How to Create a Dashboard in Looker with AI

Creating a dashboard in Looker Studio used to mean staring at a blank canvas, grappling with dimensions and metrics, and spending hours just to get the basics right. Now, by using AI, you can generate charts and reports simply by describing what you want to see. This article will show you how to use Duet AI inside Looker Studio to build dashboards using plain-English prompts, turning a time-consuming technical task into a quick conversation.

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

What is Duet AI in Looker Studio?

First, a quick clarification: Google's BI ecosystem can be confusing. Looker is the enterprise-level platform, while Looker Studio (formerly Google Data Studio) is the free, self-service tool most marketers and entrepreneurs are familiar with. The AI features we're discussing live within Looker Studio.

Duet AI is Google's integrated artificial intelligence assistant designed to help you analyze data and build visualizations faster. Instead of manually dragging and dropping fields or configuring chart settings, you can type a prompt like, "Show me website sessions by country for the last 30 days as a bar chart." Duet AI interprets your request, queries your data source, and generates the chart for you. Think of it as an analyst-in-training who does the heavy lifting, allowing you to focus on the story your data is telling.

Before You Begin: Connecting Your Data

The AI is smart, but it can't create anything without data. Before you start prompting, you need to connect a data source to your Looker Studio report. For many, this will be Google Analytics 4, but Looker Studio supports hundreds of connectors, including Google Sheets, Google Ads, and BigQuery.

For this tutorial, we will assume you're using a Google Analytics 4 property. The key is to ensure you have access to the data source and that the data itself is reasonably clean and organized. The AI works best when the underlying data has clear, logical naming conventions (e.g., "Revenue," "User," "Campaign Name").

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

Step-by-Step: Building a Dashboard with Duet AI

Let’s walk through the process of creating a simple marketing dashboard from scratch using natural language prompts.

Step 1: Create a New Report and Select a Data Source

Start by opening Looker Studio and creating a new blank report.

1. Navigate to lookerstudio.google.com.
2. Click the Blank Report button on the top left.
3. You'll immediately be prompted to add data to the report. Under Google Connectors, select Google Analytics.
4. Authorize your account if you haven't already, then select your GA4 account and property from the list. Click Add.

Once you confirm, Looker Studio will add the data source to your report, and you'll be dropped into the report editor with a blank page. You might see a default table auto-generated for you — feel free to delete it to start fresh.

Step 2: Find and Use the Duet AI Prompt Box

With your data source connected, you can now start using Duet AI. You'll find it in Looker Studio’s toolbar.

Look for an option called Insert in the menu, or just look for a chart type. When Duet AI is available for chart generation (it's gradually rolling out), you'll see an option like "Create chart with Duet AI" or a prompt-friendly interface when adding a new chart. Often, the easiest way to access it is to use the Chart dropdown from the top toolbar.

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 3: Write Your First Prompt to Generate a Chart

Let's create our first visualization. Imagine we want to see where our website visitors are coming from. A simple prompt for this would be:

Show total users by country as a map.

Type this into the prompt box and hit enter. Duet AI will analyze the request, identify "Total Users" as the metric and "Country" as the dimension, and generate a geographical map chart. This chart will automatically appear on your report canvas.

Let’s try another one. This time, we want to see our traffic trend over time.

Create a line chart of website sessions over the last 90 days.

Again, Duet AI gets to work. It knows a line chart is best for showing trends and that you need "Date" on the x-axis and "Sessions" on the y-axis. The result is a clean time-series chart showing your session data for the specified period.

Step 4: Generate a Mix of Complementary Visualizations

A good dashboard tells a story by providing different views of your data. Let's add a few more charts to round out our report.

• For user devices: what percentage of our users are on mobile vs desktop? make it a pie chart
• For top-performing pages: Top 10 landing pages by number of sessions as a bar chart
• For key conversions: Show me a scorecard of total conversions this month

As you generate each chart, you can click and drag it to the desired position on the report canvas. You don't have to get it perfect the first time, the goal is to get all the necessary components onto the page first.

Step 5: Refine and Arrange Your Dashboard

Once you have several charts on your canvas, it's time to organize them into a clear, readable dashboard.

• Reposition and Resize: Click and drag the borders of each chart to resize it. Line them up neatly to create a professional-looking layout. Place your most important KPI scorecards at the top, followed by trends and breakdowns.
• Add a Title and Date Control: Go to Insert > Text to add a title like "Monthly Marketing Performance." Then, go to Add a control > Date range control and place it at the top. This allows viewers to filter the entire dashboard by a specific time frame.
• Review and Edit: Sometimes the AI won't get it perfectly. Click on a chart you generated, and look at the right-hand panel. You can manually change the dimensions, metrics, or sorting if needed. The AI does the initial setup, but you still have full control to fine-tune it.

Following these steps, you’ve just built a marketing dashboard in minutes, a process that used to be a frustrating exercise in clicking through menus and settings.

Tips for Writing Better AI Prompts

Prompting is the new skill. While you can use "lizard brain" prompts with just a few words, being more specific will get you better results, faster. Here are a few tips:

• Specify the Chart Type: Use words like "bar chart," "line chart," "pie chart," "map," or "scorecard" in your prompt. If you don't, the AI will guess, but it might not be what you want.
• Clearly State Metrics and Dimensions: A metric is something you can measure (e.g., Revenue, Sessions, Clicks), and a dimension is what you measure it by (e.g., Country, Campaign, Device). Including both improves accuracy. Example: "show CTR (metric) by Campaign (dimension)" is better than "show me campaign performance."
• Include a Timeframe: Unless you want the default, specify a time range like "last 30 days," "this quarter," "in April 2024," or "year to date."
• Keep It Simple at First: Start by asking for one chart at a time. Trying to build a complex, multi-chart dashboard with a single, massive prompt often leads to confusing results. Generate charts one by one, then arrange them yourself.
• Use Follow-Up Prompts: If the AI gives you a chart that's almost right, you can often refine it with another prompt. For instance, after getting a chart of users by country, you could say, "now only show the top 10" or "change this to a horizontal bar chart."

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

The Hurdles of Building Dashboards with AI

Using AI in Looker Studio is a huge leap forward, but it's not a silver bullet. The process still relies on your tool's ability to connect your data sources easily and correctly, which can be a pain point for platforms outside the Google ecosystem. Connecting data from platforms like Shopify, Salesforce, or Facebook Ads often requires third-party connectors that can be slow, clunky, or hit API rate limits.

Furthermore, while the AI helps you get started, you're the one who needs to piece together the insights. It can show you a chart about Facebook Ads performance and another one about Shopify sales, but connecting the dots between your ad spend and your actual revenue still requires you to manually analyze the data across reports. The AI acts as an order taker, generating the chart you ask for, but it doesn't always provide the strategic insight or recommendations you need to make decisions.

Final Thoughts

Leveraging AI tools like Duet in Looker Studio lets you build dashboards faster than ever by skipping the manual setup and focusing directly on your data. This makes business intelligence more accessible for everyone, especially those on marketing and sales teams who need quick answers without spending hours learning a complex tool.

At Graphed, we aim to make this process even simpler. Instead of wrestling with connectors and APIs, we centralize your data from sources like Shopify, Google Analytics, Facebook Ads, and Salesforce for you. From there, you can use natural language not just to build live dashboards but to have a conversation with your data, ask follow-up questions, and get AI-powered insights and recommendations. We help you skip reporting busywork and get straight to the answers that grow your business.