How Run Controls Work in Looker

Ever opened a complex Looker dashboard and had to wait… and wait… while it chugged through a dozen different queries just to load? You’re not alone. This is where Looker's often-overlooked "run on load" setting can be a lifesaver. This article covers exactly what this feature is, why it's so important for performance and cost, and how you can enable it in just a few clicks.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

What Are Run Controls in Looker?

Looker dashboards, by default, start running all their queries the moment you open them. This is fine for simple dashboards, but for anything complex with multiple tiles, joins, and filters, it can lead to long load times and a sluggish user experience. You open the dashboard, it starts loading. You change a filter, it reloads. You change another filter, it reloads again. Each action triggers a new set of queries against your database.

Run controls, also known as disabling "Run on load," flip this behavior. When you disable this setting, the dashboard loads as an empty shell. None of the tiles will run their queries or display any data when the page first loads. Instead, you'll see a prominent "Run" button at the top of the dashboard. Only when a user clicks that button does Looker execute the queries and populate the tiles with data. This puts the user in control of when the dashboard queries the database, which has some significant benefits.

Why You Should Use Run Controls

Delaying a dashboard's execution might sound counterintuitive, but it solves a few very common and very expensive problems associated with complex reports. It’s one of the easiest ways to improve performance, save money, and create a better user experience all at once.

AI Agents for Marketing

Crash Course

Graphed

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.

Download the PDF

1. Drastically Improve Dashboard Performance

The number one reason to use run controls is to improve dashboard speed. A dashboard with 15-20 tiles, each with its own underlying LookML model and query, can hit your data warehouse hard. When "Run on load" is active, Looker tries to run all these queries simultaneously as soon as the page is opened.

This initial "query storm" can be incredibly slow, especially if:

• The queries are complex: They involve complicated joins across multiple large tables.
• The underlying data is massive: The queries are scanning millions or billions of rows.
• Multiple users are accessing the dashboard: Imagine 10 people on your team opening the same heavy dashboard at 9 AM on a Monday. Your database is now trying to handle 150+ heavy queries at once.

By disabling "Run on load," you prevent this initial performance hit. The dashboard frame loads instantly, and the user can decide when they are ready to incur the "cost" of running the queries. This feels much faster and more responsive from the user's perspective, as they aren’t stuck staring at spinning loader icons.

2. Reduce Unnecessary Database Costs

Every query sent to your data warehouse costs money. Cloud data platforms like Google BigQuery, Snowflake, and Amazon Redshift often charge based on the amount of data processed or the compute time used. Running unnecessary queries is like literally burning money.

Consider a dashboard that auto-loads with a default date range of the "last 90 days." A user might open it, realize they only need to see data from "last 7 days," and immediately change the filter. In this common scenario, you have just paid for a huge 90-day query that was immediately thrown away. You then pay again for the 7-day query the user actually wanted.

When "Run on load" is disabled, this double-spending disappears. The user opens the dashboard, sets their desired filters (change the date range, select a specific region, choose a campaign), and then clicks "Run." Only one set of optimized queries is ever sent to the database, significantly cutting down on wasted computation and lowering your data warehouse bill.

3. Create a Better User Experience for Dashboards with Filters

Nothing is more frustrating than a screen that keeps reloading every time you click a button. For dashboards with multiple user-defined filters, the default auto-run behavior creates exactly this experience.

Let’s walk through the "before" scenario:

1. You open a sales dashboard. It starts querying data for the last 30 days for all sales reps and all product categories.
2. You change the date filter to "This Quarter." The entire dashboard reloads. You wait.
3. You then filter for a single sales rep, "Jane Doe." The entire dashboard reloads again. You wait again.
4. Finally, you filter for the "Enterprise Software" product category. The entire dashboard reloads a third time.

With run controls, the flow is much more logical:

1. You open a blank sales dashboard. It loads instantly.
2. You set the date filter to "This Quarter."
3. You set the sales rep filter to "Jane Doe."
4. You set the product category filter to "Enterprise Software."
5. You click the "Run" button. The dashboard runs one single time, with all your chosen filters applied.

