Case Study: PushOwl achieves 40x faster analytics and sub-40ms response times with Citus Data

PushOwl Uses Citus Cloud to Scale Out Postgres & Speed Up Analytics Queries

PushOwl is a Shopify-focused push-notification app that helps merchants recover lost sales and track campaign performance with a real-time analytics dashboard. Rapid growth—from about 3,000 to 6,500+ stores—exposed a bottleneck in their single-node Heroku Postgres setup, with dashboard queries taking up to eight seconds and threatening user experience and future scalability.

PushOwl migrated to Citus Cloud, using its Django multi-tenant library and Citus team support to keep refactoring minimal and enable horizontal sharding by tenant. The move cut complex query times by 40x (from ~6s to ~200ms), reduced 95th-percentile response times to 30–40ms, increased request throughput (25–30 → 60 req/s), boosted monthly notifications from 6.5M to 10M, eliminated the need for an immediate DevOps hire, and now supports adding 50–100 new customers per day.

PushOwl

Shashank Kumar

Founder