Star us on Github!
Ready-to-use infrastructure for advanced product analytics

Open-source infrastructure
for product analytics modeling

Collect data designed for modeling and quickly build & run reusable models.

Built for those who like their data raw and models in code

Objectiv integrates data collection & modeling to give teams everything they need
to build and run state-of-the-art product analytics models in minutes.

Objectiv

DATA COLLECTION

Open Analytics Taxonomy

Give your datasets a generic & strict event structure designed for modeling.

Taxonomy
Lines left
Tracking SDKs

Tracking SDKs

Collect error-free user behavior data with validated instrumention.

Lines vertical
Data cloud

Data cloud agnostic

Run data models across data stores. Works with existing infra, no lock-in.

ObjectivLines vertical

DATA MODELING

Open Model Hub

Take pre-built models & functions for fast & efficient modeling.

Models
Lines right
Bach

Bach (modeling library)

Create models on your full dataset and reuse them on any project.

Objectiv

Data cloud agnostic
Run data models across data stores. Works with existing infra, no lock-in.

Data Collection

Always the right data for your models

Objectiv is built around an open analytics taxonomy: a generic classification of common event types and the contexts in which they can happen. It is designed and tested with UIs and product analytics use cases of over 50 companies.

Taxonomy overviewTaxonomy overview
Right data

No more back and forth on tracking plans

The taxonomy prescribes how to instrument your tracking, ensuring your dataset covers every common analytics use case. It is extensible to cover any custom requirements you may have.

Data designed for modeling

Work with data designed for modeling

Datasets that embrace the taxonomy are well-structured for modeling and contain the exact UI location for each event captured. You can work straight on the raw data without any prepwork.

Tracking

Tracking that keeps working, even if your product changes

The taxonomy is used to validate data collection at multiple stages of the pipeline, and our Tracking SDKs enable you to test & debug your instrumentation every step of the way.

Learn more about the open analytics taxonomyDocs - Open Analytics TaxonomyDocs - Open Analytics Taxonomy

Data Modeling

Take pre-built models and run them on the full dataset

Open your favorite notebook tool, import the open model hub, and use pre-built product analytics models & operations directly on the raw data to build your own analyses in minutes.

$ pip install objectiv-modelhub
Vertical lineSelection of models in the open model hubJupyter notebook with product analyticsJupyter notebook with product analytics
Right data

Reuse models between teams & platforms

Datasets that embrace the taxonomy are very consistent. As a result, models built on one dataset can be deployed and run on another without changing a single line of code.

Data designed for modeling

Runs straight on your data store

Objectiv's pandas-like modeling library runs SQL straight on your data store. You can use models in production directly to simplify data debugging & delivery to BI tools, dbt, etc.

Learn more about modeling with ObjectivDocs - ModelingDocs - Modeling

Supported platforms and technologies

Objectiv plays nice with most popular tools in the modern data stack.

FRONTEND

Frontend tech

DATA STORES / BACKEND

Data stores / backend tech

NOTEBOOKS

Modeling tech

OUTPUTS TO

Outputs
Anything that takes SQL or pandas as input
starThe Launchpad: a managed back-end and data store to simplify testing Objectiv. Learn morestar

See Objectiv in action

Integration between data collection & modeling enables you to build custom product analytics models in minutes. Check out these demo notebooks to see Objectiv in action.

Try the productDemo NotebooksDemo Notebooks

Join the product analytics modeling community

Our growing Slack community is a great place to learn & share about Objectiv and product analytics modeling. Join the discussion, talk to the team or stay in the loop.

Join us on SlackJoin us on Slack