How to Create a Project Portfolio Dashboard in Looker with AI

Trying to keep track of a dozen projects at once feels like juggling chainsaws. Between shifting deadlines, budget approvals, and resource conflicts, getting a clear high-level view of what’s actually happening is a constant struggle. This guide will walk you through building a project portfolio dashboard in Looker to bring all that chaos into a single, straightforward view, and show you how to use AI to get it done faster.

What is a Project Portfolio Dashboard?

A project portfolio dashboard isn't just another report, it's your command center. It gives you a bird's-eye view of all your projects in one place, pulling data from various sources to show you the overall health of your initiatives. Instead of digging through emails, spreadsheets, and different project management tools, you get one visual summary highlighting progress, budgets, resource allocation, and potential risks.

Having this single source of truth helps you:

Make Better Decisions: Instantly see which projects are succeeding and which are struggling, so you know where to focus your attention and resources.
Manage Resources Effectively: Identify over-allocated teams or individuals before burnout becomes a problem and spot opportunities to shift resources to more critical projects.
Stay Ahead of Risks: Proactively identify projects that are falling behind schedule or going over budget, allowing you to intervene before it's too late.
Save Time on Reporting: Automate status updates and eliminate the weekly headache of manual report building, so you can report to stakeholders in seconds.

Step 1: Planning Your Dashboard Before You Touch Looker

Jumping straight into Looker without a plan is a recipe for frustration. A great dashboard starts with clear questions and goals. Before you build a single chart, take a moment to outline what you need to see.

Define Your Key Metrics (KPIs)

What questions does your dashboard need to answer? Your KPIs will form the backbone of the entire report. Here are some of the most common and valuable metrics for a project portfolio dashboard:

Project Status: A simple categorical status like On Track, At Risk, or Off Track. This gives an immediate health check.
Budget vs. Actual Spend: How much have you spent compared to what you planned to spend? You can view this per project or for the entire portfolio.
Task Completion: A percentage showing how many tasks are completed out of the total tasks for each project.
Milestone Progress: Tracks the completion of key project phases or deliverables against their due dates.
Resource Utilization: Shows how much of your team's available time is allocated to projects. This is critical for preventing burnout and spotting capacity.

Identify Your Data Sources

Where does all this information live right now? Your data is likely scattered across several platforms. Make a list of all of them. Common sources include:

Project Management Tools: Jira, Asana, Monday.com, Trello, etc.
Spreadsheets: Google Sheets or Excel files for tracking budgets, timelines, or custom project data.
Financial Software: QuickBooks, Xero, or other accounting tools that track project expenses.
Time-Tracking Software: Tools like Harvest or Toggl that log hours worked on specific projects.

The goal is to bring data from these sources together into a central location that Looker can access, usually a cloud data warehouse like Google BigQuery, Snowflake, or Amazon Redshift.

Sketch Your Dashboard Layout

Grab a piece of paper or open a whiteboard tool and sketch a rough layout. Where should the most important information go? Think about the story you want to tell.

Top Section: High-level KPIs and summaries. Use scorecards to show total projects, overall budget spent, and percentage of projects on track.
Main Body: The detailed visualizations. A bar chart comparing Budget vs. Actual for the top 5 projects, a pie chart showing the breakdown of projects by status, and a timeline for upcoming milestones.
Detailed Table: At the bottom, include a table with all projects listed, allowing users to drill down into specifics.

This simple exercise helps solidify your vision and makes the building process in Looker much smoother.

Step 2: Connecting Your Data and Modeling it with LookML

This is where Looker’s power - and complexity - really comes into play. Unlike some simpler BI tools, Looker sits on top of a central database or data warehouse. You don’t upload CSVs directly, you connect Looker to an established database where your data already lives.

Once connected, you use Looker’s modeling layer, known as LookML, to define your business logic. Think of LookML as a translator between your raw database tables and your business users. You write simple code to tell Looker things like:

"The spend_usd column is a currency."
"To calculate Profit, subtract the cost column from the revenue column."
"The projects table is related to the tasks table through the project_id field."

