How to Create an Analytics Dashboard in Looker

Creating a Looker dashboard can turn a mountain of raw data into your most valuable business asset. But staring at a blank screen and figuring out where to start can feel overwhelming, especially with concepts like LookML and "Explores" to learn. This guide will unpack the jargon and walk you through the entire process step-by-step, from initial setup to building a powerful, interactive dashboard that your team will actually use.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

First Things First: What is a Looker Dashboard?

In Looker, a dashboard is more than just a collection of charts and graphs, it's a live, interactive hub for your most important metrics. Think of it as a central control panel for your business, department, or specific project. While a traditional report might give you a static snapshot of last month's performance, a Looker dashboard provides a real-time view that you and your team can interact with to uncover insights.

The primary goal is to centralize key performance indicators (KPIs) in one place so you can:

• Monitor business health at a glance.
• Share data-driven insights across your organization.
• Empower team members to answer their own questions without needing an analyst.
• Drill down into the data to understand the "why" behind the numbers.

Before You Build: Laying the Groundwork

Jumping straight into building without a plan is the fastest way to create a dashboard nobody looks at. A few minutes of prep work will save you hours of frustration and ensure you build something genuinely useful.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

1. Define Your Audience and Goals

Before you drag and drop a single chart, ask yourself two simple questions:

1. Who is this for? The needs of a C-suite executive are very different from those of a social media manager. An exec might want a high-level view with KPIs like Monthly Recurring Revenue (MRR) and Customer Acquisition Cost (CAC). A social media manager will need granular data on campaign engagement, click-through rates, and conversions per platform.
2. What questions should it answer? A great dashboard answers specific business questions. Instead of aiming for "a marketing dashboard," define the questions it needs to solve, such as, "Which marketing channels are driving the most qualified leads this quarter?" or "How is our sales team's pipeline velocity trending month-over-month?"

A mission statement like, "This dashboard will help the marketing team understand campaign ROI by tracking spend vs. generated revenue for our top channels" is far more effective than a vague goal.

2. Understand Looker's Core Concepts

Looker has its own terminology, and understanding these few core concepts is essential. You don't need to be an expert, but knowing what these terms mean will make the process much smoother.

• LookML: This is Looker's modeling language. Under the hood, data engineers use LookML to define the dimensions (the things you group by, like Date, Country, or Product Category) and measures (the things you calculate, like Total Sales, Average Order Value, or User Count) available to you. You can think of it as the structured rulebook that helps Looker understand your raw data. For the most part, you won't need to write LookML yourself, but it's helpful to know it’s what powers everything.
• Explore: An Explore is the starting point for asking questions of your data. It’s an interactive environment where you can select dimensions and measures to build a query, filter a dataset, and create a single visualization. Your Looker admins will have set these up for you (e.g., an "Orders" Explore, a "Users" Explore).
• Look: A Look is a single, saved data visualization. Once you build a useful table or chart in an Explore, you can save it as a Look to use again later or add to a dashboard.
• Dashboard: A collection of multiple Looks (often called "tiles") arranged on a single page.

The Main Event: Building Your Dashboard, Step-by-Step

With the planning phase complete, it's time to get hands-on and start building. We'll start by creating individual visualizations ("Looks") and then assemble them into a dashboard.

Step 1: Start with an Explore

Everything in Looker starts with an Explore. In the main navigation, click on Explore. You'll see a list of available data models. Let's say we're building a sales dashboard and want to see sales over time. You would choose the pre-configured Orders Explore.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Step 2: Build Your First Visualization (a "Look")

Once you're in the Explore interface, you’ll see a left-hand panel with all the available dimensions and measures defined in your LookML model.

1. Select your data: First, select a dimension. Let's start with a time-based one. Under the "Orders" dimension group, you might see "Created Date." Click on that. Next, select a measure, like "Total Sale Price."
2. Run the query: Click the Run button in the top right. Looker will now query your database and display the results in a simple data table beneath the settings — you now have a list of total sales for each creation date listed.
3. Choose a visualization type: A big table of data isn't very insightful. Above the results, you'll see a Visualization tab. Click it. Looker provides a variety of chart types like Bar, Column, Line, Area, and Pie. Because we're looking at data over time, a Line chart is a perfect choice. Select it. Looker will automatically render your data as a line chart.
4. Customize your chart: You can click the Edit button in the Visualization tab to adjust colors, axis labels, reference lines, and more. For example, you can give the Y-axis the label "Total Revenue" to make it clearer for your users.

