How to Create a Project Management Dashboard in Looker with AI

A great project management dashboard does more than just show you charts - it gives you a single source of truth to see what's on track, what's falling behind, and where your team's efforts are going. We'll walk you through how to build one in Looker (now part of Looker Studio), leveraging its powerful features and a touch of AI to give you real-time visibility into your projects. This guide will cover everything from defining your key metrics to building the visualizations that matter.

First, Why Build a Project Management Dashboard at All?

In a busy team, information often gets scattered across different tools - your project management app, spreadsheets, timesheets, and budget trackers. A centralized dashboard brings all that disconnected data together into a clear, actionable view. Let's face it: guesswork rarely works when managing timelines and budgets.

A good dashboard empowers you to:

• Spot bottlenecks early: See where tasks are piling up or which stage of a project is causing delays.

• Monitor team workload: Understand who has too much on their plate and who has capacity, making resource allocation fair and efficient.

• Track budget vs. actuals: Keep an eye on spending in real-time to avoid costly overruns.

• Improve stakeholder communication: Replace lengthy email updates with a simple, shareable link that always shows the latest data.

Laying the Groundwork: Before You Touch Looker

The success of any dashboard depends on the prep work you do before you start building. Jumping straight into a B.I. tool without a clear plan is a recipe for a cluttered, confusing dashboard that no one uses.

Define Your Key Project Management Metrics (KPIs)

Start by asking yourself: "What questions do I need this dashboard to answer?" The answers will help you identify the right Key Performance Indicators (KPIs) to track. You don’t need to track everything, focus on the metrics that directly reflect project health and team performance.

Here are some essential PM KPIs to consider:

• Task Completion Rate: The percentage of tasks completed out of the total tasks assigned. This gives you a high-level view of progress.

• Tasks by Status: A simple breakdown of tasks in categories like "To Do," "In Progress," "Blocked," and "Completed." This is great for daily stand-ups.

• On-Time Completion Rate: The percentage of tasks or milestones completed by their deadline. This is a critical indicator of whether you'll hit your final delivery date.

• Cycle Time: The average time it takes to complete a task from start to finish. A shorter cycle time generally means a more efficient process.

• Team Workload: The number of tasks or estimated hours assigned to each team member. This helps prevent burnout and balance responsibilities.

• Budget Variance: The difference between your planned budget and your actual spending. Track this closely to keep finances on track.

• Project Timeline & Milestones: A visual representation of key project phases and their deadlines against the original plan.

Prep Your Data Source

Looker needs clean, structured data to work its magic. Your project data likely lives in tools like Jira, Asana, Monday.com, Trello, or even a detailed Google Sheet. No matter the source, your goal is to have the data organized in a way that's easy to analyze.

A good practice is to ensure your source data is in a "tidy" format, where:

• Each row represents a single task.

• Each column represents a distinct attribute of that task (e.g., Task Name, Assignee, Due Date, Status, Project, Estimated Hours, Actual Hours).

If you're using a Google Sheet, your columns might look something like this:

Spending an hour cleaning up your data source now will save you a dozen hours of frustration inside Looker later.

Building Your Dashboard in Looker: A Step-by-Step Guide

With your KPIs defined and your data prepared, it's time to build your dashboard. This process in Looker generally involves connecting your data, defining your logic with LookML, creating individual charts (called "Looks"), and arranging them into a cohesive dashboard.

Step 1: Connect Your Data to Looker

Looker supports a wide range of databases and file integrations. Log into your Looker instance and navigate to the Admin panel to set up a new connection.

1. Go to Admin > Database > Connections.

2. Click Add Connection.

3. Select your database dialect (e.g., Google BigQuery, PostgreSQL, Google Sheets).

4. Fill in the connection details. For a Google Sheet, this often just involves connecting via OAuth and selecting the correct sheet.

Step 2: Create a Project and Model Your Data with LookML

This is where the unique power (and sometimes the challenge) of Looker comes in. Looker uses a modeling layer called LookML to define dimensions (the things you group by, like "Assignee" or "Status") and measures (the things you calculate, like "Count of Tasks" or "Average Hours Logged"). LookML acts as a blueprint for your data, ensuring that everyone in your organization calculates metrics the same way.

• A Project is a collection of LookML files that describe your data. Start by creating a new LookML project and connecting it to the database connection you just made.

