Retool is a leading platform for rapidly building business apps and custom internal tools. Instead of starting from a blank codebase, you drag pre-built components onto a canvas, connect them to your data, and customize with code when needed.
I tested the platform to see if it lives up to the hype. It's impressively easy to use, and the new AI features help you get started faster. But in the end, you're still working with Retool's set of building blocks.
In this article, we’ll cover:
- What Retool is and how it works
- Its strengths and weaknesses based on hands-on testing
- How Retool compares to alternatives like Superblocks
What is Retool? The 30-second answer
Retool is a low-code development platform for building custom business applications like admin panels, dashboards, and workflows. You can use prompts to generate the initial app, then use the visual editor to tweak it.
Key features
- Drag-and-drop app builder: Retool has over 100 pre-built UI components (tables, forms, charts, buttons) that you drag onto a canvas and connect to your data sources. The components are customizable with JavaScript and CSS, but you can also bring your own custom React components.
- Enterprise controls: Retool provides role-based access controls, SSO integration, audit logs, and Git-based version control. You can also self-host it for full control over your data and infrastructure.
- Workflows and automation: The visual workflow builder automates multi-step business processes. It supports conditional flows, automatic retries, and scheduled actions.
- External apps: It supports white-labeled portals for customers, vendors, or partners. This extends Retool beyond internal software to customer-facing applications.
- AI actions and Agents: Retool connects to major LLM providers (OpenAI, Anthropic, Google, or custom models). It has pre-built AI actions for text generation, classification, and document extraction. You can also build Agents to execute multi-step tasks across your tools.
- AI-assisted app generation: Assist is the new tab in Retool IDE that generates or modifies apps from natural language prompts. It works by assembling an app from Retool's library of pre-built components.
- Mobile app development: Retool Mobile allows you to create internal tools for iOS and Android with pre-built components. You can use native device features to take photos, scan barcodes, and read NFC tags.
- Database and API integrations: Retool connects to over 70 databases and services, including PostgreSQL, MySQL, MongoDB, Stripe, and REST/GraphQL APIs. Most integrations work with a point-and-click setup, though developers can build custom connections for unsupported systems.
How does Retool work?
Retool assembles internal tools that sit on top of your data.
It has 3 main layers:
- Components (UI layer): These are the tables, forms, modals, charts, buttons, and other building blocks. You can configure their properties and bind them to data.
- Queries (Data layer): This is where Retool connects to your databases, APIs, Google Sheets, etc. Each query is a reusable block that you can pull, push, or transform data. You can trigger the queries through UI events like button clicks or automatically when data changes.
- Scripting and logic (Glue layer): You can write inline logic using JavaScript. It also supports Python, though that’s limited.
To build apps, use AI, start from a template, or drag components on a blank canvas. Write JavaScript directly in the app to add custom logic like calculating a value.
The workflow builder uses a visual flowchart approach. You drag logical blocks (API calls, database queries, conditional steps) onto a canvas and connect them to create multi-step automations. These can run on schedules, respond to webhooks, or trigger from app actions.
Retool AI and agent capabilities
Retool has gone all-in on AI recently. The star of the show is Assist, which lives in a tab in your IDE. You can type a prompt like, "Build me an order dashboard," and it will scaffold a working app for you using Retool's components.
It also has an "Ask AI" shortcut for generating or fixing SQL and JavaScript on the fly and pre-built AI actions for tasks like summarizing text.
You can build agents that connect to your business tools or APIs to automate processes between them. You can trigger agents from apps, workflows, or email.
Retool vs. Superblocks: What's the difference?
Retool and Superblocks are internal app builders, but Superblocks is more open and extensible.
Here’s how these two platforms compare:
Superblocks wins for teams that want true AI-native development and maximum flexibility.
With Retool, the AI doesn't generate open, editable code. It generates Retool-specific components that stay locked in its framework. While you can create your own custom components, that's a far cry from being able to fully edit your app in a real IDE.
Superblocks doesn’t constrain you. You can export the app that the AI generates and customize it in your favorite IDE (like VS Code or Cursor). The 2-way live sync keeps your visual edits and code changes aligned.
Deployment is another big difference. Retool’s self-hosted option means you manage the entire infrastructure. Superblocks uses a lightweight on-premises agent. You keep data control without the pain of managing full infrastructure.
What I liked and didn't like about Retool
I signed up for the free plan and gave Retool a try. The Assist feature was surprisingly helpful when setting up my first app. It only took a few minutes to generate a simple refund processing tool. I could tweak what it built either through prompts or visually by dragging components around and configuring properties in the editor.
