How to Create a Project Budget in Looker
Tired of manually updating spreadsheets to track if your projects are on, under, or way over budget? Using a tool like Looker can transform that tedious manual process into an automated, real-time dashboard. This article will walk you through the steps to prepare your data, model it, and build an interactive project budget dashboard in Looker to keep all your stakeholders informed.
Why Use Looker for Project Budgeting Anyway?
While spreadsheets are familiar, they’re static and prone to human error. Every time you need an update, you're back to exporting CSVs and copy-pasting data. A Looker dashboard, on the other hand, connects directly to your live data sources.
This approach gives you a single source of truth that’s always up-to-date. Instead of a report that's stale the minute you send it, you get a dynamic dashboard everyone on the team can access to see real-time project financial health.
- Real-Time Data: See exactly where your budget stands right now, not last Tuesday when you last crunched the numbers.
- A Single Source of Truth: Eliminate confusion from multiple versions of the same spreadsheet floating around in inboxes.
- Interactive Visuals: Drill down into the data to see which tasks, team members, or categories are driving costs.
Step 1: Get Your Data in Order
Before you build anything in Looker, you need organized, accessible data. Looker doesn’t store your data, it queries the data where it lives. This means your first job is to centralize your project finance data in a database or a structured source that Looker can connect to.
Most project budget data comes from a few places:
- Accounting Software: Tools like QuickBooks or Xero for actual expenses.
- Project Management Tools: Systems like Jira, Asana, or Monday.com for tasks and project details.
- Time-Tracking Software: Apps like Harvest or Toggl for billable hours.
- Spreadsheets: A Google Sheet is a great place to start, especially for laying out the initial budgeted amounts per project.
Your goal is to get all of this information into a single table or a set of related tables in a database (like BigQuery, Snowflake, or Redshift). If you lack a formal data warehouse, starting with a well-structured Google Sheet that Looker can connect to is a fantastic starting point.
Essential Data Fields
Your source data should contain distinct columns (or fields) that clearly define each part of your budget. Aim for a structure that looks something like this:
project_id- A unique identifier for each project.project_name- The name of the project.expense_date- The date an expense was incurred.expense_category- What the money was spent on (e.g., Software, Freelancers, Advertising, Travel).line_item_description- A specific description of the cost.team_member- Who incurred the cost.budgeted_amount- The planned spend for a project or line item.actual_spend- The actual amount spent.
Your data might be split, for example, you could have one table with budgeted amounts and another with a running list of actual expenses. That's perfectly fine - you can join them later in Looker's modeling layer.
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.
Step 2: Define Your Metrics with LookML
Once your data source is connected to Looker, the next step is to create a model. This is where Looker’s real power lies. You use LookML (Looker’s modeling language) to define your business logic - telling Looker what your data means, how tables are related, and how to calculate metrics.
Think of LookML as a recipe book for your data. You write the recipes (calculations like 'Budget Variance' or 'Percentage Spent') once, and then any user in your company can use those metrics without having to know the underlying code.
Creating Dimensions and Measures
In LookML, you work with "Dimensions" and "Measures."
- Dimensions are your data attributes - things you can group by. Think Project Name, Expense Category, or Date.
- Measures are aggregations or calculations performed on your dimensions. Think Total Actual Spend, Total Budgeted Amount, or Average Cost.
Here’s a simplified LookML example for a view called project_budget:
view: project_budget {
sql_table_name: `your_database.project_costs`,
dimension: project_name {
type: string
sql: ${TABLE}.project_name ,
}
dimension: expense_category {
type: string
sql: ${TABLE}.expense_category ,
}
dimension_group: expense_date {
type: time
timeframes: [date, week, month, year]
sql: ${TABLE}.expense_date ,
}
measure: total_budget {
type: sum
sql: ${TABLE}.budgeted_amount ,
value_format_name: usd_0
}
measure: total_actual_spend {
type: sum
sql: ${TABLE}.actual_spend ,
value_format_name: usd_0
}
}Crafting Custom Calculations for Deeper Insights
The real magic happens when you create custom measures that answer key business questions. Here are three essential metrics for any project budget dashboard that you can define directly in LookML.
