How to build your first Retool dashboard

Dashboards are one of the most popular Retool use cases. They transform raw data into interactive visualizations that teams actually use—and with the platform's low-code approach, you can build custom applications and business dashboards in an afternoon instead of weeks.

In this tutorial, you'll connect sample data to build a revenue dashboard with interactive charts and tables, filters, and drill-downs that make your data actionable.

What we're building

A revenue dashboard that includes:

KPI cards: Total revenue, order count, average order value
Trend chart: Weekly revenue visualization
Category breakdown: Revenue by product category
Order table: Recent orders with interactive filters
Drill-downs: Click to view order details in a slide-out drawer

This covers the essentials of building powerful internal tools: data visualization, interactivity, and real-time data updates.

Follow along: Download sample data to build this dashboard:

orders.csv - 150 orders from Sept-Oct 2025
categories.csv - 6 product categories

Step 1: Connect your data source

Before building, connect to your data so Retool can read your schema and generate queries automatically. Having your data connected first allows the AI to understand your tables, columns, and relationships, resulting in more accurate generated queries.

Option A: Retool Database (fastest)

Use RetoolDB—our managed PostgreSQL database with a spreadsheet-like interface:

1.
Go to Resources > Create new > Resource
2.
Select Retool Database
3.
Select the + button on the left, then select Import from CSV and upload orders.csv and categories.csv
4.
Tables are created automatically with the correct schema

For production data, connect to your existing data sources:

1.
Go to Resources > Create new
2.
Select your database (Postgres, Snowflake, MySQL, BigQuery, Databricks) or API
3.
Configure connection details
4.
Name it clearly (e.g., analytics_warehouse)

Step 2: Generate your dashboard with a prompt

Next, it’s time to build your app by starting with a prompt. Be specific about what you want while letting the AI handle the implementation details.

Open the Assist tab

1.
From your Retool home screen, click Create > App
2.
Click the Assist icon in the lower-left corner
3.
If you don't see it, check Settings > AI and verify that the Retool AI and Assist Tab settings are toggled on.

Write your prompt

Be specific about components, layout, and data sources. Here's an example of what an effective prompt looks like:

Build a revenue dashboard using my @RetoolDatabase orders table.

Include:
- KPI cards showing total revenue, order count, and average order value
- A line chart showing weekly revenue trends
- A bar chart breaking down revenue by category (join to categories table)
- A table of recent orders with columns: id, created_at, customer_id, region, total, status
- Filters: date range (default last 90 days), region multiselect, category multiselect

Use the 12-column grid layout. Place KPIs at the top, charts in the middle, and the table at the bottom.

Prompting best practices:

@mention your data sources (@RetoolDatabase) so Retool can read schema
Specify exact UI components you want (KPI cards, line chart, table)
Define layout preferences to get the structure you envision
Request simple, maintainable queries that other team members can understand

The more specific you are about the structure, the less manual refinement you'll need afterward. Think of it as giving clear requirements to a colleague—the AI will make intelligent decisions within those constraints.

Review and accept

Retool generates:

SQL queries that connect and query data (you’ll be asked for approval before any queries are run)
Text, Chart, and Table components
Bindings between queries and components
A 12-column grid layout

This isn't a black box—you can see exactly what code is generated before it's added to your app. Review the diff to understand what's being created, then click Accept changes.

Step 3: Refine queries and layout

Now that you have your initial concept, it’s time to refine. You can edit manually by writing your own queries and arranging your components, you can edit with prompts, you can drop to custom code, or switch between all three approaches.

Adjust SQL queries

Open any generated query to modify it. You can add filters, transform the data returned from your query, or configure event handlers that run when the query succeeds or fails. For example, show a success notification when data updates, or display helpful error messages to users when a query fails.

If you don’t want to modify your queries yourself, you can do it from a prompt. Try, "Add a filter for category when [your category filter] has a value."

The choice is yours: edit SQL directly for precise control, or describe what you want and let Assist handle the implementation.

Refine the layout

Just like you can refine your queries manually or with AI, you can edit your UI the same way.

