How to Create a Project Dashboard in Looker with AI

Juggling project tasks, timelines, stakeholders, and budgets can feel like a masterclass in controlled chaos. Creating a project dashboard in Looker helps you turn that chaos into clarity by providing you with a real-time, single source of truth about how everything is progressing. This guide will walk you through building a project management dashboard in Looker step-by-step and show you how to leverage its AI-assisted features to get it done faster.

Graphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Build Your First Dashboard

Why Build a Project Dashboard in the First Place?

Before jumping into the "how," let's quickly cover the "why." A well-built project dashboard isn't just a collection of charts, it's a command center for your entire project. It centralizes scattered data from your various tools (like Asana, Jira, or spreadsheets) and helps you see the bigger picture at a glance. The main benefits include:

• Centralized Hub: No more jumping between tabs or bugging team members for status updates. Everything you need is in one place.
• Real-Time Tracking: Instantly see which tasks are on track, which are falling behind, and where your budget stands right now, not last Tuesday.
• Proactive Problem-Solving: Spot potential bottlenecks or resource constraints before they turn into full-blown crises.
• Clear Stakeholder Communication: Easily share progress with leadership or clients, providing transparency and building confidence.

Ultimately, a good dashboard gives you back time and mental energy, allowing you to focus on strategy and guidance instead of data collection.

Step 1: Define Your Goals and Key Metrics (What Story Do You Want to Tell?)

The most common mistake people make is diving straight into building without a clear plan. A successful dashboard tells a story. Before you drag and drop a single chart, ask yourself: What questions do I need this dashboard to answer?

Grab a pen and paper or open a notepad and list the essential metrics you need to track. Think in terms of project health, team performance, and financial status.

Here are a few common project management metrics to get you started:

• Project Progress:
• Team & Resource Management:
• Financial Tracking:

Don't try to track everything. Pick the 5-7 most important metrics that align with your project's goals. This initial brainstorming session will be your blueprint for the entire build process.

AI for Data Analysis

Crash Course

Graphed

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

Download Free Guide

Step 2: Connect Your Project Management Data to Looker

Looker gets its power from sitting on top of a clean, structured database. This means you first need to get your project data into a data warehouse that Looker can connect to, like Google BigQuery, Snowflake, or Amazon Redshift.

Your data likely lives in tools like:

• Jira
• Asana
• Trello
• Monday.com
• Airtable
• Or even a detailed Google Sheet or Excel workbook.

Getting this data into your warehouse is often the most technical part of the process, and you might need support from your data team to set up the data pipeline. Once the data is connected, it’s modeled using LookML (Looker’s modeling language), which defines your metrics and business logic. While LookML gives developers incredible control, it can be a hurdle for less technical users. Fortunately, this is where newer AI features can start to assist.

Step 3: Creating Your First "Look" with AI Assistance

In Looker, a single chart or data table is called a "Look." You create multiple looks and then assemble them on a dashboard. You can begin exploring your data from an "Explore," which is the user-friendly interface for querying your LookML model.

Traditionally, this required you to know which dimensions (fields to group by, like "Assignee" or "Due Date") and measures (calculations to perform, like "Count of Tasks") to select from the sidebar. However, with the integration of Duet AI in Google Cloud, you can now start this process with conversational language.

Using Natural Language in the Explore Interface

In the Explore section sidebar, you may see an option to ask a question. This allows you to describe what you want to see in simple English.

For example, you could type a prompt like:

show me the number of completed tasks by month for this year

The AI assistant will translate your request into a query, selecting the appropriate fields and filters for you. It might select the "Tasks" count measure, group it by the "Completion Date" dimension (set to monthly), and add a filter for the current year. It may not get it perfectly right every time, but it’s an incredible accelerator that helps you get 80% of the way there in seconds. From there, you can fine-tune the visualization, change the chart type to a line or bar chart, and save it as a "Look" for your dashboard.

Graphed

Your AI Data Analyst to Create Live Dashboards

Connect your data sources and let AI build beautiful, real-time dashboards for you in seconds.