This is a calmer, more efficient, and far less frustrating way for users to interact with filter-heavy reports.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

How to Enable Run Controls on a Looker Dashboard

Putting run controls in place is incredibly simple and takes less than a minute. Here’s a step-by-step guide to disabling "Run on load."

Step 1: Open Your Dashboard in Edit Mode

First, navigate to the Looker dashboard you want to modify. In the top right corner, you'll see a button labeled "Edit dashboard." Click this button to enter edit mode. You must have the appropriate permissions to see this option.

Step 2: Open the Dashboard Settings

Once you are in edit mode, a new toolbar will appear at the top of the screen. On this toolbar, you'll find a "Settings" option, usually on the left side. Clicking "Settings" will open a configuration pane for the entire dashboard.

Step 3: Locate the "Run on load" Option

Inside the Settings pane, you'll see a tabbed interface. Stay on the main "Settings" tab. Look for an option titled "Run on load." It's usually near the top, right under the description box. By default, this option will be enabled (the toggle will be blue or the box will be checked, depending on your Looker version).

Step 4: Disable "Run on load"

Click the toggle or a checkbox next to "Run on load" to disable it. When disabled, the toggle will turn grey. This one action is all it takes to switch the dashboard from auto-loading to manual run control.

AI Agents for Marketing

Crash Course

Graphed

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.

Download the PDF

Step 5: Change How Filters Affect the Dashboard (Optional but Recommended)

Right below the "Run on load" setting, there is another critical setting that dictates how filters behave. By default, changing a filter value will trigger an automatic rerun of the dashboard queries. To get the full benefit of run controls, you want to prevent this.

Uncheck the box that says, "Rerun dashboard on every filter change." This ensures that users can set up all their filters first before intentionally clicking "Run." This step is essential for creating the streamlined user experience described earlier.

When you are finished, ensure that your dashboard has a primary execution button for users to action the report. If this "Run" button does not exist on your dashboard, you can add it through the dashboard Settings as well via 'Add Filter'. Within the Modal > search & add 'Button'.

Step 6: Save Your Changes

Finally, click the "Save" button on the Settings pane, and then click "Save" again on the main dashboard toolbar to exit edit mode. Your changes are now live. The next time anyone loads this dashboard, it will appear blank with a "Run" button waiting for their command.

Best Practices for Using Run Controls

"Run Controls" can be super effective when implemented into your regular business workflow. It's not a silver bullet, however - use these principles to guide decisions when deploying a BI dash for your org:

• Be Selective: Don't disable "Run on load" for every dashboard. A simple, fast-loading dashboard with just a few tiles and very few filters likely benefits from auto-loading. Reserve this feature for your most complex, heavy, and filter-driven dashboards where performance and cost are real concerns.
• Educate Your Users: The first time a user opens a dashboard that doesn’t auto-load, they may be confused. Make sure to communicate this change to your team. Let them know why the dashboard is loading empty and that they now need to set their filters and click "Run" to see the data. A simple text tile at the top of the dashboard with instructions like "Set filters below and click RUN to load the dashboard" can work wonders.
• Position the Run Button Intuitively: You can move the Run button (which is often configured as a Filter Bar) around in Edit Mode just like any other tile. Make sure it’s in a prominent, easy-to-find location, typically near the filters at the top of the dashboard.

Final Thoughts

Disabling "Run on load" is a simple but powerful technique in your Looker arsenal. It gives you the power to tame complex dashboards, prevent runaway database costs driven by unnecessary queries, and provide a user experience that is intentional and far less frustrating, especially on dashboards loaded with filters.

Managing the fine-grained settings in tools like Looker highlights how complex traditional business intelligence can be. At Graphed, we’ve taken a different approach. We believe getting insights shouldn't require you to become an expert on performance tuning or database query optimization. You just connect your data sources - like Google Analytics, Salesforce, or Shopify - and use plain English to ask for the dashboard you need. We handle the process of creating efficient queries and real-time dashboards automatically, so you can spend less time configuring settings and more time acting on your data.