Drag and drop components on the canvas to adjust their positioning— or add new components. Common refinements include:

Resizing charts
Repositioning filters
Rearranging KPI cards
Expanding the table

You can also prompt the Assist tab with layout changes: "Move the filters to the left side and make the table span the full width" or "Stack the KPI cards horizontally at the top."

You can freely mix both approaches—drag a few components, then prompt for the next adjustment.

Style components

Once you’ve laid out your components how you’d like, Retool gives you extensive control on how you style them.

Appearance section

Use the Appearance section of the Inspector to customize the appearance of any component in Retool. You can change row colors in tables, border radiuses for containers, padding for text components, and so much more. Read more about appearance customization here.

Conditional formatting

Make data stand out visually with conditional formatting. To modify a column's background color, select your table, navigate to Columns, choose the column you want to style, and open the Appearance section. For example, set the total column's background color to {{ currentSourceRow.total > 200 ? '#9ed692' : '#d68b7c' }} to highlight high vs. low order values.

Themes

Apply organization-level themes with your brand colors and typography. Navigate to Settings > Themes to create custom themes that can be reused across all your apps.

Bulk editing

You don’t have to style each component individually. Select multiple components (click and drag, or Cmd/Ctrl + click) and edit them simultaneously in the Inspector. Change colors, fonts, or spacing for several components at once.

Or prompt the Assist tab for styling: "Make revenue values bold and green," "Apply a red background to orders with status 'cancelled'," or "Update the application to use our brand colors, fonts, and logo."

Step 4: Add interactivity

Static dashboards display data. Interactive dashboards help teams make decisions. Event handlers transform your dashboard from a report into a tool that drives action.

Retool supports many interactive elements:

Chart interactions: Click on data points to filter other components or trigger queries
Row actions: Add buttons to table rows for actions like view, edit, or delete
Drill-downs: Open drawers or modals to show detailed information without leaving the page
Form submissions: Update data and refresh displays automatically
Component triggers: Chain actions across multiple components with success handlers

Add chart interactivity

Let’s add interactions to your bar chart. Using event handlers, we’ll make it so that when users click a bar, the entire dashboard filters to that category, letting them drill from high-level trends to specific details without leaving the page.

Manual setup:

Select your bar chart
Go to Interactions section of the Inspector
Add event handlers for the Select event
Trigger the queries powering your table, KPI cards, and trend chart
Update those queries to filter on the selected category

Or, you could put a prompt like this in the Assist tab:

Make my category bar chart interactive. When a user clicks a bar, 
filter the orders table, KPI cards, and trend chart to show only 
data for that category. When nothing is selected, show all data.

Retool will add the necessary event handlers and modify your queries to include conditional filtering. Review the changes to see how it references the selected data point in your SQL WHERE clauses.

Step 5: Secure and share

Retool apps ship with strong defaults: SOC 2 Type II compliance, granular permissions, audit logs, and environment-level resource access. You can deploy on Retool Cloud or self-host on your own infrastructure, integrate with your existing identity provider, and configure role-based access control to ensure apps meet your organization's security requirements.

Once you've built your dashboard, you can also configure permissions to control who can access and modify it.

Control access with permission groups

Create permission groups that reflect how your team works. Common patterns include:

Analysts: View-only access with Use permissions to run queries and filter data
Operators: Can edit specific fields with Use + Edit permissions but can't modify queries
Admins: Full control with Own permissions to modify queries, manage permissions, and deploy changes

From the Retool homepage, navigate to Settings > Permission to create custom groups or use the default groups (All Users, Admin). For each group, set app permissions to Use, Edit, or Own. You can also configure resource permissions by environment—for example, giving developers Edit access to staging resources while restricting production access.

Where to go next

Once you've built your first dashboard, try building another one on top of your own data.The patterns you learned here—KPI cards, interactive charts, filtered tables, and permission controls—apply to virtually any dashboard use case.

Retool's combination of drag-and-drop components, SQL and API integrations, and AI-assisted building makes it straightforward to create custom internal tools that scale across your organization.