📊 Visualizations in Looker

• 1. Types of Visualizations
• 2. Customizations
• 3. Dashboards
• 4. LookML Dashboards (Optional)
• 5. Drill Fields & Interactions

1. Types of Visualizations

Looker supports the following chart types in its Explore and Dashboard UI — ordered from most frequently used to least:

• Table
• Bar chart
• Line chart
• Column chart
• Pie chart
• Single value
• Area chart
• Scatter plot
• KPI (with trend line)
• Donut chart
• Funnel chart
• Gauge
• Boxplot
• Map (points)
• Map (regions)
• Waterfall chart
• Treemap
• Timeline
• Word cloud
• Custom visualizations (via JavaScript plugin)

2. Customizations

Looker provides multiple ways to customize visualizations to match user needs or business branding.

🎯 Axes, Series, Colors

• Toggle x/y axes, gridlines, and label orientation
• Control color schemes, series ordering, and labeling
• Apply value formatting (currency, %, decimals) directly in the UI

These settings are available per tile in the “Edit” > “Series” and “Plot” tabs.

➕ Table Calculations and Custom Fields

• Table calcs allow you to write Excel-style formulas on top of query results (e.g., ${revenue} / ${orders})
• Custom fields can be built via point-and-click UI (no LookML needed)
• Use percent_of_previous, rank, moving average, and more

Useful for lightweight transformations without touching backend LookML.

🧠 HTML Tooltips (Advanced)

Looker supports custom HTML in tooltips to show dynamic context when hovering over chart elements.

✅ Where to use:

• Line, bar, scatter, and other charts with hover elements

✅ How to define: In your view file, you can define an html block inside a dimension or measure.

Example:

measure: avg_order_value {
  type: average
  sql: ${TABLE}.order_amount ;;
  html: "<b>Avg:</b> \${value}"
}

✅ Supported tokens:

• \${value} — shows the measure’s rendered value
• \${name} — shows the field label
• You can embed other HTML elements like <b>, <br>, <i>, and inline styling

Tooltips are especially useful when:

• You want to show multiple metrics without cluttering the chart
• You want to show derived context (e.g., % change from previous)

📌 Notes:

• Only available in LookML-defined fields, not custom fields in Explore
• Helps make dashboards feel more interactive and insightful

3. Dashboards

Dashboards in Looker are visual collections of saved queries and charts — often created directly from the Explore UI.

🛠️ Creating from Explore vs LookML

• From Explore (UI-based):
  • Most common method
  • Create a visualization → click “Save to Dashboard”
  • Allows drag-and-drop editing of tiles
  • Saved in the Looker front-end (not part of Git versioning)
• From LookML:
  • Hand-coded .dashboard.lookml files
  • Covered separately in section 4

🎛️ Dashboard-Level Filters

• Filters can be added to a dashboard to apply globally across multiple tiles
• Fields used in filters must exist in all tiles where the filter is applied
• You can set default values and allow user input

Example use cases:

• date filter affecting all charts
• region filter syncing across multiple visualizations

🧱 Layout Options (Tiles, Spacing)

• Tiles = individual visualizations on the dashboard
• You can adjust:
  • Width / height
  • Spacing between tiles
  • Order of appearance (drag to rearrange)
• Text tiles can be added for titles, sections, and instructions

📌 Notes:

• You can schedule dashboards for email delivery (PDF, CSV)
• Dashboards can be shared with groups or embedded externally
• Good layout = fewer tiles, clear labels, grouped metrics

4. LookML Dashboards (Optional)

LookML dashboards are defined as code using .dashboard.lookml files — giving you version control, reusability, and programmatic flexibility.

🧱 Syntax: dashboard: and element: blocks

Basic structure:

dashboard: user_summary {
  title: "User Summary Dashboard"
  filters: [ ... ]
  elements: [ ... ]
}

Inside the dashboard block:

• filter: defines global filters
• element: defines each chart/tile
• You can reference saved Looks, inline queries, or explores

Example element:

element: total_users_tile {
  type: single_value
  query: {
    model: ecommerce
    explore: users
    fields: [users.total_users]
  }
}

📊 Static vs Dynamic Dashboards

• Static: Hardcoded dashboards with predefined tiles and filters
• Dynamic: Can include templated filters, variables, and links that change based on user inputs or URL params

You can use Liquid variables to customize behavior dynamically. For example:

• User attribute reference:

  {{ _user_attributes['region'] }}
  

• Conditional logic:

  {% if parameter_name._value == 'X' %}
    Show this content
  {% else %}
    Show alternate content
  {% endif %}
  

  These work inside fields like label, html, and other blocks that support dynamic runtime evaluation.

🤔 When to Use LookML Over UI

Use LookML dashboards when:

• You need Git version control
• Dashboards are part of a deployment pipeline
• You want to reuse tiles across models/environments
• You need complex filtering or dynamic linking
• You’re building for embedded analytics

📌 Notes:

• LookML dashboards are managed in Git and require deploy to update
• They support Markdown, text blocks, and layout customization via code

5. Drill Fields & Interactions

Drill behavior lets users click into a value on a chart and either:

• See a deeper breakdown (e.g., transactions behind total sales)
• Navigate to another Explore or dashboard
• Jump to an external tool (e.g., Salesforce, internal URL)

🧠 How to Define Drill Behavior

Drills are defined at the dimension or measure level in the view file.

Example:

dimension: user_id {
  type: number
  sql: ${TABLE}.user_id ;;
  drill_fields: [orders.order_id, orders.order_amount]
}

This means clicking a user_id will show a table of related order IDs and amounts.

🔗 Linking to Explores or Dashboards

Instead of showing fields, you can link to another Explore or dashboard.

Example:

link: {
  label: "View User Details"
  url: "/explore/ecommerce/users?fields=users.id,users.email&user_id="
}

You can also link to LookML dashboards using:

url: "/dashboards/user_summary_dashboard?user_id="

🌍 Jump-to-URL or Deep Links

Use link: blocks with url: and Liquid templating (``) to:

• Jump to 3rd-party tools (e.g., Zendesk, Amplitude, Jira)
• Pass parameters dynamically
• Embed values in external systems

📌 Notes

• drill_fields: = in-app table of details
• link: = navigation to another page or tool
• Drills enhance interactivity without cluttering the main tile