Marketing ops · Production

Behind the Scenes of Canva's DesignDNA Campaign: Generative AI Delivers 95 Million Personalized Year-in-Review Experiences

The problem

Canva wanted to create a personalized year-in-review experience for millions of users but could not access personal design content due to strict privacy policies, and could not manually create content at the required scale.

Workflow diagram · grounded in source
1
Audience selection and consent
trigger
“we selected our target audience based on a minimum threshold of user design activity and engagement levels on Canva in the past year, and the user's consent for us to personalize our marketing communications to them”
2
User data retrieval
integration
“we could retrieve a user's locale, the total number of designs they created, and infer their top-created design type (for example, "Presentations")”
3
Template metadata inference
ai_action
“Canva has a vast library of public templates users use. These templates are tagged with style and theme metadata (you can even search by these in the Marketplace!). So, we used this template metadata to infer the user's top themes and st…”
4
Keyword-based design trend matching
ai_action
“We developed an algorithm to give each user a score for each design trend based on the keywords that matched the design trend from the templates they had used. The highest-scoring design trend would be considered the user's emerging desi…”
5
AI keyword expansion for unmatched users
ai_action
“we used generative AI to expand the set of keywords in each trend and select the most contextually relevant keywords from our curated list. Using this approach, we matched 99% of the users in our target audience to a design trend.”
6
Design personality generation
ai_action
“We used Magic Write for each segment to define a design personality name and description, translating the content for different locales using AI. We used Canva's Dream Lab to generate a hero image that aligned with the content of the des…”
7
AI poem generation at scale
ai_action
“We created unique prompts for each locale and then provided the top 3 styles we wanted to generate a poem for. The result was a million generated poems across 9 different locales.”
8
Human and AI poem review
human_review
“Asking our Localisation team to review a sample of poems generated in non-English locales. We took their feedback into account to help fine-tune our prompt for generating the poems. - Flagging poems containing potentially sensitive words…”
9
Personalized DesignDNA URL delivery
output
“For each user, we added the tailored content to the URL and produced a link the user could access that would generate their personalized copy of the DesignDNA”
Reported outcome

Canva delivered 95 million unique DesignDNAs, matched 99% of its target audience to a personalized design trend, and generated over a million poems across 9 locales using generative AI.

Reported metrics
unique DesignDNAs created95 million
Users matched to design trend (keyword matching)95%
users matched to design trend (after AI expansion)99%
Poems generateda million
Show all 5 reported metrics
unique DesignDNAs created95 million
users matched to design trend (keyword matching)95%
users matched to design trend (after AI expansion)99%
poems generateda million
locales covered9
Reported stack
Dream Labgenerative AI
Source
https://www.canva.dev/blog/engineering/behind-the-scenes-of-canvas-designdna-campaign/
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Canva delivered 95 million unique DesignDNAs, matched 99% of its target audience to a personalized design trend, and generated over a million poems across 9 locales using generative AI.

What tools did this team use?

Dream Lab, generative AI.

What results were reported?

unique DesignDNAs created: 95 million; Users matched to design trend (keyword matching): 95%; users matched to design trend (after AI expansion): 99%; Poems generated: a million (source-reported, not independently verified).

How is this marketing ops AI workflow structured?

Audience selection and consent → User data retrieval → Template metadata inference → Keyword-based design trend matching → AI keyword expansion for unmatched users → Design personality generation → AI poem generation at scale → Human and AI poem review → Personalized DesignDNA URL delivery.

Related marketing ops cases