Case Study: Kami scales from 80M to nearly 2B weekly events with ClickHouse

How Kami scaled from 80M to nearly 2B events per week with ClickHouse

Kami, a New Zealand-based edtech company, needed a new way to handle explosive event-tracking growth when the COVID-19 pandemic pushed weekly volume from about 80 million events to nearly 2 billion. Its Postgres-based analytics system, used to track feature engagement and user behavior, became too slow to query and was no longer practical for reporting or product insights.

ClickHouse provided the solution, with Kami migrating from Postgres to a Kafka-to-ClickHouse pipeline built to ingest and analyze data at scale. The new ClickHouse-powered stack lets Kami query 140 billion+ rows in minutes instead of hours, and their dbt-powered aggregations cut DAU query time from 3–4 minutes to 1 second on 1.3 billion rows.

Kami

Jordan Thoms

Co-Founder, CTO