The AI design giant Canva just announced the first-ever major round of layoffs in its 13-year history. As part of the restructuring, 1,271 people—around 22% of the company’s overall staff—will either be laid off, asked to change roles, or relocate to the company’s headquarters in Sydney, Australia.
News of the layoffs was announced on November 3 (a perfect day to bury some bad news) in a press release from Canva cofounder and CEO Melanie Perkins. In the note, Perkins explains that the layoffs are part of a larger plan to make Canva’s teams more agile and less siloed, including by folding product, growth, and platform teams into one unit and reducing the number of “subspecialities” at the company.
“I’ve come to the difficult conclusion that small changes won’t be enough,” Perkin’s letter reads. “To set ourselves up for the next decade, we need to refound Canva, getting back to our roots of small, empowered teams with the freedom to make decisions, take ownership and focus on what matters most for our community.”
While Canva didn’t provide a broader business reason for the layoffs, the restructuring comes at a time when it’s working through some mounting growing pains and questions about the timing of a potential IPO. After spending the last several years going all-in on its AI-powered product suite, Canva is now facing the challenge of squaring its business model with an increasingly voracious user base.
Canva’s growth conundrum
In May, Fast Company wrote that design was entering its “frenemies” era.
At the time, Anthropic had just launched Claude Design . The product seemingly came as a surprise to incumbent companies including Adobe and Figma, which had been operating under a multiyear collaboration with Anthropic to integrate Claude into their products, and didn’t expect the AI giant to debut its own design competitor. Meanwhile, Canva told Fast Company that it had actually helped Anthropic develop Claude Design, despite Canva’s own collaboration deal with the company.
For Adobe, Figma, and Canva, working with frontier models offers the advantage of bringing new tools to market faster and benefiting from the sophistication of frontier models. The main downside is that using these frontier models also takes a lot of computing power, and that racks up major costs. As Canva has made a major push —by building its own models and acquiring AI companies—to become an “AI platform” over the last several years, those costs have become a significant problem.
In early August, Canva sent a note to its investors with an update: the company was downgrading its revenue growth forecast for 2026 from 30% to 20%. At the time, Perkins told shareholders that the “average cost of serving an AI task was too high” as the company relied heavily on its partnerships with frontier models to complete users’ requests.
For context, by the end of 2025, Canva had more than 265 million monthly active users and more than 31 million paid users, meaning that some large portion of AI traffic on the site was likely generated by free users. In an August interview with Fortune , Perkins said that users’ demand for new AI features had “significantly exceeded” the company’s expectations, which came with both pros and cons.
“This validated the demand, but also showed us we needed to reduce the cost of completing an AI task to support a broad rollout,” Perkins told the publication. “Rather than broadly rolling out a product before the underlying economics were ready, we decided to slow the rollout while we rebuilt the architecture, reduced unit costs and strengthened the business model.”
To reduce the intermediary costs that Canva incurs every time an AI task requires a frontier model, Canva has been working on developing first-party tools that can take up some of the work. For example, Perkins told Fortune that, since launching Canva 2.0 in April, the company’s cost-per-task has reduced by 90%. This October, Canva launched another product called Canva Brush, which it said reduced cost-per-task by 46% compared to the previous design agent.
These new features have made improvements, but, based on its downgraded revenue forecast, Canva still has a ways to go before it fully irons out its economic model. The fact that user demand for Canva’s AI tools is outpacing its own capabilities seems to be a bullish sign for AI design at large. Now, Canva needs to figure out how to make the most of the surge—and the company’s first major layoffs could be part of its plan to get back on track.