Setting up LookML requires some technical patience, but it’s what enables business users to later build reports and explore data without writing any code themselves. It ensures that everyone in the company is using the same definitions for key metrics, creating consistency across all reports.

Step 3: Building Your Dashboard Visualizations in Looker

With an effective plan and your data model in place, building the dashboard is the fun part. Inside Looker, you'll create a new dashboard and start adding "Tiles," which are the individual charts, tables, and visualizations.

Here’s how you could create some of the core elements of your project portfolio dashboard:

Tile 1: Portfolio Health Scorecards

Start with the most important numbers at the top. Use the "Single Value" visualization to create scorecards for:

Total Active Projects
Overall Budget Consumed (%)
Projects On Track (Count)
Projects At Risk (Count)

Tile 2: Project Status Breakdown (Pie Chart)

Every executive loves a pie chart. It’s perfect for showing the proportion of your projects in different states.

From an Explore, select your Project Status dimension and your Project Count measure.
Choose the Pie Chart visualization type.
Customize the colors to be intuitive: green for "On Track," yellow for "At Risk," and red for "Off Track."

Tile 3: Budget vs. Actual Spend by Project (Bar Chart)

This visualization quickly shows which projects are draining your budget.

Bring in your Project Name dimension.
Add your Budgeted Amount and Actual Spend measures.
Select the Bar Chart visualization. This will create side-by-side bars for each project, making comparisons obvious.

Tile 4: A Detailed Project List (Table)

Sometimes, you just need to see the raw data. A table is perfect for this.

Simply select all the important dimensions and measures: Project Name, Project Manager, Status, Due Date, Budget, Actual Spend, and Budget Variance.
Choose the Table visualization. You can enable totals and use conditional formatting to highlight rows where Budget Variance is negative (over budget).

Finally, don't forget to add dashboard filters! Add filters for Project Manager and Department at the top of the dashboard to make it interactive and useful for different team leads.

Where AI Comes In: Using Looker's AI to Accelerate Analysis

Building dashboards manually is powerful, but this is where AI can step in to make your life dramatically easier. More recent versions of Looker include generative AI features designed to turn natural language questions into answers and visualizations instantly.

Conversational Ad-Hoc Reporting

Instead of manually selecting dimensions, measures, and chart types, you can simply ask Looker a question in plain English. For example, during a project review meeting, someone might ask, "Which departments have the most projects at risk?" You can type that directly into Looker’s conversational interface and get a chart instantly.

Here are a few example prompts:

“Show me the budget variance for projects managed by Elena.”
“Compare task completion percentage by month as a line chart.”
“Create a table of all projects due in the next 30 days that are currently 'At Risk'.”

This feature drastically lowers the learning curve and allows anyone on your team to answer their own data questions on the fly, without needing a dedicated analyst.

Proactive Insights and Forecasting

AI isn't just for answering your questions, it can also tell you what questions you should be asking. Looker can automatically detect anomalies and trends in your data, such as a project whose spending is accelerating faster than planned or a team whose resource utilization is unsustainably high. These proactive alerts help you move from reactive problem-solving to proactive management.

Final Thoughts

Building a project portfolio dashboard in Looker organizes your project chaos into a clear, actionable picture, helping you manage resources, mitigate risks, and keep stakeholders informed. By starting with a solid plan, creating a reliable data model in LookML, and then building intuitive visualizations, you can create a powerful command center for all your company’s initiatives.

The time you spend connecting platforms and building out reports is a common bottleneck. This is exactly why we built Graphed. We turn hours of complex setup and dashboard building into a short, 30-second conversation. Simply connect your project management tools and spreadsheets, then ask in plain English for the dashboard you need - Graphed builds it for you in real-time, instantly streamlining your entire reporting process without the steep learning curve.