 90 days
• Traffic source breakdown (donut or bar — 5 sources max)
• One red/amber/green status indicator if there’s a performance threshold to track

What doesn’t belong here:

• More than 5 chart types
• Anything that requires the viewer to calculate what it means

Page 2: Traffic Acquisition Detail

Audience: Marketing team, SEO/PPC manager Goal: Answer “what channels are growing/declining and which drive the most valuable sessions?”

Charts:

• Table: Channel grouping × Sessions, Users, Conversion Rate, Revenue (sortable)
• Time series: Sessions by channel (last 30 days, channel comparison lines)
• Scorecard comparison: This month vs same period last year

Page 3: Content / Pages Performance

Audience: Content team, SEO manager Goal: Answer “which pages drive the most traffic and engagement?”

Charts:

• Table: Page path × Pageviews, Unique Users, Average Engagement Time, Conversions
• Bar chart: Top 10 pages by pageviews
• Filter: Allow drill-down by page category or section (if you have custom dimensions)

Page 4: Ecommerce Performance (if applicable)

Audience: Ecommerce manager, buyer Goal: Answer “what’s selling, what’s not, and what’s the purchase funnel health?”

Charts:

• Funnel chart: view_item → add_to_cart → begin_checkout → purchase
• Table: Product name × Units sold, Revenue, Revenue per item
• Scorecard: Transactions, Revenue, AOV, Return Rate (if tracked)

---

Building the Scorecards

Scorecards are the highest-value elements on an executive dashboard. Set them up correctly:

1. Insert → Scorecard
2. Select your metric (e.g., Sessions)
3. Comparison → enable → select “Previous period” or “Same period last year”
4. Style → set font size large enough to read at a glance; use colour for positive/negative delta

Calculated metrics for Looker Studio:

Some metrics aren’t available natively and need to be created:

• Bounce Rate (GA4 definition): Bounced Sessions / Sessions — where GA4 defines a bounce as a session with < 10 seconds engagement time and no second event

  • Create a calculated field: 1 - (Engaged sessions / Sessions)

• Average Order Value: Purchase revenue / Transactions

  • Calculated field: Purchase Revenue / Transactions

• Revenue per Session: Purchase Revenue / Sessions

To create a calculated field: In your data source → Add a field → write the formula.

---

Useful Chart Types for GA4 Data

Time Series Chart

The workhorse of analytics dashboards. Use for:

• Sessions over time
• Revenue trend
• Multiple metrics on one timeline (use secondary axis for metrics at different scales)

Settings to get right:

• Set default date range to “Last 30 days” — users can change it
• Enable “Breakdown dimension” to compare channels on one chart
• Smooth lines if there’s too much noise (weekly data often looks cleaner than daily)

Geo Map

Useful for businesses with geographic spread. Shows which countries/regions drive traffic.

• Zoom area: Set to your primary market (e.g., United Kingdom)
• Colour by: Sessions or Revenue

Funnel Chart

For ecommerce conversion funnel analysis. Native Looker Studio doesn’t have a funnel chart type — use a Bar Chart with manual step data, or use a community funnel visualisation from the Community Visualisations gallery.

Alternatively, use a table showing each funnel step count and calculate drop-off rate as a calculated field.

Pivot Table

For cross-dimensional analysis — e.g., Channel × Device Category with Sessions as the metric. Shows data density well but can become complex. Use only when the 2D comparison adds genuine insight.

---

Blending Multiple Data Sources

Looker Studio’s Data Blending feature lets you combine GA4 data with Shopify, Google Ads, or other sources on a single chart.

Blend GA4 with Google Ads (ROAS Dashboard)

1. Add two data sources: GA4 and Google Ads
2. In a chart → Data → Blend data
3. Join key: Date (both sources have daily data)
4. From GA4: Revenue, Sessions
5. From Google Ads: Cost, Impressions, Clicks
6. Calculated field in the blend: ROAS = Revenue / Cost

This gives you a single table with spend and revenue side by side.

Blend GA4 with Shopify (via Sheets)

Export Shopify reports to Google Sheets (manually or via a Shopify export app), then blend:

1. Google Sheets connector → your Shopify export sheet
2. Blend with GA4 on Date
3. Compare GA4 Revenue vs Shopify Revenue to track the discrepancy percentage

Calculated field: GA4 Revenue / Shopify Revenue — shows the tracking accuracy ratio over time.

---

Filters and Controls

Controls make dashboards usable for different audiences without building separate reports.

Date Range Control

Always add a Date Range Control widget. Set the default to “Last 30 days” — but let users change it. Place it top-right where it’s conventional.

Dropdown Filter

Add dropdown filters for common segmentation dimensions:

• Device category (Desktop / Mobile / Tablet)
• Country
• Traffic source / medium
• Page section (if using a custom dimension for content categories)

Segment Switcher (Advanced)

Using data blending and control groups, you can simulate GA4 segments in Looker Studio:

1. Create multiple data sources filtered to different segments (e.g., one filtered to mobile, one to desktop)
2. Blend them with conditional display using a Dropdown Control
3. Charts update based on the selection

This is complex to build but gives non-technical users self-service segment analysis.

---

Performance Optimisation

Slow dashboards don’t get used. Common causes and fixes:

Too many charts pulling from different date ranges: Each unique date range triggers a separate data query. Use consistent date ranges across all charts on a page — they can share cached results.

No data freshness caching: Looker Studio caches data for 12 hours by default. For daily reporting, this is fine. For real-time monitoring, set cache to “None” but accept slower load times.

BigQuery queries without date filters: If your BigQuery connector runs a query that scans all historical data, every dashboard load is expensive. Always include WHERE _TABLE_SUFFIX >= ... in your BigQuery queries.

Too many pages: A 15-page dashboard loads slowly and overwhelms users. Focus on 4–5 pages maximum.

---

Sharing and Access Management

When sharing a Looker Studio report:

• View access: Anyone with the link can view — no Google account required if you set it to “Anyone with the link”
• Edit access: Limit to your team — editors can modify the report
• Embed: Reports can be embedded in client portals using the embed URL (no Looker Studio account required to view)

Scheduling email delivery: Looker Studio reports can be emailed on a schedule:

1. Share → Schedule email delivery
2. Set recipients, frequency (daily/weekly/monthly), time
3. Recipients receive a PDF snapshot of the report

This is useful for executive stakeholders who won’t log in to view a live dashboard.

---

A Template to Start From

Rather than building from scratch, start with one of Google’s Looker Studio Gallery templates — several GA4-specific templates exist. Then customise the layout, add your branding (logo, brand colours in the theme settings), and swap metrics for the ones relevant to your business.

We build custom Looker Studio dashboards that combine GA4, Shopify, Google Ads, and other data sources for e-commerce and SaaS clients. If you need a dashboard your whole team will actually use — not just open — book a free consultation.