How to Build Looker Dashboards

Building a dashboard in Looker for the first time can feel like a steep climb, but it boils down to a repeatable process. Once you understand the basic building blocks and the steps to connect them, you'll be creating insightful, interactive reports for your team. This tutorial breaks down how to plan, build, and refine a Looker dashboard from scratch, transforming raw data into clear, actionable visualizations.

Before You Build: Understanding Looker’s Core Concepts

To build anything in Looker, you first need to get familiar with its terminology. These three concepts are the foundation of everything you'll create.

• Models: Think of a model (.model file) as the blueprint for your data. A developer on your team uses LookML (Looker’s language) to define the database connections and specify which tables to use and how they join together. You won’t usually interact with this directly, but it’s what powers everything else.
• Explores: An Explore is your starting point for analysis. It's a curated view of your data, designed by a developer, that allows you to start pulling in fields without writing any SQL. For example, you might have an "Orders Analysis" Explore that gives you access to metrics like sales, customer info, and product details.
• Looks: A Look is a single, saved visualization. It’s a single chart or a table you’ve created from an Explore that answers one specific question. You can use these Looks as individual building blocks for your dashboards.

In simple terms: you use an Explore to create a Look (a single chart), and then you combine multiple Looks to create a Dashboard.

Step 1: Plan Your Dashboard with a Clear Goal

The most common mistake people make is jumping straight into Looker without a clear plan. A great dashboard isn’t just a collection of cool-looking charts, it’s a tool that answers specific questions and drives decisions. Before you drag and drop a single element, ask yourself these three questions:

Who is this dashboard for?

Are you building this for the executive team, the marketing department, or an account manager? An executive likely wants high-level KPIs like total revenue and profit margin. A marketing manager needs to see campaign performance, cost-per-acquisition, and conversion rates by channel. Knowing your audience dictates which metrics you'll include.

What question(s) should it answer?

Get specific. Instead of a vague goal like "show sales data," define the exact questions your audience needs answered. For a marketing team, this could be:

• Which marketing channels are driving the most user signups this quarter?
• What is our return on ad spend (ROAS) for the Facebook Ads campaign?
• How is website traffic from organic search trending month-over-month?

Writing these questions down first gives you a checklist for the visualizations you'll need to build.

How should the story flow?

A good dashboard tells a story. Think about how you want to present the information. A common and effective layout follows a top-down approach:

1. Top Section: Display high-level, "headline" KPIs. These are the main numbers your audience cares about (e.g., Total Revenue, New Customers, Total Marketing Spend).
2. Middle Section: Show trends and comparisons that provide context for the headline numbers. These are typically line or bar charts that visualize performance over time or across different categories.
3. Bottom Section: Provide granular, detailed data in tables. This allows users to dig into the specifics if a trend in the middle section piques their interest.

Step 2: Creating Your Visualizations (Tiles) from an Explore

With a solid plan, it's time to create the individual charts and tables, also known as "tiles," that will make up your dashboard. The primary way to do this is from an Explore.

1. Navigate to an Explore: From the main navigation, select "Explore" and choose the relevant starting point for your analysis (e.g., "Shopify Orders," "Salesforce Leads").
2. Select Dimensions and Measures: On the left-hand field picker, you'll see two types of fields:

• Dimensions: These are the things you want to group your data by. They are usually text, date, or geographic fields like ‘Order Date’, ‘Campaign Name’, or ‘Country’.
• Measures: These are the numbers you want to calculate. They are numeric fields or aggregations like ‘Sum of sales’, ‘Average order value’, or ‘Count of sessions’.

To build your query, simply click on the fields you need. For example, to see sales by country, you would select the ‘Country’ dimension and the ‘Total Sales’ measure.

1. Run the Query: Once you’ve selected your fields, click the "Run" button in the top right. Looker will generate a raw data table based on your selection.
2. Choose Your Visualization: Right above the data table, you’ll see the "Visualization" pane. Looker offers various chart types, including Bar, Column, Line, Pie, Scatter Plot, and more. Select the one that best illustrates your data. For our "sales by country" example, a Map or a Bar chart would work well.
3. Customize the Visualization: Click the "Edit" gear icon next to the visualization type to open the settings panel. Here you can customize everything from colors and labels to axes and trend lines. Take a moment to give your chart a clear, descriptive title.

Step 3: Assembling the Dashboard

Once you’ve created a visualization you’re happy with, it’s time to add it to a dashboard. You can group related visualizations together to create a comprehensive report.

Building from a Blank Dashboard

The cleanest way to start is by creating a new, empty dashboard and adding tiles one by one.

1. Create a New Dashboard: Navigate to the folder where you want to save your work, click "New" and select "Dashboard." Give it a clear name like "Q3 Marketing Performance."
2. Add Your First Tile: From your new dashboard, click "Add Tile." A window will pop up asking you to choose an Explore to begin building your tile. This will take you through the same process described in Step 2: pick your dimensions and measures, run the query, select and customize your visualization, and finally, give your tile a name and save it to the dashboard.

Saving a Look and Adding It

An alternative method is to save visualizations from an Explore as "Looks" first.

1. Save as a Look: After creating a visualization in an Explore, click the gear icon in the top right and select "Save." You can choose to save it as a Look. Give it a title and save it to a folder.
2. Add the Look to Your Dashboard: Go to your dashboard and enter "Edit mode." Click "Add," then "Look," and search for the one you just saved. This adds a copy of your saved Look as a tile on the dashboard.

Repeat this process for all the questions you outlined in your planning phase until every tile is on the dashboard. Don't worry about the layout yet, just get all the necessary components onto the canvas.

Arranging Your Dashboard

Now, click and drag your tiles to arrange them according to the story you planned in Step 1. Place your headline KPIs at the top, trends below them, and detailed tables at the bottom. You can easily resize tiles by dragging their corners to create a visually appealing and logical flow.

Step 4: Making Your Dashboard Interactive with Filters

A static dashboard is useful, but an interactive dashboard is powerful. Filters allow you and your team to slice and dice the data to answer follow-up questions without needing to build new reports.

1. Enter Edit Mode: On your dashboard, click the "Edit dashboard" button.
2. Add a Filter: At the top of the dashboard view, click "Filters" then "Add Filter."
3. Configure the Filter: A window will appear. First, give your filter a title, like "Date Range" or "Campaign." Next, select the type of control, such as a date range slider, radio buttons, or a dropdown menu.
4. Link the Filter to Your Tiles: This is the most important step. Under "Tiles to Update," you’ll see a list of every tile on your dashboard. For each tile you want the filter to affect, choose the data field it should apply to. For example, for a "Date Range" filter, you’d link it to the ‘Order Date’ field on your sales tiles, the ‘Session Start Date’ on your website traffic tiles, etc.

Click "Add" once you've configured it. Now, when a user changes the date range at the top of the dashboard, all the linked tiles will update automatically to reflect that new period.

Final Thoughts

With your tiles created, layout finalized, and filters configured, you now have a functional, interactive Looker dashboard. Building an effective dashboard is a skill that blends data familiarity with thoughtful design - it’s both a science and an art that improves with practice, moving from just presenting data to telling insightful stories.

Creating powerful reporting shouldn't require a steep learning curve or hours spent wrestling with complex settings. At Graphed, we've designed a platform that skips the manual setup. Instead of clicking through menus to create tiles and link filters, you can just ask a question in plain English, like "Show me a dashboard of Shopify sales vs. Facebook Ad spend for the last quarter," and see it built for you in seconds. If you want to build professional dashboards with zero friction, you can create a free account with Graphed today.