Graphed LogoGraphed
Login

How to Create a Restaurant Dashboard in Looker

Cody Schneider•

Building a successful restaurant is about more than just great food, it's about understanding the health of your business day in and day out. But getting a clear picture often means juggling data from your POS system, online ordering platforms, reservation software, and marketing accounts. This article will show you how to centralize that information by creating a powerful restaurant dashboard in Looker, turning disconnected numbers into actionable insights.

Why Your Restaurant Needs a Centralized Dashboard

If your reporting process involves downloading CSV files from multiple systems and trying to stitch them together in a spreadsheet, you know how painful and time-consuming it can be. By the time you’ve built the report, the data is already old. A dedicated business intelligence dashboard solves this by connecting directly to your data sources, providing a single, always-updated view of your restaurant's performance.

Here are the core benefits:

  • A Single Source of Truth: Everyone from the general manager to the head chef and marketing lead can work from the same live data, eliminating conflicting reports.

  • Spot Trends in Real-Time: Instantly see if lunch sales are dipping week-over-week or if a specific menu item is suddenly gaining popularity. You can react quickly instead of waiting for a messy weekly summary.

  • Improve Operational Efficiency: Track key operational metrics like table turnover, food cost percentage, and labor costs to pinpoint areas for improvement and boost profitability.

  • Make Data-Driven Decisions: Confidently answer crucial questions like, "Should we adjust our staffing on Tuesday nights?" or "Is our new Facebook ad campaign actually bringing in reservations?"

Step 1: Plan Your Restaurant Dashboard Before You Build

Jumping directly into a tool like Looker without a plan is a recipe for a confusing and ineffective dashboard. A great dashboard answers specific questions. To start, you need to know which questions are most important.

Identify Your Key Performance Indicators (KPIs)

What are the vital signs of your restaurant? Your KPIs are the measurable values that tell you whether you're hitting your goals. Group them by category to create a logical flow for your dashboard.

Sales & Revenue KPIs

  • Gross Revenue: Total sales over a period (daily, weekly, monthly).

  • Average Check Size: Total revenue divided by the number of covers. Helps you understand customer spending habits.

  • Revenue Per Available Seat Hour (RevPASH): A powerful metric that measures how effectively you're generating revenue from your available seats.

  • Sales by Category: Breakdown of sales by food, beverage, merchandise, etc.

  • Top 5 Best-Selling Menu Items: Understand what's popular and drives your revenue.

  • Top 5 Least-Selling Menu Items: Identify items that may need to be re-engineered or removed from the menu.

Operational & Cost KPIs

  • Table Turnover Rate: How many times a table "turns" during a service period. Crucial for maximizing capacity.

  • Cost of Goods Sold (COGS): The direct cost of ingredients for the menu items sold.

  • Food Cost Percentage: Your COGS divided by your food revenue. One of the most critical metrics for profitability.

  • Labor Cost Percentage: Your labor cost divided by total revenue. Helps you optimize staffing.

  • Order Volume by Channel: The number of orders from different sources like Dine-In, Delivery (Uber Eats, DoorDash), and Takeout.

Customer & Marketing KPIs

  • Reservations vs. Walk-ins: Understand guest booking behavior.

  • New vs. Returning Customers: A key indicator of customer loyalty and satisfaction.

  • Online Review Ratings: Track average ratings from platforms like Yelp and Google.

  • Marketing Campaign ROI: Compare the revenue generated from specific marketing initiatives against their cost.

Sketch a Low-Fidelity Mockup

Grab a piece of paper or open a simple diagramming tool and sketch out how you want your dashboard to look. Where will the most important KPI (like daily sales) go? How will you group the different charts and tables? A simple visual plan makes the building process in Looker much faster and more focused.

Step 2: Connect and Model Your Data in Looker

This is the most technically demanding part of the process. Looker doesn't just pull data in, it works best when it sits on top of a central data repository, otherwise known as a data warehouse (like Google BigQuery, Snowflake, or Amazon Redshift).

Consolidate Your Data Sources

Your restaurant's data is likely spread across various platforms:

  • Point of Sale (POS): Toast, Square, Lightspeed, etc. (for sales, menu items, and checks).

  • Reservation Platform: Resy, OpenTable, Tock (for bookings and customer data).

  • Delivery Aggregators: Uber Eats, DoorDash, Grubhub (for online order data).

  • Accounting Software: QuickBooks, Xero (for COGS and other financial data).

  • Marketing Platforms: Google Analytics, Facebook Ads (for campaign performance and website traffic).

You need to use an ETL (Extract, Transform, Load) tool like Fivetran or Stitch to pipe the data from these discrete sources into your data warehouse. Setting this up is a project in itself but is fundamental for a holistic view of your business.

