How to Create a Sales Report in Looker with AI

Building a rich, insightful sales report in Looker can feel like a heavy lift, often requiring you to be well-versed in LookML or have a data team on standby. But by using Looker's AI capabilities, you can skip the complex parts and create powerful sales dashboards just by asking questions in plain English. This article will show you how to prepare your data and use natural language to build a detailed sales report, step by step.

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

Why Use AI for Sales Reporting in Looker?

Traditionally, creating a new report in a powerful BI tool like Looker involves a manual, multi-step process. You have to select data sources, choose dimensions and measures, configure filters, and pick the right visualization type - all done through a series of clicks and drags. If you're not a Looker expert, this can be slow and intimidating.

AI changes the game entirely. Instead of navigating menus, you simply describe the report you want. This approach offers a few huge advantages:

• Speed: What once took minutes of clicking can now be done in seconds with a single sentence. You can generate charts and reports almost as fast as you can think of the questions you want to ask.
• Accessibility: You no longer need to be a data analyst or learn LookML to get answers from your sales data. Anyone on the sales team, from a rep to the VP of Sales, can ask questions and build their own reports without technical assistance.
• Deeper Insights: The conversational nature of AI encourages exploration. When a chart sparks a new question, you can immediately ask a follow-up, drilling down into specifics without losing your train of thought or building a new report from scratch.

Getting Your Data Ready for AI Reporting

The quality of your AI-generated report depends entirely on the quality of the data it's drawing from. Before you start asking questions, make sure your foundation is solid. The "garbage in, garbage out" rule definitely applies here.

Step 1: Connect Your Core Sales Data Sources

First things first, your sales data needs to be in Looker. This typically means connecting your CRM or the database where your deals live. Common sources for sales reporting include:

• Customer Relationship Management (CRM) platforms like Salesforce or HubSpot.
• Databases like BigQuery, Redshift, or Snowflake where your sales data is stored.
• Spreadsheets (like Google Sheets) for smaller datasets or supplementary information.

The connection process is done within Looker's admin settings and is the necessary first step before any analysis can begin.

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 2: Understand Your Data Model and Key Metrics

Looker's AI assistant is smart, but it's not a mind reader. It relies on the underlying data model (the LookML project) that defines your company's data. Having a basic understanding of what fields and metrics are available will help you ask better questions.

Before you dive in, familiarize yourself with the names of key sales metrics in your Looker instance, such as:

• Dimensions (the "what"): Sales Rep Name, Deal Name, Account, Region, Product Line, Close Date, Lead Source.
• Measures (the "how much"): Deal Amount, Annual Recurring Revenue (ARR), Pipeline Amount, Number of Deals Won.

You don't need to know the technical details, but just having a general idea of these field names will make your prompts much more effective.

Step-by-Step: Creating a Sales Report with Looker's AI Features

With your data connected and a basic grasp of your metrics, you're ready to start building. The core workflow involves asking a question, refining the result, and asking follow-up questions to dig deeper.

1. Start a Conversation with Your Data

Find the natural language or "Ask" feature within your Looker instance. This is typically located within an Explore or on an existing dashboard. This is your command center for creating reports conversationally.

Begin with a clear, straightforward question about your sales performance. Think about a high-level KPI you'd want to see on a dashboard.

Example Prompts for High-Level KPIs:

• Show me total sales revenue for this quarter
• What is our win rate for the last 6 months?
• Create a bar chart of pipeline value by sales stage

Looker's AI will interpret your request, select the appropriate dimensions and measures, and instantly generate a visualization and a data table.

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

2. Fine-Tune Your Visualization with Follow-up Prompts

The first chart the AI creates is just a starting point. Often, you'll want to modify it to better suit your needs. You can do this by continuing the conversation instead of using the manual settings.

Let's say you started by asking, Show me sales revenue by month this year, and it produced a table. You can easily pivot from there.

Example Refining Prompts:

• Change the chart type: Change this to a line chart
• Break it down further: Now break this down by sales region
• Filter the data: Only show me deals from the Enterprise segment or Remove deals from Q1
• Add another metric: Add average deal size to the chart

Each command builds upon the previous one, allowing for rapid iteration until the chart shows exactly what you need.

3. Drill Down to Uncover Deeper Insights

This is where the true power of conversational analytics shines. Once a visualization sparks your curiosity, you can dig deeper without breaking your flow. This turns reporting from a static task into a dynamic investigation.

Imagine your chart of sales revenue by region shows that "North America" is outperforming everyone else. You can immediately investigate why:

• Initial Prompt: Show me our total sales by region as a bar chart this year.
• Follow-up 1: For North America, which sales reps are closing the most revenue?
• Follow-up 2: Let's focus on the top rep, Jane Doe. What were her sales by product line last quarter?
• Follow-up 3: Ok, change that to a pie chart showing which lead sources are driving her deals.

In just a few simple questions, you've moved from a high-level overview to a granular insight about your best-performing channels for your top rep in your strongest region. Attempting this same analysis manually would have involved creating several different reports and applying multiple filters.

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

4. Building an Entire Dashboard with AI-Generated Tiles

Don't stop at just one chart. You can use this AI-powered process to build an entire sales dashboard tile by tile. Simply create each visualization with a natural language prompt, and then pin it to a new or existing dashboard.

A good sales dashboard tells a story. Here’s a sample flow for building one:

1. KPI Overview: Start with prompts for your main metrics.
2. Team Performance: Build a leaderboard to track rep performance.
3. Pipeline Health: Visualize the future state of your sales.
4. Geographic Breakdown: See where your revenue is coming from.

After a few minutes of conversation, you can have a comprehensive, multi-faceted sales dashboard ready to share with your team, all without getting bogged down in the technical minutiae of Looker's editor.

Best Practices for Getting the Most Out of AI in Looker

To make your interactions with Looker's AI productive and accurate, keep a few best practices in mind:

• Be Clear and Specific: Vague prompts lead to vague outcomes. Instead of "Show sales," ask "What was our total closed-won deal value in the APAC region last month?" The more detail you provide, the better the result will be.
• Start Small, then Iterate: Don't try to get the perfect multi-layered chart with your first prompt. Start with a simple question to get your base data (e.g., Revenue by month). Once that looks right, add layers of complexity with follow-up questions (e.g., Now break that down by product or Only show results from the West region).
• Review and Verify: AI is a powerful assistant, but it's not perfect. Always take a moment to sanity-check the visualization it produces. Does the chart make sense? Did it choose the correct fields for "revenue" and "close date"? Treat it as a "trust but verify" relationship.
• Use Natural Language, but Think in Terms of Data: Frame your questions conversationally, but try to use terms that align with the fields in your data model. Using "deal value" instead of "money" or "sales rep" instead of "person" helps the AI quickly find the right data.

Final Thoughts

Using AI turns the process of creating sales reports in Looker from a technical, click-heavy chore into a fast, intuitive conversation. It empowers every member of your team to explore data, uncover insights, and make better decisions without needing a data background or waiting on specialized help.

We built Graphed because we believe the simplicity of conversational analytics shouldn't be limited to just one tool. While platforms like Looker have immensely powerful data modeling capabilities, they often still require significant technical setup. We designed Graphed to skip that complexity entirely, allowing you to connect sources like Salesforce, HubSpot, and Google Analytics in a single click and begin creating dashboards with natural language from the very first minute, with no LookML or BI expertise required.