That said, you’re still confined to Retool’s design system. You don’t have full creative control over the UI.
What works well
Once you get past the first bit of “where do I click?” confusion, Retool starts to feel fast and capable.
Here’s what stood out:
- AI features fit into the flow: The AI assistant lives right inside the Retool IDE. Adding AI-powered elements like LLM chat or agent components is as simple as dragging them onto the canvas. You can connect to Retool’s vector database to give AI queries more context. Workflows can even feed this vector store data from tools like Slack.
- Built-in resources: Retool includes its own database, storage, email, and vectors, so you don’t need to leave the platform for that functionality.
- Large library of templates: Apps, workflows, and agents all came with templates. The agent builder also has a configuration AI assistant that accepts prompts and helps shape the logic. It made building an agent much easier than I expected.
Where it falls short
Two areas stood out as consistent pain points, both for me and in Retool reviews:
- UI design limitations: I tried tightening up a form layout and moving buttons exactly where I wanted them, but the grid kept snapping components back. If you care about pixel-perfect control, it gets frustrating.
- Workflow builder gets cluttered: Each step in a workflow shows up as a node with arrows connecting them. With just a few nodes, it’s easy to follow. But once you add more, the canvas becomes difficult to navigate.
Who uses Retool?
A wide range of professionals in technical and operational roles across organizations use Retool. Well-known Retool customers include Ramp, Plaid, Twilio Segment, and DoorDash.
It’s popular with:
- Data engineers: Automate ETL jobs, schedule recurring tasks, and integrate data from multiple sources (SQL, NoSQL, APIs) into unified dashboards for monitoring pipelines.
- Backend engineers: Build frontends for backend services and APIs, giving teams usable interfaces without writing a full UI layer from scratch.
- Product managers: Create internal reporting dashboards from product, sales, and support data.
- Business analysts: Replace static spreadsheets and BI reports with interactive apps that let them drill into data, trigger actions, and make decisions faster.
Retool integrations and supported databases
Retool connects to a wide variety of data sources out of the box, which is a big part of its appeal. You don’t need to build custom connectors for the tools you already use.
On top of that, there’s the Retool database, file storage, and vector database for AI use cases. Right now, it supports over 70 native connections.
Retool supports both transactional databases and analytical stores, including:
- Relational databases: AlloyDB, MySQL, PostgreSQL, and OracleDB
- NoSQL databases: Cassandra, Amazon DynamoDB, Google Cloud Datastore, and Redis
- Data warehouses: BigQuery, Snowflake, Amazon Redshift, and Databricks SQL
- Indexing engines: Elasticsearch
Beyond databases, Retool lets you pull data from your SaaS platforms through APIs. You can connect to Amazon S3, Firebase, Slack, GitHub, Google Sheets, Jira, and other services. If there isn’t a built-in connector, you can add any REST or GraphQL API as a custom resource.
Setting up a connection is usually a point-and-click process. You pick the resource type, drop in credentials or an API key, and Retool makes it available in your app. Once connected, you can query the data directly, bind it to UI components, and trigger actions like updates or API calls.
Retool pricing and plans
Retool pricing is seat-based. The platform charges more for standard users (who build apps) and less for end users (who only use them). External users are priced separately, and AI agents are billed by the hour.
Here’s how the plans break down:
Who should use Retool? My take
Retool is great for internal tools, portals, and workflows where getting something usable quickly matters more than owning every line of code. But it’s not for every team.
Retool is perfect for:
- Teams with mobile internal tool needs: If your field teams, warehouse staff, or service technicians need native mobile apps with barcode scanning, offline mode, and real-time data sync, Retool Mobile handles this well.
- Companies building customer or partner portals: The External Apps feature lets you create white-labeled portals where clients, vendors, or partners can access their data securely. This extends your internal tools to external users without building separate systems.
- Organizations requiring fully on-premise deployments: If regulatory requirements or security policies mandate that everything runs in your infrastructure with no cloud components, Retool's full self-hosted option delivers this.
Skip Retool if you:
- Want open AI app generation: Retool’s AI assembles apps using its own pre-defined components. You won’t get fine-grained control over UI, UX, or architecture.
- Need a fully open, exportable codebase: Retool keeps you within its platform. If you want true AI-native development where you describe what you need and get a working app with no lock-in, use Superblocks.
How to get started with Retool in 4 steps
You can usually get up and running with Retool quickly if you already have a database or API to connect to.