Define Your Business Logic with LookML

Once your data is in a warehouse, Looker’s unique advantage comes into play: LookML. LookML is a modeling layer where you use code to define your business metrics and their relationships, independent of any specific chart or report an end user creates. For example, you define what "Revenue" means (e.g., sum of check totals minus discounts) or how to calculate "Labor Cost Percentage" just once.

This is powerful because it ensures consistency. Every single person in your organization who uses Looker to analyze revenue will be using the exact same definition. Creating a solid LookML model is often a job for a data analyst or developer, as it requires a good understanding of both your business logic and data structures.

Step 3: Building Visualizations ("Looks") in Looker

With a well-defined LookML model, building the actual visualizations becomes much simpler. In Looker, individual charts or tables are called "Looks." You create Looks using the "Explore" interface, which exposes the dimensions (attributes like 'Date', 'Menu Item') and measures (calculations like 'Total Sales', 'Average Check Size') from your LookML model.

Example 1: Creating a Daily Sales Trend Line Chart

  1. Navigate to an Explore where your sales data lives.

  2. In the left-hand panel, find and select your 'Order Date' dimension and your 'Total Revenue' measure.

  3. Looker will automatically generate a data table. Change the visualization type to a Line Chart.

  4. You can customize labels, colors, and axes.

  5. Once you're happy, click "Save" and name your Look (e.g., "Daily Sales - Last 30 Days").

Example 2: A Pie Chart for Sales by Channel

  1. Start a new Explore with your order data.

  2. Select the dimension 'Order Channel' (e.g., Dine-in, Takeout, DoorDash) and the measure 'Total Revenue'.

  3. Change the visualization type to a Pie Chart to see the percentage contribution of each channel.

  4. Save this visualization as another Look named "Sales Mix by Channel."

Example 3: A Table for Top-Selling Menu Items

  1. In an Explore connected to your POS data, select the 'Menu Item' dimension.

  2. Then, add the measures you want to see, such as 'Items Sold' and 'Total Revenue'.

  3. Keep the visualization as a Table.

  4. You can sort the table by 'Total Revenue' in descending order to see your most popular items at the top.

  5. Save this as a Look named "Top Selling Items."

Step 4: Assembling and Filtering Your Dashboard

Now it's time to bring all your individual Looks together onto a single dashboard.

Add and Arrange Your Tiles

In Looker, create a new "Dashboard". You can then add the "Looks" you just created as tiles. Arrange them based on your initial sketch, placing the most critical, high-level KPIs at the top (like big numbers for today's revenue and covers). You can drag, drop, and resize each tile until the layout is clean and intuitive.

Add Interactive Filters

A static dashboard is useful, but an interactive one is far more powerful. Looker allows you to add dashboard-level filters that apply to all relevant tiles simultaneously.

The most important filter for a restaurant is a date filter. Users can set it to "Today," "Last 7 Days," "This Month," or a custom range to see how performance changes over time. If you have multiple locations, a location filter is also essential. This allows your team to view data for a single restaurant or compare performance across all of them.

Best Practices for a Valuable Dashboard

  • Keep It Simple: A dashboard crowded with dozens of charts becomes overwhelming and unusable. Focus only on the KPIs that drive decisions. If a chart doesn't inspire an action, consider removing it.

  • Use the Right Visualization: Use line charts for time-series data, bar charts for comparisons between categories (like sales by location), and tables for detailed-level information. Keep pie charts to a minimum.

  • Tell a Story with Layout: Group related metrics together. For example, have a section for sales, another for operations, and a third for marketing. This guides the user's eye and helps them connect the dots.

  • Schedule and Set Alerts: Automate your reporting by scheduling the dashboard to be emailed to key stakeholders every morning. Additionally, you can set up alerts in Looker to notify you if a KPI crosses a certain threshold (e.g., if food costs go above 35%).

Final Thoughts

Creating a restaurant dashboard in Looker is a powerful step towards running a more data-informed business. It transforms raw data from your POS, reservation system, and marketing tools into a single, cohesive view that helps you spot trends, control costs, and ultimately drive growth. The process requires careful planning of your KPIs and a significant technical investment in setting up data pipelines and developing a LookML model.

That technical barrier is precisely why we built our platform. For many businesses, setting up a data warehouse and learning dedicated BI tools like Looker is simply too slow and expensive. We created Graphed to be your AI data analyst, letting you skip the complexity entirely. You can connect your marketing and sales sources in a few clicks and simply describe the dashboard you want in plain English. Instead of spending weeks on setup, our AI can build real-time, interactive dashboards instantly, allowing you to get answers and make decisions in seconds, not months.