How to Create an E-commerce Dashboard in Looker with AI

Creating an e-commerce dashboard that pulls all your data into one place can feel like a game-changer. Suddenly, you're not juggling tabs between Shopify, Google Analytics, and your ad platforms just to see what's working. This guide will walk you through building a comprehensive e-commerce dashboard in Looker, using its AI features to make the process faster and more intuitive.

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 Centralize Your E-commerce Data in the First Place?

Most e-commerce businesses operate across multiple platforms. Your sales happen in Shopify, traffic data lives in Google Analytics 4, ad spend is tracked in Facebook Ads Manager, and customer relationships are managed in a CRM like Klaviyo or HubSpot. Looking at each platform individually only gives you a tiny piece of the story.

When you pipe all that data into a single dashboard, you unlock a complete view of your business performance. You’re no longer guessing, you’re seeing the full customer journey, from ad click to final purchase and beyond. Centralizing your data allows you to:

• Get a Holistic View: See how ad spend on Facebook influences traffic in Google Analytics and ultimately drives sales in Shopify.
• Calculate True ROI: Accurately attribute revenue back to a specific marketing channel or campaign instead of relying on each platform's biased reporting.
• Make Faster, Smarter Decisions: When you see that a specific campaign’s traffic isn’t converting, you can act on it immediately, rather than waiting for manual reports at the end of the week.
• Uncover Deeper Insights: Identify trends you'd miss by looking at isolated datasets, such as which geographic regions have the highest Average Order Value (AOV) but the lowest ad spend.

Step 1: Get Your Data Ready for Looker

Before you can build anything, you need to bring your data sources together in a place Looker can access. Unlike some tools that have direct, one-click integrations, Looker is a traditional BI tool that typically sits on top of a data warehouse. This setup process is often the most challenging part for marketing and e-commerce teams without dedicated data engineers.

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

Connect Your E-commerce Platforms

First, identify all the places your data lives. For a typical e-commerce business, this includes:

• Store platform: Shopify, BigCommerce, Magento, etc.
• Web analytics: Google Analytics 4
• Advertising platforms: Google Ads, Facebook Ads, TikTok Ads
• Email marketing / CRM: Klaviyo, HubSpot, Mailchimp
• Payment processors: Stripe, PayPal

Choose a Data Warehouse

A data warehouse is essentially a central, high-powered database designed to store large amounts of data from different sources. Think of it as the single source of truth for your business. Common choices include Google BigQuery, Snowflake, and Amazon Redshift. They are all powerful but require technical knowledge to set up and manage.

Pipe Your Data into the Warehouse

Once you have a warehouse, you need a way to get your data from your platforms (like Shopify and Facebook Ads) into it. This is handled by a process called ETL (Extract, Transform, Load). You’ll likely need a third-party tool like Fivetran, Stitch, or Airbyte to create these data pipelines. These tools automatically pull data from your connected sources and load it into your data warehouse on a set schedule.

This initial setup is a significant hurdle and often requires help from a data engineer or IT specialist. Getting this foundation right is critical, as everything else depends on it.

Step 2: Building Your Dashboard in Looker with AI Features

With your data flowing into a warehouse, you can finally connect Looker and begin building. Looker has a notoriously steep learning curve, primarily because of its powerful modeling layer, LookML. However, its new AI features are designed to lower this barrier and speed up the development process for everyone.

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

Working with LookML: The Foundation

Before AI can help, you need to define your business logic in LookML. LookML is Looker’s specialized language for describing your data. It’s where you tell Looker what your data means and how tables relate to each other.

For example, you’ll create “dimensions” (attributes to group by, like 'Campaign Name' or 'Country') and “measures” (metrics to calculate, like 'Total Sales' or 'Average Order Value'). You'd define total_sales as the sum of the price column from your Shopify orders table.

This step ensures that everyone in the company is using the same definitions for key metrics, eliminating confusion. It's a powerful feature, but it’s code-based and requires time to learn.

Using Generative AI in Looker

Once your LookML model is set up, you can start leveraging Looker's AI capabilities. This is where the magic happens for non-technical users. Instead of navigating complex menu systems or writing code to build visualizations, you can use natural language.

You can simply type a question into the search bar, like:

Show me total revenue by marketing channel for the last 90 days as a bar chart

Looker's AI will interpret your request, query the data warehouse based on your LookML definitions, and automatically generate the bar chart for you. From there, you can ask follow-up questions to drill down further:

Okay, now filter for only the 'USA' and break it down by week

This conversational approach radically transforms the dashboard creation process. It turns an hours-long task of clicking, dragging, and formatting into a quick, intuitive conversation. The AI can also suggest relevant questions you might not have thought to ask, helping you explore your data more deeply.

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

Key E-commerce KPIs to Track in Your Dashboard

A great dashboard answers your most important business questions at a glance. Here are some essential KPIs every e-commerce dashboard should include. Start with these, and build them out using Looker’s AI prompts.

Overall Business Health

• Total Revenue: The top-line number for all sales.
• Total Orders: The volume of transactions.
• Conversion Rate: The percentage of website visitors who make a purchase. This is a critical indicator of how well your site and marketing are performing. (Prompt: "What was my overall conversion rate last month?")
• Average Order Value (AOV): The average amount customers spend per transaction. (Prompt: "Show me AOV trending over the last 6 months.")

Marketing & Customer Acquisition

• Customer Acquisition Cost (CAC): The total cost of sales and marketing to acquire a new customer. (Formula: Total Ad Spend / New Customers Acquired).
• Return on Ad Spend (ROAS): The revenue generated for every dollar spent on advertising. This is crucial for evaluating campaign profitability. (Prompt: "Compare ROAS for my Google Ads vs. Facebook Ads campaigns.")
• Revenue by Marketing Channel: A breakdown of where your sales are coming from (e.g., Organic Search, Paid Social, Email).

Customer Value & Retention

• Customer Lifetime Value (LTV): The total revenue you can expect from a single customer over their lifetime. A healthy business has an LTV that is significantly higher than its CAC.
• Returning vs. New Customers: A ratio showing how many of your sales come from repeat buyers versus first-time purchasers.

Tips for a Dashboard that Actually Gets Used

A dashboard is only valuable if it leads to action. Here’s how to make sure yours doesn't just collect dust.

• Start with Questions, Not Data: Before building anything, define the top 5-10 questions you need to answer every day or week. Your dashboard should be built to answer those questions.
• Keep it Simple: A cluttered dashboard is an ignored dashboard. Use plenty of white space and focus on a few key metrics per section. A simple bar chart is often much clearer than a complex scatter plot.
• Tailor to Your Audience: A dashboard for a CEO should show high-level metrics like overall revenue and profitability. A dashboard for a paid ads manager needs granular details like ROAS by campaign and CPC.
• Make it Interactive: Use filters to allow users to slice and dice the data themselves. Adding filters for date ranges, marketing channels, and countries empowers your team to explore the data and find their own answers.

Final Thoughts

Creating a centralized e-commerce dashboard in Looker can provide incredible clarity for your business. The process involves a significant technical lift upfront - setting up a data warehouse and defining your LookML model - but once it's done, Looker's AI features can make data exploration fast and accessible for your whole team.

Of course, wrestling with data pipelines, warehouses, and data modeling isn't for everyone, especially for busy marketing and e-commerce teams. We built Graphed to deliver the same powerful insights without any of the complex setup. You just connect your data sources like Shopify, Google Analytics, and Facebook Ads in a few clicks, and our AI does the rest. Instead of learning LookML, you can ask questions in plain English and get real-time, shareable dashboards built for you in seconds.