Build Your First Dashboard

Step 4: Building Your Project Dashboard Tile by Tile

With your first Look saved, you're ready to start assembling your dashboard. A Looker dashboard is essentially a collection of "tiles," where each tile can be a Look you created, a stylized text box, or another element. Go to the dashboard creator ("New Dashboard") and start adding your tiles.

Here are some essential tiles to include for a comprehensive project overview:

1. At-a-Glance KPIs

Start your dashboard with several "single value" visualizations at the top. These act as your headline numbers, giving anyone who looks at the dashboard an instant summary of project health.

• Total Open Tasks: A simple count of all incomplete tasks.
• Overdue Tasks: The count of tasks whose due dates are in the past. Highlight this one in red!
• Budget Spent (%): A calculation showing the percentage of the budget that has been used so far.

2. Task Status Breakdown (Pie or Donut Chart)

This tile visualizes the current state of all tasks. It quickly answers the question, "Where does the work stand?" A pie or donut chart is perfect for this, as it intuitively shows the proportion of tasks in each status ("To Do," "In Progress," "Blocked," "Completed").

Natural Language Prompt Idea: Count of tasks grouped by status for Project Alpha

3. Team Workload (Bar Chart)

Avoid team burnout by visualizing how work is distributed across team members. A simple bar chart showing the number of open tasks assigned to each person is incredibly effective. It helps project managers rebalance a lopsided workload.

Natural Language Prompt Idea: show me a list of assignees and their count of open tasks

4. Budget Burn-Down (Line or Area Chart)

Are you spending money too fast? Too slow? A burn-down chart maps your actual cumulative spend over time against your planned budget. This visualization is critical for financial forecasting and making sure you don't run out of funds unexpectedly.

Natural Language Prompt Idea: Plot cumulative project spend per day versus our total budget this quarter

5. Upcoming Deadlines (Table)

A simple, actionable table showing tasks due in the next 7 or 14 days keeps the team focused on what's immediately ahead. Include columns like Task Name, Assignee, and Due Date, and sort it with the nearest deadline at the top.

AI for Data Analysis

Crash Course

Graphed

Free PDF Guide

AI for Data Analysis Crash Course

Learn how to get AI to do data analysis for you — the best tools, prompts, and workflows to go from raw data to insights without writing a single line of code.

Download Free Guide

Step 5: Filtering and Fine-Tuning for Maximum Impact

A static dashboard is useful, but an interactive dashboard is powerful. Looker's filters allow you and your stakeholders to slice and dice the data to answer their specific questions. You can add dashboard-level filters that apply to all tiles simultaneously.

Recommended filters for a project dashboard include:

• Project Name: If your dashboard covers multiple projects.
• Date Range: To zoom in on performance during a specific week, month, or quarter.
• Team Member: To see a specific individual's workload and progress.

Once your filters are set up, take a moment to refine the layout. Arrange your tiles to tell a logical story, starting with the high-level KPIs at the top and flowing down to more detailed charts and tables. Use text tiles to add headings, explanations, or key takeaways directly on the dashboard so everyone knows exactly what they’re looking at.

Final Thoughts

Creating a high-quality project dashboard in Looker turns complex data into a clear, actionable management tool. By strategically planning your metrics, building insightful visualizations tile by tile, and making it interactive with filters, you can build a single source of truth that keeps your entire team aligned and on track. Leveraging modern AI-assisted features can speed up the process, making it more accessible to those who aren’t already Looker experts.

While Looker is a powerfully robust tool, we know its setup process - from configuring data pipelines to learning LookML - can feel intimidating for teams who don't have dedicated data engineers. For marketing, sales, and operations teams who need answers now, all that complexity gets in the way. We built Graphed to solve exactly that problem. You can connect your project management tools, ad platforms, and CRM in just a few clicks, and then build entire real-time dashboards simply by describing what you want in plain English. Instead of just helping you start a query, Graphed's AI builds the whole report for you in seconds, letting you go from question to insight without waiting for anyone.