Step 3: Save Your Visualization to a Dashboard

Now that you have a chart you're happy with, it's time to save it so it can be added to a dashboard.

1. In the top right corner of the Explore page, click the gear icon and select Save.
2. A dropdown will appear. Choose To an existing dashboard if you have one ready, or As a new Look if you just want to save it on its own first. For this guide, let’s go directly to a dashboard. Select To a dashboard.
3. Give your Look a descriptive title, like "Revenue Over Time."
4. Looker will now ask you which dashboard you want to save it to. Since this is our first one, there won’t be one to select. Instead, a prompt in the "Add to Dashboard" pop-up will appear, and you can create one by naming it "Sales Performance Dashboard" and clicking Ok.
5. Click Save to Dashboard. Congrats, you've just created your first dashboard and added your first tile!

Step 4: Arranging Your Dashboard Layout

Navigate to the folder where you saved your new dashboard and open it. You'll see the single chart we just created. To add more and organize them, click on Edit dashboard in the top right.

You are now in edit mode. From here, you can:

• Add more tiles: Repeat Steps 1-3 to create more Looks for things like "Sales by Region" or "Top Selling Products." When you save them, add them to your new "Sales Performance Dashboard." Alternatively, in Edit mode, you can click "Add Tile" to create a new one directly from the dashboard.
• Resize tiles: Click and drag the bottom-right corner of any tile to resize it. Larger tiles naturally draw more attention, so make your most important KPIs bigger.
• Rearrange tiles: Click and drag tiles to move them around. A common best practice is to place high-level summary numbers (like total revenue or total orders) at the top, followed by more detailed charts and tables below.

Level Up: Adding Interactive Filters

A static dashboard is good, but an interactive dashboard is great. Filters empower your users to segment the data and explore it for their specific needs, reducing the number of one-off report requests you get.

How to Add a Dashboard Filter

Let's add a date range filter so users can view sales performance for a specific period.

1. While in Edit dashboard mode, click Filters from the top navigation bar and select Add Filter.
2. A configuration window will pop up. First, give the filter a Title, such as "Date Range."
3. Choose the Control type. For dates, a "Date Range" is perfect.
4. In the "Tiles To Update" tab, you'll need to tell Looker which field in each tile this filter will apply to. For our "Revenue Over Time" tile, search for and select the Created Date field. If you had other tiles on the dashboard (like "Orders by Region"), you would link the filter to their date fields as well. This is a critical step — forgetting it will mean the filter doesn't do anything for that specific tile.
5. Configure a Default Value if you’d like, such as "Last 90 days."
6. Click Add. The new filter will appear at the top of your dashboard.

Remember to save your changes by clicking Save. Now, anyone viewing the dashboard can change the date filter, and all the connected charts will update automatically.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Tips for a Dashboard People Actually Use

Building something functional is one part of the puzzle, making it effective is the other. Following a few design principles can make a world of difference.

• Keep it simple: A cluttered dashboard can cause confusion. Don't try to answer every possible question on one screen. If needed, create separate dashboards for different audiences or objectives (e.g., an "Executive Summary" dashboard vs. a "Campaign Performance" dashboard).
• Tell a story: A good dashboard has a logical flow. Start with high-level summaries at the top ("what happened"), then move to trends and comparisons in the middle ("why did it happen?"), and end with granular data at the bottom for those who want to dig deeper.
• Use clear titles and descriptions: Every tile should have a title that clearly states what it’s showing. Use text tiles to add context, explanations of metrics, or summaries of key insights for your viewers.
• Choose the right chart: Use line charts for trends, bar/column charts for comparisons, pie/donut charts for proportions (with few categories), scatter plots for correlations, and scorecards for single, critical KPIs.

Final Thoughts

Building a dashboard in Looker is about transforming raw data logs into a clear, compelling story. By carefully defining your goals, translating your questions into visualizations, arranging them logically, and adding interactive filters, you can create a powerful analysis and reporting tool that truly helps your whole team make smarter, data-informed decisions.

The biggest hurdle with platforms like Looker is often the time it takes to grasp the intricacies of the data model and the manual effort of building each visualization. It's an incredibly powerful system, but it comes with a steep learning curve. At Graphed, we’ve worked to remove this friction by letting you create dashboards using simple, natural language. Instead of wrangling with settings and learning new software, you can just ask questions like ‘build a dashboard showing my top ad campaigns by ROI from Facebook Ads versus Google Ads’ and get a live, automated dashboard in seconds. If you want to skip the complexity and go straight to the insights, you can give Graphed a try.