How to Make a Chart in Google Analytics with AI
Creating charts in Google Analytics with a few simple keystrokes is now a reality. This article will show you exactly how to use GA4's built-in "Ask Analytics Intelligence" feature to generate charts and reports by simply typing questions in plain English.
What Is Google's "Ask Analytics Intelligence"?
You may have noticed the search bar at the top of your Google Analytics 4 property that says "Search or ask a question." This isn't just for finding reports, it's a powerful natural language query tool. Think of it as a direct line to your data, allowing you to bypass menus and complex report configurations to get quick answers.
This feature uses AI to understand your questions and instantly pull the relevant data. You can ask for specific metrics ("How many new users did we get last week?"), get definitions ("What is engagement rate?"), or navigate directly to a report. But its most powerful use case is generating quick visualizations and charts on the fly, which is what we’ll focus on here.
Instead of manually building an exploration report, filtering dimensions, and adding metrics, you can just ask a question and let Google’s AI do the heavy lifting for you.
How to Generate Charts with Simple Questions in GA4
Turning a question into a chart is a straightforward, three-step process. Let's walk through it with a practical example. Say you want to see which of your marketing channels drove the most users in the past month.
Step 1: Locate the Analytics Intelligence Search Bar
Log in to your Google Analytics 4 account. At the top of every page, you'll see the search bar. This is your command center for asking questions.
Step 2: Type Your Question in Plain English
Click inside the search bar and type your request. To get the best results, you should specify three things:
The chart type: "bar chart," "line chart," "pie chart," etc.
The metric: "users," "sessions," "conversions," "revenue."
The dimension: "country," "device category," "session default channel group."
The time frame: "last 30 days," "this quarter," "in June 2024."
Continuing our example, you could type:
Show me a bar chart of users by session default channel group for the last 30 days
Press enter. The AI will process your request based on the data in your GA4 property.
Step 3: Analyze the Generated Chart
Within seconds, Analytics Intelligence will present a dropdown with its answer. In this case, it will display a bar chart visualizing the number of users from each channel group (like Organic Search, Direct, Paid Search, etc.) over the last 30 days. You can hover over each bar to see the exact numbers.
Step 4 (Optional): Add to a Custom Report
If you find the chart useful and want to refer to it later, you don't have to repeat the question every time. At the top right of the chart pop-up, you'll see a small icon with a plus sign. Clicking this allows you to add the visualization as a card to one of your custom dashboards or reports within GA4, making it easy to monitor that metric regularly.
Tips for Writing Effective AI Prompts in GA4
While the AI is quite intelligent, its output is only as good as your input. Here are a few tips to help you write prompts that generate accurate and useful charts.
Be Clear and Specific
Vague questions lead to vague answers. Instead of asking "show me traffic," try "show me a line chart of daily active users over the last 28 days." Adding details like the desired chart type, a specific timeframe, and the exact metric and dimension you're interested in will give the AI clear instructions to follow.
Vague: "How are my campaigns doing?"
Specific: "Bar chart comparing conversions by session campaign for the last 90 days"
Use Google Analytics Terminology
The AI is trained on GA4's specific language. While it can often infer what you mean, using the official names for dimensions and metrics will get you the most reliable results. You don't need to memorize everything, but getting familiar with the basics helps tremendously.
Here are a few common ones:
Instead of "channels," use "Session default channel group."
Instead of "traffic sources," use "Session source / medium."
Instead of "phone users," use "device category = mobile."
Instead of "where people live," use "country" or "city."
Start Simple, Then Build Complexity
If you have a complex question, start with a simple version and refine it. For example, begin by asking for "total users last month." Once you see the result, you can ask a more targeted follow-up question in the search bar, like "compare users from mobile vs desktop last month," or "what were the top 5 countries for users?". This iterative approach helps you drill down into your data quickly.
Ask for Comparisons
The AI tool is great for comparative analysis. Use terms like "compare" or "vs" to create charts that show different segments side-by-side. This is perfect for understanding performance trends without having to build two separate reports and place them together manually.
Compare users from the United States vs Canada last 90 days.
Show me session conversion rate YTD vs last year.
Examples of AI Prompts to Make Charts in Google Analytics
Ready to try it yourself? Here are some copy-and-paste prompts you can adapt for your own GA4 property.
Audience and Demographic Charts
Show me a pie chart of my audience by country for the past year.
Bar chart of new users by City for the last 28 days.
Compare users in California vs New York for this quarter as a line chart.
Line chart of daily active users who use Android vs iOS over the past week.
Traffic Acquisition Charts
Bar chart of sessions by session default channel group for this month to date.
Compare Organic Search vs Paid Search sessions over the last 6 months.
Top 10 session source / mediums ranked by engaged sessions.
What are my top 5 session campaigns by conversions?
Behavior and Engagement Charts
Bar chart of top 10 pages by views last month.
What are the most popular landing pages for Organic Search users?
Line chart of user engagement over time for the 'Pricing Page'.
Compare average engagement time for blog articles vs landing pages.
The Limitations of GA4's AI Feature
As useful as Google's AI assistant is, it's important to understand its limitations. It's designed for rapid, on-the-fly analysis, not for building comprehensive, multi-layered dashboards.
Here are a few things it can't (yet) do:
Analyze Data Outside of Google Analytics: Its primary weakness is that it's siloed. You can’t ask it to chart your GA4 sessions against your Facebook Ads cost, Shopify sales data, or Salesforce lead conversions. True business insights often require combining data from multiple platforms, which is beyond this tool's scope.
Handle Extremely Complex Queries: While it’s smart, it can get confused by very complex, multi-conditional questions. Requests like "Show me users from Organic Search who landed on the blog, visited the pricing page, and converted, but only on weekends" are generally too advanced.
Create Full Dashboards: This feature generates single charts or data points, not fully interactive dashboards. You can add the cards to a report, but the initial creation process is one visualization at a time.
Final Thoughts
Using the "Ask Analytics Intelligence" feature in GA4 is a fantastic way to quickly visualize your website data. By asking clear, specific questions, you can generate insightful charts in seconds, saving yourself the time you would have spent navigating through menus and report settings.
While GA4's tool is great for instant analysis of your website data, a complete picture of your business performance often requires connecting it to sales and marketing data from other places. Since you can't ask GA’s AI about your Facebook Ad spend or CRM conversions, we built Graphed to do exactly that. You can connect all your data sources - Google Analytics, Shopify, Facebook Ads, Salesforce, HubSpot - and then ask questions that span across them, like "create a dashboard comparing our ad spend on Google and Facebook against sales revenue from Shopify.” This allows you to measure your true return on investment in a unified view.