1. Budget Variance
This shows you whether you're over or under budget. It's a simple subtraction of actual spend from the budget.
measure: budget_variance {
type: number
sql: ${total_budget} - ${total_actual_spend} ,
value_format_name: usd_0
}2. Percent of Budget Spent
This gives an immediate sense of pacing. Are you 25% through the project timeline but have already spent 75% of the budget? That’s something you need to know now.
measure: percent_of_budget_spent {
type: number
sql: ${total_actual_spend} / NULLIF(${total_budget}, 0) ,
value_format_name: percent_2
}3. Budget Status
This is an incredibly useful field that automatically categorizes projects for at-a-glance monitoring. You can create a custom dimension using a CASE WHEN statement to flag projects.
dimension: budget_status {
type: string
case: {
when: {
sql: ${percent_of_budget_spent} > 1 ,
label: "Over Budget"
}
when: {
sql: ${percent_of_budget_spent} BETWEEN 0.9 AND 1 ,
label: "At Risk"
}
when: {
sql: ${percent_of_budget_spent} < 0.9 ,
label: "On Track"
}
else: "No Budget"
}
}By defining this logic once in LookML, you empower your entire team to use these flags correctly across all their reports.
Step 3: Build Your Interactive Dashboard
With your data modeled, it's time to build the fun part: the dashboard. In Looker, dashboards are collections of “Tiles,” which are individual reports or visualizations you create in the “Explore” interface.
Navigate to the Explore for your project budget model and start building tiles that answer key questions.
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.
Key Visualizations for a Project Budget Dashboard
Here are some essential tile ideas to get you started:
1. At-a-Glance KPI Scorecards
Use the "Single Value" visualization to create large, bold readouts of your most important top-line metrics. These should be at the very top of your dashboard.
- Total Budget
- Total Actual Spend
- Overall Budget Variance
2. Budget vs. Actual by Project
A grouped bar chart is perfect for this. Set your Dimension to Project Name and your Measures to Total Budget and Total Actual Spend. This gives you a clear visual comparison of planned vs. actual costs for every project.
3. Budget Pacing Over Time (Burndown/Burnup Chart)
A line chart showing cumulative spending helps you understand your budget burn rate. Set your Dimension to Expense Date (grouped by week or month) and your Measure to Total Actual Spend (using a running total). You can even add a reference line representing your budgeted spend to see if your actuals are pacing above or below the plan.
4. Detailed Project Status Table
Use a table visualization to create a detailed project breakdown. Select dimensions like Project Name and Budget Status, along with measures like Total Budget, Total Actual Spend, Budget Variance, and Percent of Budget Spent. Use Looker’s conditional formatting to color-code the Budget Status field: red for "Over Budget", yellow for "At Risk", and green for "On Track".
5. Adding Interactivity with Filters
The true power of a Looker dashboard is its interactivity. Add filters to the top of your dashboard that allow users to slice and dice the data themselves. Common and useful filters include:
- Date Range: Let users view performance for This Quarter, Last Month, or a custom range.
- Project Name: Allow drilling down into a single project.
- Expense Category: Answer questions like "How much are we spending on freelance contractors across all projects?"
Final Thoughts
Building a project budget tracker in a BI tool like Looker takes some initial setup, but the payoff is huge. You transform a painful, manual reporting task into an automated, always-on command center that keeps stakeholders informed and projects on track with real-time financial data.
For many teams, the main challenge with traditional BI tools is the technical expertise and setup time required, especially for creating the data models. Instead of learning LookML or wrangling data pipelines, we use AI to help handle that heavy lifting. With a modern tool like Graphed, you can securely connect your data sources (like QuickBooks, Salesforce, or Google Sheets) and simply ask in plain English, "Create a dashboard comparing budgeted versus actual spend by project for this quarter," and the dashboard is instantly built for you - no complex modeling required.
Related Articles
How to Create Facebook Ads API for Claude Code
Create a Meta Marketing API access token that never expires, using a system user, so Claude Code can pull Facebook ad reports and manage campaigns.
ClickHouse vs Postgres: When to Switch to Columnar
Comparing ClickHouse vs PostgreSQL means matching storage architecture to workload. Row storage wins for transactions. Columnar wins for scan-heavy analytics.
Google Ads for Orthodontists: Get More Booked Consults
Most orthodontic Google Ads campaigns waste money on low-intent searches. The problem is structure, not the platform.