How to Create a Financial Dashboard in Looker with AI
Creating a financial dashboard from scratch can feel like a major project, but new generative AI features in Looker are making it faster and more accessible than ever. Instead of spending hours clicking through menus to configure charts, you can now use plain English to generate insights. This guide will walk you through how to use Looker's AI to build a real-time financial dashboard, step-by-step.
First, Why Build a Financial Dashboard in Looker?
While spreadsheets can track numbers, a dynamic dashboard turns raw data into a decision-making tool. Looker sits on top of your live database, which means your reports are always up-to-date, eliminating the tedious and error-prone process of manually exporting and wrangling CSV files every week.
The core benefit of Looker is its centralized data modeling layer, Looker Modeling Language (LookML). LookML allows you to define all your business metrics and logic in one place. Your company's specific definition of "Monthly Recurring Revenue" or "Gross Profit Margin" is coded once, ensuring everyone in the organization is working from the same source of truth. This consistency is critical for financial reporting, and it's the foundation that makes Looker’s new AI features so powerful and accurate.
Laying the Groundwork: What You'll Need
Before you start building, you need to have a few key components in place. Proper preparation is what separates a helpful, accurate dashboard from a confusing one.
Connecting Your Data Sources
Your financial data probably lives in several different places. You'll need to consolidate it into a central data warehouse that Looker can connect to, such as Google BigQuery, Snowflake, or Amazon Redshift. Common financial data sources include:
Accounting Software: QuickBooks, Xero, NetSuite
Payment Platforms: Stripe, PayPal, Square
Enterprise Resource Planning (ERP) Systems: SAP, Oracle
Subscription Management: Chargebee, Recurly
CRMs: Salesforce, HubSpot (for tracking pipeline and sales data)
Defining Your Key Performance Indicators (KPIs)
A good dashboard tells a story, and your KPIs are the main characters. Don't try to track everything, focus on the handful of metrics that truly reflect the health of your business. Meet with stakeholders from finance, sales, and leadership to define what matters most.
Essential financial KPIs often include:
Revenue Metrics: Monthly Recurring Revenue (MRR), Annual Recurring Revenue (ARR), Revenue Growth Rate.
Profitability Metrics: Gross Profit Margin, Net Profit Margin, Earnings Before Interest, Taxes, Depreciation, and Amortization (EBITDA).
Expense Metrics: Operating Expenses (OpEx), Customer Acquisition Cost (CAC), Burn Rate.
Cash Flow Metrics: Operating Cash Flow (OCF), Free Cash Flow (FCF).
Company Health Metrics: Cash Runway, Customer Lifetime Value (LTV), LTV:CAC Ratio.
Access and a Defined LookML Model
To follow along, you’ll need authoring access in Looker and a connection to your data warehouse. Most importantly, a data engineer or analyst must have already created a LookML model that defines your dimensions (like "Date," "Department," "Product Line") and measures (like "Total Revenue," "Transaction Count," "Average Deal Size").
This LookML model is what makes Looker’s AI so intelligent. It provides the AI with the necessary context about your business, ensuring that when you ask for "sales," it knows exactly which metric you mean.
A Step-by-Step Guide to Building Your Financial Dashboard
Once your data is connected and your LookML model is ready, you can start building. We'll leverage Looker Studio, which now deeply integrates with Looker and houses its generative AI features, powered by Duet AI.
Step 1: Create a Looker Studio Report Connected to Your Looker Data
Looker and Looker Studio work together. Think of Looker as the engine for data modeling and governance, and Looker Studio as the canvas where you visualize it all. First, you'll need to create a blank Looker Studio report and connect it to your Looker instance as a data source.
Open Looker Studio and start a new, blank report.
In the "Add data to report" panel, find and select the "Looker" connector.
Authorize the connection and navigate to the Looker instance you want to use.
Select the specific Looker Explore that contains your pre-defined financial data and click "Add."
Now you have a canvas to build on, with Looker Studio understanding the structure of your business as defined in your LookML model.
