Case Study Deskcasestudydesk.com
Design SoftwareSourced

Case Study: Canva built user-facing analytics 10x cheaper and 5x faster with Tinybird

Canva Case StudySourced & dated by Case Study Desk
Key facts · TL;DR
Company
Canva
Industry
Design Software
Challenge
Warehouse couldn't power user-facing analytics
Headline result
10x lower compute cost

Key results

10x
Lower compute cost
vs. prior warehouse approach
25,000x
Faster end-to-end latency
Monthly batch to real time
<50ms
p99 API latency
1.4PB processed per month
8x faster
Development cycle
MVP in 2 weeks vs. 4 months

The challenge

Canva ran a warehouse-centric data platform that could not meet the latency and concurrency requirements for user-facing analytics at reasonable cost. It needed an engine purpose-built for real-time, in-product analytics, separate from its business intelligence and data science workloads.

The solution

Canva moved its user-facing analytics use cases to Tinybird, ingesting large event volumes and exposing low-latency SQL-based API endpoints to power in-product Insights features for creators.

When we moved those use cases to Tinybird, we shipped them 5x faster and at a 10x lower cost than our prior approach.

GN
Guy Needham
Staff Backend Engineer, Canva

The results, in context

Canva reported 10x lower compute costs and shipped user-facing analytics 5x faster than its prior approach, cutting end-to-end latency 25,000x from a monthly batch to real time. It held sub-50ms p99 API latency while processing 1.4PB per month and 31M requests per day, and shipped 7 new features in a year.

Products used

Tinybird Tinybird