Open-source product analytics infrastructure with a generic event taxonomy
Use Objectiv to capture validated user behavior data straight into your data warehouse,
and speed up product analytics projects with pre-built & reusable models.
Capture validated user behavior data &
feed it straight into your data warehouse
Eliminate complexity with a tracker that feeds high quality user behavior data directly into the
heart of your product analytics stack. No cleaning, transformations or tracking plans required.
A typical data collection workflow before using Objectiv
A typical data collection workflow after using Objectiv
A taxonomy to ensure quality & consistency
Objectiv's tracker validates all incoming events against an open analytics taxonomy. This ensures it is well-structured, clean and ready for modeling.
It describes classes for common user interactions and their contexts. A tracking plan is no longer needed as the requirements for effective analysis are carried by the design of the taxonomy.
Get the full context
Objectiv's tracker captures the structure of your product's UI inside the dataset. Events contain the exact location where they were triggered in a hierarchical stack of locations.
This not only makes events easily identifiable, it also enables data slicing on a very granular level without doing a ton of manual mapping first.
Speed up product analytics projects
with pre-built & reusable models
Take granular control over your data with pre-built models that run on the full dataset.
Share & reuse any model and convert them to SQL with a single command.
A typical modeling & analysis workflow before using Objectiv
A typical modeling & analysis workflow after using Objectiv
Pandas-like modeling on the full SQL dataset
The Objectiv Bach modeling library combines the scalability of SQL with the agility of Pandas.
You can build models using dataframes and pandas-like operations and run them on the full dataset as SQL. If you know Pandas, you'll feel right at home.

Take pre-built models off the shelf
Objectiv includes pre-built models for a wide range of product analytics use cases. You can chain them together to answer common product analytics questions quickly.
You're free to customize them (or build your own) for specific in-depth analyses.
Reuse anyone's models
High data consistency means models and datasets are intercompatible and can be shared and reused.Pandas compatible
Pandas compatibility enables you to tap into the rich ecosystem Pandas is well-known for, including all ML libraries.Works with the open taxonomy
Bach includes operations that are specifically designed to effectively work with datasets that embrace the open analytics taxonomy.Optimized for machine learning
A number of built-in optimizations for popular libraries, like scikit-learn, will enable you to incorporate ML into your analyses faster.Run your entire product analytics
workflow from a notebook
Make your notebook the headquarters of your product analytics operations.
Convert models to SQL with a single command
On command, Objectiv converts your entire model to a production-ready SQL query, which you can directly use to feed into your tools and products.
Combine this with the fact that raw Objectiv data is model-ready straight from the tracker, and you have a very efficient workflow that enables you to experiment freely without the typical overhead.
As a result, you can adapt to changing product questions much faster and keep all your product analytics projects in one place.

To simplify sharing insights with other team members, Objectiv comes with
built-in integration for the open-source BI platform Metabase.
What's in the box?
Objectiv is open-source and self-hosted. It includes what you need to answer common product
analytics questions fast and accurately, while providing a solid foundation for advanced modeling.

Run it locally or in the cloud for reliable event handling.

Plugs into your Snowplow backend for event handling at scale with BigQuery.
Try the Objectiv local demo
Follow the Quickstart Guide to run a fully functional Objectiv setup locally.