LimeJourney is an open-source customer engagement platform with AI capabilities that help businesses create personalized, data-driven user journeys at scale.
LimeJourney evolved from a closed-source platform called "Stapl," born out of a real need I had while running Teal. I needed a way to communicate with customers at the right time, ideally triggered by specific actions. However, I encountered two major challenges with existing tools:
- Many were expensive.
- They required switching our entire communication stack to their platform.
To address these issues, I created LimeJourney with two key principles:
-
Open-Source Accessibility: We've made LimeJourney open-source, allowing smaller teams and developers to use it freely. For those preferring a managed solution, we offer a cloud version.
-
BYOI (Bring Your Own Integrations): LimeJourney allows you to connect your existing tools, eliminating the need for a complete overhaul of your communication stack.
I'm always excited to chat about LimeJourney and how it can help your business. Feel free to reach out at [email protected] or try our hosted demo [Insert Demo Link Here] to experience LimeJourney for yourself.
-Tobi Okewole, Founder LimeJourney
- 🤖 AI-Driven Insights: We use AI models to help you understand your customers better
- 🛠 Visual Journey Builder: The visual journey builder allows you to create complex user journeys with an intuitive drag-and-drop interface
- 🎯 Dynamic Segmentation: We allow you to create dynamic segments that are based on either user events or attributes - You can create segments like "Users who have not purchased in the last 30 days"
- 🔗 Integration: You can connect your existing tools and services to LimeJourney, and we are working hard on adding more integrations especially around data ingestion
Ready to revolutionize your customer engagement strategy? Try Our Demo | Book a Call with Our CEO
LimeJourney is built on a modern, scalable architecture designed to handle high-volume data processing and real-time interactions.
It is a fullstack monorepo application built with TypeScript across the board.
We use:
- Turbo for managing the monorepo
- Next.js for the web app
- TailwindCSS for styling
- Express for the API
- TSOA for generating API routes and OpenAPI documentation
- PostgreSQL as the main datastore
- Redis for a KV store for managing Journey<>Trigger mappings
- ClickHouse for managing high volume event/entity data as well as for real time segmentation
- Temporal for executing the journeys made on the visual journey builder
- Kafka (hosted on Upstash) as the central event bus for the system
-
Event Streaming Backbone At the heart of our system lies a Kafka-based event streaming platform. This acts as the nervous system of LimeJourney, enabling:
- High-throughput processing of user and system events
- Decoupled communication between different parts of the system
- Reliable event sourcing for data consistency and replay capabilities
-
Real-time Segmentation Engine Leveraging ClickHouse's powerful querying capabilities, our segmentation engine allows for:
- Dynamic user categorization based on properties and behaviors
- Lightning-fast segment calculations, even at massive scale
- Real-time updates triggered by incoming events
-
Journey Orchestration Powered by Temporal, this component:
- Manages complex, long-running user journeys
- Ensures reliable execution of workflows, even in the face of failures
- Allows for pausing, resuming, and modifying journeys on the fly
-
Entity Management This part of the system handles all user-related data:
- Stores and retrieves user profiles and properties
- Emits events for entity changes
- Provides a unified view of the user across the platform
- User actions (e.g., page views, purchases) generate events that are published to Kafka.
- These events are consumed and stored in ClickHouse for large-scale data storage.
- The Segmentation Engine queries ClickHouse to update user segments in real-time based on the latest events and entity data.
- Changes in segmentation or specific events can trigger the Journey Orchestration.
- Journeys, managed by Temporal, orchestrate a series of actions over time, which may include:
- Querying ClickHouse for user data and event history
- Checking current segmentation
- Triggering external actions (e.g., sending emails, push notifications)
- Each significant action or state change generates new events, feeding back into the system.
- ClickHouse: Stores large-scale entity and event data, enabling fast queries for segmentation and user insights
- PostgreSQL: Handles relational data like journey definitions and system configurations
- Redis: Provides fast key-value store for managing journey<>trigger mappings
- Event-Driven Design: Enables scalability, flexibility, and real-time responsiveness
- ClickHouse for Large-Scale Data: Leverages ClickHouse's column-oriented structure for efficient storage and querying of high-volume entity and event data
- Modular Component Design: Allows for focused development and easier maintenance of different system aspects
- Workflow as Code: Temporal allows us to define complex journeys as code, simplifying management and versioning
We're working on comprehensive documentation/setup for self-hosting LimeJourney. In the meantime, you can get started with the following steps:
-
Clone the repository:
git clone https://github.com/limejourney/limejourney.git
-
Navigate to the project directory:
cd limejourney -
Copy the example environment file and update it with your configurations:
cp .env.example .env
-
Use Docker Compose to spin up the services:
docker-compose up -d
Or Deploy to Render:
Stay tuned for more detailed self-hosting documentation! If you are considering self-hosting, please reach out to me at [email protected] and we can help you get started.
We welcome contributions from the community! Check out our Contributing Guide for more information on how to get started.
Whether you're fixing bugs, improving documentation, or proposing new features, your efforts are appreciated.
Distributed under the AGPLv3 License. See LICENSE for more information.