Apache Unomi vs Tracardi

Apache Unomi is a Java server for managing customer, lead and visitor data and personalizing customer experiences. It tracks users, builds profiles progressively and associates events with those profiles. Its rules can respond when profile events are collected, while dynamic segments classify profiles into groups. The privacy REST API lets integrators provide controls for tracking, data review, anonymization and deletion, but Unomi has no default interface for these controls, so developers must expose them. Unomi runs as an OSGi application inside Apache Karaf, uses Elasticsearch for storage and communicates through HTTP REST requests with JSON. Its plugin system supports conditions, actions, beacon tracking and push notifications. Listed uses include website personalization, mobile app data collection, store beacon connections and sending scored leads to a CRM. The project names CMSs, CRMs, issue trackers and native mobile apps as integration targets. It is self-hosted, with Docker images and a Java quick-start guide; the project describes installation as unzip and run. Apache Unomi is an Apache open-source project under the Apache License v2 and costs nothing.

Tracardi is an open-source, composable, API-first customer data platform engine for building a tailored CDP. It collects behavioral customer data from websites, mobile apps, web services, CRMs, e-commerce shops, CSV files, APIs, queues and databases. It can merge that information into unified profiles and automatically deduplicate customer records. Journey automation supports data collection, engagement, messaging, targeting, recommendations and reminders. Documentation lists integration resources including HubSpot, Salesforce, Mailchimp, Mautic, Mixpanel, Airtable, RabbitMQ, Slack, Discord and Telegram. Tracardi does not send marketing campaigns, but it can gather consent and subscription data and connect marketing data with other platforms. Deployment documentation describes open-source and commercial Docker installations, plus Helm-based production installation. The production guide lists Redis, Elasticsearch, Apache Pulsar and MySQL as dependencies, and recommends encrypted API connections, separating event collection and GUI APIs, and restricting GUI access to trusted networks. The free open-source plan is limited in event rate, tenancy, automation and event validation. Commercial enterprise and multi-tenant options require contacting the vendor; the latter includes installation and integration support.