Behind the scenes of Canva's DesignDNA campaign

Summary

Canva launched DesignDNA in December 2024 as a personalized year-in-review campaign to highlight user achievements and showcase generative AI capabilities. Because internal privacy rules strictly prohibited inspecting personal designs, the engineering and creative teams inferred user preferences from the style and theme metadata tags on public templates. An initial keyword-matching algorithm paired 95% of users with one of seven emerging design trends, while generative AI keyword expansion increased coverage to 99%. The team also synthesized over one million localized poems and distinct design personalities using tools like Magic Write and Dream Lab. Finally, dynamic Canva template elements populated through URL parameters assembled 95 million distinct multi-page stories across nine locales.

Context

Canva wanted to build DesignDNA, a personalized year-in-review campaign for millions of users celebrating their achievements while showcasing generative AI capabilities. However, Canva's privacy policies strictly prohibited accessing personal user design content to analyze styles and themes, requiring alternative methods to understand user behavior and generate personalized marketing stories at scale.

Approach / What changed

The team leveraged public template metadata to infer top styles and themes from user activity. They matched users to seven design trends using keyword matching and genAI keyword expansion, achieving 99% coverage. For design personalities and poems, they categorized users into audience groups, generated copy via Magic Write, produced hero imagery using Dream Lab, and created over one million localized poems across nine locales with automated tone and safety reviews. Completed stories were delivered dynamically by populating tagged Canva template elements via URL parameters, generating 95 million unique DesignDNAs.

Takeaways

• To protect user privacy without inspecting personal design content, Canva inferred individual styles and themes using the metadata tags of public templates users interacted with.
• Combining direct keyword matching with generative AI keyword expansion increased user-to-trend matching coverage from 95% to 99% across the target cohort.
• Canva produced personalized stories at scale by dynamically replacing tagged template elements with user-specific data passed through URL parameters across 95 million generated DesignDNAs.

Related reading

• Canva · Jun 20, 2025

  Measuring Commercial Impact at Scale at Canva

  Canva needed a scalable, standardized way to calculate the commercial impact of thousands of annual experiments on key business metrics like Monthly Active Users and Annual Recurring Revenue. Previously, disparate teams spent over six hours per experiment performing manual, error-prone calculations across fragmented spreadsheets and inconsistent data models. To resolve this, Canva built the IMPACT app using Snowflake, Streamlit, Snowpark, and Cortex to provide a self-serve platform tied directly to its central finance model. The application scales local uplift by actual audience exposure and supports pre-experiment scenario modeling alongside post-experiment tracking. A custom deployment workflow generates pull-request-isolated Streamlit environments in Snowflake stages, reducing time-to-insight to under ten minutes while enabling multiple developers to safely build and demo features in parallel.

  Jun Ye

• Canva · Jan 28, 2025

  Image replacement in Canva designs using reverse image search