Here’s how to get started:
- Sign up and set up an account: Head to the Retool website and create an account. From the dashboard, go to the Apps section and name your first app.
- Build the UI: Retool organizes components and code into pages. To start, describe your app in the Assist tab or drag and drop a component to the canvas.
- Connect to data: Add a resource query and connect it to a data source. Retool supports popular databases, APIs, and its own built-in PostgreSQL and vector databases. You can also connect through the REST API integration if you’re using a custom service.
- Update the component with your data: Components will load with demo data. You’ll need to update it with your own. Open the inspector, change the data source to the name of the resource query you used in step 3, and the component will refresh with live data.
This quick start only scratches the surface. Check out the Retool docs for an in-depth guide.
Retool best practices I wish I knew earlier
When I first started with Retool, I rushed to get apps working and only later realized I could’ve saved time (and headaches) with a few best practices. Knowing these upfront makes Retool projects easier to maintain and scale.
Essential practices:
- Name everything clearly: Queries, components, and variables pile up fast. Use descriptive names like get_users instead of query2 so you don’t lose track later.
- Group logic into modules: Instead of stuffing everything into one app, use Retool modules for shared UI layouts.
- Lean on templates: Retool’s templates are shortcuts to production-ready patterns you can adapt instead of rebuilding from scratch.
- Limit unused queries and data: Regularly clean up unused queries and avoid fetching unnecessary columns or rows. It reduces lag in performance.
Common mistakes to avoid:
- Bloated and unscoped apps: Trying to pack too much functionality into a single app leads to slow load times and maintainability issues. Keep app scope intentional and clear.
- Inconsistent design: Navigation, actions, and styling should feel unified. Inconsistencies confuse users and increase training time.
My verdict on Retool
After testing Retool, I’d recommend it for teams comfortable operating inside its ecosystem. It’s packed with built-in resources, including a database, storage, email, and vector store, plus one of the broadest template libraries out there. The app generation feature and configuration assistant for agents make setup faster and automation more approachable.
However, you can’t export or fully own the underlying code and customization within Retool only goes so far. If you want more open, code-level flexibility or hybrid deployment options, Superblocks gives you more control.
Secure AI-native app development with Superblocks
Superblocks gives business and engineering teams the speed of AI-assisted building with the flexibility and control of full-code development.
The platform makes this possible with its extensive set of features:
- Flexible development modalities: Teams can use Clark to generate apps with AI from natural language prompts, then refine them in the WYSIWYG drag-and-drop visual editor or in code. Changes you make in code and the visual editor stay in sync.
- Context-aware AI app generation: Every app built with Clark automatically abides by organizational standards for data security, permissions, and compliance. This addresses the major LLM risks of ungoverned shadow AI apps.
- Centrally managed governance layer: It supports granular access controls with RBAC, SSO, and audit logs, all centrally governed from a single pane of glass across all users. It also integrates with secret managers for safe credentials management.
- Keep data on prem: It has an on-prem agent you can deploy within your VPC to keep sensitive data in-network.
- Extensive integrations: It can integrate with any API or databases. These integrations include your SDLC processes, like Git workflows and CI/CD pipelines.
- Forward-deployed engineering support: Superblocks offers a dedicated team of engineers who’ll guide you through implementation. This speeds up time to first value and reduces workload for your internal platform team.
If you’d like to see Superblocks in action, book a demo with one of our product experts.
Frequently asked questions
What is Retool used for?
Retool is used to build internal tools, dashboards, workflows, and portals quickly. It gives teams a drag-and-drop editor for UI components, built-in integrations with databases and APIs, and options to add custom logic with SQL or JavaScript.
How much does Retool cost?
Retool costs $10 per standard user per month and $5/end-user (billed annually) on the entry-level plan. AI agent usage is billed by the hour.
Is Retool good for developers?
Yes, Retool is good for developers because they can still write custom SQL, JavaScript, and Python where needed. The trade-off is that you’re working inside Retool’s platform rather than owning a fully exportable codebase.
What is the best internal tool builder?
Superblocks is the best internal tool builder because it lets you scaffold apps and workflows with AI prompts within your security and governance controls.
Can Retool connect to any database?
Retool can connect to almost any database using an API or JDBC. It also offers support for 70+ native connectors out of the box, including PostgreSQL, MySQL, OracleDB, MongoDB, DynamoDB, BigQuery, and Snowflake.
Is Retool a replacement for a custom software development company?
No, Retool is not a replacement for a custom software development company because it still has limited UI customization options and backend support for complex workflows. If you need highly polished, customer-facing apps, a development firm is still the better choice.