Step 2: Use Natural Language to Generate Charts
This is where the magic happens. Instead of manually dragging and dropping dimensions and metrics, you can simply describe the chart you want to create.
On your blank dashboard canvas, locate the generative AI feature (often indicated by a sparkle icon and a prompt box that says "Create a chart by describing it"). Click on it and start typing. Here are a few examples to get you started:
Generate a Revenue Trendline
Start with a high-level view of your top-line revenue.
Looker Studio's AI will parse your request, identify the "MRR" measure and the "Date" dimension from your LookML model, apply the 12-month filter, and create a perfectly structured line chart for you. No clicks needed.
Create an Expense Breakdown
Next, understand where the money is going. This helps you track budget vs. actuals.
The AI will generate a bar chart showing OpEx broken down by finance, marketing, sales, engineering, and so on, depending on how "Department" is defined in your data.
Visualize Profitability
Add a KPI card to highlight your most important profitability metric.
The AI will create a large, clear KPI widget displaying the single number that shows your current quarterly gross profit margin.
Build a Combo Chart
More complex visualizations are also possible. For instance, you can analyze the relationship between what you spend and what you earn.
This command creates a single, insightful chart that helps you see the correlation between marketing spend and incoming revenue, all without navigating complex chart-building menus.
Step 3: Ask Instant Follow-up Questions
The real power of conversational AI is its iterative nature. The first chart often leads to more questions. Instead of building a new chart from scratch, you can modify the existing one with simple follow-up prompts.
Click on the MRR line chart you created and ask:
"Now show it weekly."
"Change this to a smoothed line chart."
"Break this down by new customer revenue vs. expansion revenue."
Each time, the AI will revise the existing chart instantly. This conversational flow turns data analysis into a fast-paced conversation, allowing you to dig deeper and uncover insights much faster than traditional methods.
Step 4: Arrange and Finalize Your Dashboard
Once you have generated your essential charts, it's time to arrange them into a cohesive dashboard. A logical layout makes the information easy to digest at a glance.
Place Key Metrics Up Top: Put your most important scorecards, like overall revenue and profit margin, at the top for immediate visibility.
Group Related Information: Keep revenue charts close to profit visualizations, and expense charts near cash flow metrics.
Add Interactivity: Use Looker Studio's date range filters and dropdown controls so viewers can slice the data themselves without needing to ask for a new report.
Style and Branding: Adjust colors, fonts, and chart styles to match your company's branding for a professional look and feel.
Best Practices for Effective Financial Dashboards
Just because you can build a complex dashboard doesn't mean you should. Simplicity and clarity are your biggest allies.
Know Your Audience: An executive dashboard should be high-level with top KPIs, while a financial analyst's dashboard might include more granular details and breakdowns. Create different dashboards for different stakeholders.
Tell a Coherent Story: Your dashboard should guide the viewer's eyes from top-level summaries (the "what") to detailed charts (the "why"). For example, an unexpected drop in a revenue trendline should be physically close to a table breaking down revenue by source, making it easy to identify the cause.
Focus on a Single Goal: Each dashboard should answer a primary question, whether it's "How healthy is our business right now?" or "Are we spending our budget effectively?" Avoid jamming unrelated metrics onto one canvas.
Final Thoughts
Building a fully functional financial dashboard in Looker used to be a task reserved for those with deep technical knowledge. With the integration of generative AI, the process is now driven by natural language, making powerful financial analytics accessible to a much broader audience. You can move directly from a business question to a data-backed answer in seconds.
While tools like Looker have made strides, we created Graphed because we believe there's an even easier way. For many teams, the setup cost of data warehouses and LookML modeling is still prohibitive. With our tool, you connect your QuickBooks, Stripe, Salesforce, and other accounts directly, and then use simple, conversational language to build dashboards and get insights instantly - no data modeling or BI software expertise needed. Our platform handles the entire process, empowering you to create the same real-time financial dashboards without a dedicated data team.