Case Study: Knock cuts data streaming costs 77% with WarpStream

KNOCK cuts data streaming costs 77% with WarpStream

The customer, Knock, an agent-native customer engagement platform, faced high infrastructure costs and operational burdens from their Amazon Kinesis data streams. Their Elixir-based services relied on a problematic community driver for Kinesis, which led to a complex and expensive "manifold" architecture for fanning out events. This setup was a major source of wasted spend and occasional inconsistencies, prompting them to seek a new solution from vendor WarpStream.

WarpStream provided a drop-in, Kafka-compatible replacement built on object storage. The migration to WarpStream was completed in a month by one engineer, moving five topics and roughly 20 consumers without service disruption. The solution delivered a 77% reduction in annual data streaming costs, from $112,000 to $25,000. It also simplified their architecture with native Kafka features like consumer groups and fan-in, improved operational reliability, and streamlined their privacy compliance by keeping data in their own AWS account.

Knock