• A View file in LookML corresponds to a data table. You'll create a view for your tasks table, defining each column as a dimension.

• A Model file defines how views relate to each other and exposes them for exploration.

For a non-developer, LookML has a steep learning curve. However, Looker can often generate a basic LookML model from your database table, giving you a solid starting point that you can then refine.

Step 3: Create Individual Visualizations ("Looks")

Once your LookML model is set up, you can start exploring your data and building visualizations. In Looker terminology, a single chart or data table is called a "Look."

Navigate to the Explore section and select the model and view you created for your project data. You'll see a user-friendly interface where you can pick dimensions and measures from the left-hand panel, apply filters, and choose a visualization type.

Example Looks for Your PM Dashboard:

• Tasks by Status (Pie Chart): Select the "Status" dimension and the "Count" measure. Choose the pie chart visualization to get a quick visual breakdown.

• Team Workload (Bar Chart): Use the "Assignee" dimension and the "Count" measure. A bar chart is perfect for comparing task loads across team members.

• On-Time vs. Overdue Tasks (KPI Tiles): Create a measure in your LookML to calculate the percentage of on-time completions. Display this as a single number using the "Single Value" visualization. Do the same for overdue tasks.

• Timeline/Gantt View (Table with Bars): Use a table visualization and configure it to show a timeline. Use dimensions like "Task Name," "Start Date," and "End Date." Some marketplace visualizations can create more advanced Gantt charts.

For each visualization you create, click Save and select "As a Look." Give it a descriptive name like "Task Status Breakdown."

Step 4: Arrange Your Looks into a Dashboard

Now it's time to bring all your individual Looks together.

1. Navigate to the folder where you saved your Looks and click New > Dashboard.

2. Give your dashboard a name like "Project Health Dashboard - Q4."

3. Start adding the Looks you created by clicking Add Tile or dragging and dropping them from the sidebar.

4. Arrange the tiles logically. Put high-level KPIs at the top, followed by more detailed charts and tables below.

5. Add filters! One of Looker's best features is the ability to add dashboard-wide filters. Add filters for "Project," "Assignee," and "Date Range" so you and your team can slice the data without having to edit each chart individually.

Supercharging Your Looker Dashboard with AI

Building the dashboard is a huge step, but the real value comes from deriving insights. Looker integrates AI and machine learning features to help you go from data to decisions faster.

Asking Questions with Natural Language

Google has been integrating its Gemini AI model into Looker, allowing users to ask questions in plain English instead of manually building reports. This functionality, often called "Explore with Conversational Analytics," enables you to type a question like, "What was the total number of tasks completed last week, grouped by assignee?" Looker's AI will translate that into a query and generate the visualization for you. This functionality greatly lowers the barrier to entry, letting less technical stakeholders find their own answers.

Automated Anomaly Detection and Alerts

You don't have to stare at your dashboard all day to catch problems. AI-powered features can monitor your data for you. You can set up alerts to get notified automatically if a metric crosses a certain threshold or behaves unusually. For example, Looker can alert you if:

• The number of "Blocked" tasks suddenly spikes.

• A project's estimated completion date, based on the current pace, moves past the deadline.

• The budget spend for a project accelerates unexpectedly.

AI for Forecasting and Predictions

Many data platforms are beginning to integrate forecasting capabilities. Within Looker, you can create fields that use forecasting models to predict future performance. For project management, this could be used to forecast the project's final cost based on the current burn rate or predict the likely completion date based on your team's velocity over the last few weeks. This moves your dashboard from being purely reactive (what happened) to proactive (what is likely to happen).

Final Thoughts

Building a project management dashboard in Looker provides a powerful, single source of truth that aligns your team, simplifies stakeholder reporting, and helps you proactively manage risks. By starting with clear goals, clean data, and a thoughtful layout, you can create a tool that drives efficiency and keeps your projects humming along.

If dealing with the complexities of LookML and the deep setup of traditional business intelligence platforms feels daunting, you're not alone. We created Graphed to cut through that complexity. Instead of modeling data structures and manually building charts, you can connect your sources (like Asana, Jira, and your spreadsheets) in a few clicks, then just describe the dashboard you want in plain English. Graphed uses AI to build interactive, real-time dashboards for you in seconds, saving you from the hours-long learning curve and letting you get straight to the insights you need to keep your projects on track.