• Canva’s user base tripled to 200M, driving demand for scalable product analytics. • Addressed cost, speed, and correctness of data insights for product decisions. • Implemented schema management to improve data correctness across analytics. • Shifted from AWS Firehose/S3 to Snowflake Snowpipe Streaming for real‑time ingestion. • Snowpipe Streaming reduces small‑file problem and cuts ingestion costs. • Continuous improvements enable faster, cheaper, and more accurate product decisions.
Article Summaries:
- snowflake The foundations of Canvaâs continuous data platform with Snowpipe Streaming Leveraging Snowpipe Streaming to build a continuous data platform. In a previous blog post, we discussed the important role product analytics plays in helping teams measure the impact of their product features and, subsequently, make data-driven decisions. Over the past 5 years, Canvaâs monthly active users have more than tripled from 60 million active users to over 200 million. In that time, Canva has also grown to over 4500 employees. As a result, the multiplicative effect of more questions from more intern
- Canva, whose monthly active users have tripled to over 200 million, has revamped its product‑analytics data pipeline to meet growing scale and cost pressures. The company moved from AWS Data Firehose-whose 5 KB minimum record size drove up costs-to Snowpipe Streaming on Snowflake. This change eliminates the Firehose billing model, improves data ingestion speed, and supports auto‑scaling without manual support tickets. Alongside better schema management and streamlined event collection, the new platform delivers faster, more accurate analytics while cutting the data‑landing cost that previously accounted for nearly half of Canva’s analytics spend.
Sources: