Quick Start Guide
This guide walks you through the basics of a full integration of GrowthBook for both feature flagging and A/B testing. You can use GrowthBook for feature flags, running no-code experiments with a visual editor, analyzing experiment results, or any combination of the above. Feel free to skip to the sections that best apply to your use case.
In our documentation, we use A/B test and experiment interchangeably.
Set Up GrowthBook
Use GrowthBook hosted on our cloud or self-host it. For the easiest setup, sign up for a free account. If you prefer to self-host, follow the instructions here or on our GitHub page.
Once you have a GrowthBook account set up, you're ready to start feature flagging and experimenting!
Feature Flags
Step 1. Add an SDK Connection
To use Feature Flags, use an SDK Connection to connect GrowthBook to your app.
From the menu, choose SDK Configuration → SDK Connections. Then, click Add SDK Connection.
Name your SDK connection, select the language you're using in your application, and configure additional options. Don't worry! These settings can be changed at any time. To use GrowthBook with multiple languages, create a separate SDK Connection for each one.
With the SDK Connection created, you're now ready to install the SDK in your application.
Step 2. Integrate GrowthBook Into Your Application
Use GrowthBook SDKs to evaluate feature flags and run experiments. Tailored instructions based on your selected language and settings will guide you through the installation process. The full SDK docs are also available.
The basics of installing the SDKs are:
- Grab the GrowthBook SDK package for your language.
- Add the basic initialization code to your application.
- Make sure
clientKeyandapiHostare set correctly. - Optionally, add any targeting attributes you wish to use for feature flags and experiments.
If your application has multiple languages or platforms, you can create a separate SDK Connection for each one. GrowthBook features and experiments will work the same across all languages, platforms, and environments!
Step 3. Create a Feature Flag
On the Features page, create your first feature flag.
The Feature Key is what you will reference in your application and cannot be changed later.
Feature flags in GrowthBook are robust, supporting advanced targeting, powerful rule evaluation, prerequisite features, JSON schema validation, and more. In this quick start, we'll just stick with a simple boolean feature flag that is always on or off for everyone.
Step 4. Use the Feature in Your Application
When you create your first feature, you'll see instructions on how to use it in your application. Here's an example using our JavaScript SDK:
if (gb.isOn("my-feature")) {
console.log("It's On!")
}
It really is that simple to get started! For next steps, we recommend reading our Feature Flag Basics page, which goes into more depth.
Feature flags are the foundation for powerful experimentation. In the next section, we show you how to use them to run no-code experimentation using our Visual Editor. Need more advanced options? Dive into our complete guide for code-based experimentation.
No-Code Experimentation
Requires a Pro or Enterprise license. Start a 14-day free trial to try it out.
Step 1. Create an SDK Connection
Navigate to SDK Configuration → SDK Connections. Then, add a new SDK Connection. For the language, choose any no/low-code options: Webflow, Wordpress, Shopify, or Script Tag.
Step 2. Add the Script Tag to Your Website's Head
When you create an SDK Connection, GrowthBook displays tailored instructions based on your no-code platform and settings. Typically, these instructions involve adding a single script tag to the head of your website. It will look something like the example below (replace YOUR_CLIENT_KEY with your uniquely generated client ID):
<script async
data-client-key="YOUR_CLIENT_KEY"
src="https://cdn.jsdelivr.net/npm/@growthbook/growthbook/dist/bundles/auto.min.js"
></script>
Step 3. Install the GrowthBook Chrome Extension
Before using the Visual Editor, you must install our Chrome Extension.
This extension enables you to make changes to your website by pointing and clicking. Use it to test different headlines, CTA text, hero images, and more.
Step 4. Create a New Visual Experiment
To use the visual editor, add a new experiment. Go to Experiments and click Add Experiment.
Choose Design a New Experiment. Then, fill out the fields (hypothesis, variation names, goal metrics, etc.). Don't worry, you can change these values later.
Once you created an experiment, you will be prompted to launch the Visual Editor.
Use the editor to make changes to your site. To return to GrowthBook, click the Done button. We have an entire guide dedicated to setting up and using the Visual Editor, so check it out if you get stuck or want to try some of the more advanced features like drag-and-drop reordering.
Step 5. Start Your Experiment!
When you are ready, click the Start Experiment button. Within seconds, GrowthBook will begin to bucket users into your variations. You can just as easily stop or make changes if needed.
Follow the Experiment Analysis section below to learn how to connect GrowthBook to your data and view results. Depending on the analytics tool and data warehouse that you use, it can take up to 24 hours for results to start showing up after you start an experiment.
Experiment Analysis
Step 1. Connect to Your Data Warehouse
GrowthBook is warehouse native: it connects to your data warehouse and queries it to get the results of your experiments. We support all the popular SQL data warehouses such as BigQuery, Snowflake, Postgres, MySQL, Redshift, Databricks, and even Mixpanel. GrowthBook is extremely flexible and can support almost any schema structure with a bit of configuration.
Our most popular option is to have Google Analytics 4 (GA4) sync data to BigQuery. We also provide guides on working with other popular data sources.
To connect GrowthBook to your data warehouse, add a data source. Go to Metrics and Data → Data Sources. From here, click Add Data Source. Follow the instructions to connect to your data warehouse and event trackers. Providing your data warehouse information enables GrowthBook to connect to your data warehouse and execute queries. The event tracker information allows GrowthBook to generate more accurate SQL templates given the schema the event tracker uses.
Additional guides on how to set up the data source for your specific data warehouse are available.
Step 2. Add a Metric or Two
GrowthBook needs to know what metrics you want to measure for your experiments. For the easiest setup, we recommend using Fact Tables. Most users create a Fact Table for each type of event (Sign Up, Purchase, etc.). With Fact Tables in place, you can then quickly create a library of metrics (Sign-Up Rate, Revenue per User, Items per Order, etc.).
Go to Metrics and Data → Fact Tables. Click Add a Fact Table and edit the default SQL if needed. Here's an example SQL query for an Orders fact table.
SELECT
user_id,
timestamp,
qty,
amount
FROM orders
Once the fact table is created, you can easily add metrics on top of it. For example, consider a simple proportion metric, which measures the percentage of experiment users who complete an action at least once. In this case, it'd measure the percentage of users who made a purchase.
Metrics in GrowthBook are powerful and have many advanced settings. Read about them on our Fact Tables page.
Step 3. View Experiment Results
Go to the Experiments page. If you implemented an experiment using feature flags or our Visual Editor, your experiment will already be listed. Click through to it. If you ran your experiment outside of GrowthBook instead, that's fine, too! Add a new experiment and select the Analyze Existing option.
Once inside your experiment, go to the Results tab. Edit your analysis settings and pick the metrics you created earlier. The results will show up in a table like this:
We have an entire page in our docs just about these results and how to interpret them.