The Long Way to an Obvious Idea

In 2007, Melanie Perkins was teaching design programs to students at the University of Western Australia. The work revealed a peculiar mismatch. The students wanted to communicate an idea, assemble a yearbook page, or make something visually convincing. The software demanded that they first learn its internal logic: layers, tools, file formats, menus, and procedures accumulated for professional users over many years. Their first hour belonged to the software rather than to the thing they wanted to create.

Perkins read this not as a training problem but as a design failure. Powerful software had made expertise in the interface a prerequisite for expression. The emerging web suggested another arrangement: creation could be collaborative, assets could be searchable, and users could manipulate a page directly. Her ambition quickly became enormous—make design accessible to everyone—but an enormous ambition is not yet an investable company or a usable product.

Perkins and Cliff Obrecht began with school yearbooks. Fusion Books gave them a contained market with painful existing workflows and a deadline no customer could ignore. Schools needed many students and teachers to contribute photographs and pages, but desktop files and specialist software made that collaboration awkward. A browser-based system with templates and drag-and-drop editing could replace coordination work as well as technical design work.

The founders did nearly everything: contacting schools, building the product, answering support requests, printing books, and shipping them. The business spread across Australia and later to other markets. Every finished yearbook supplied a more valuable form of evidence than a slide about democratizing design. Non-designers could make complicated visual artifacts together online, and they would pay for a workflow that made the result attainable.

Fusion Books proved the interaction but not the universal platform. Yearbooks had known dimensions, recurring layouts, and a predictable annual cycle. Canva would have to support a shifting universe of presentations, posters, invitations, social posts, marketing materials, and documents. It needed a flexible editor that felt simple even when the possible outputs multiplied. That required more engineering, design leadership, capital, and credibility than the Perth team initially possessed.

Perkins and Obrecht moved the center of their effort toward Sydney and repeatedly traveled to the United States. The famous count of investor rejections compresses this period into a test of stubbornness. In reality, each meeting exposed a different missing bridge: the market sounded too broad, the founders were geographically distant, the product was not yet complete, or the team lacked the senior technical counterpart investors expected for such an ambitious platform.

Perkins kept revising the story without shrinking its central claim. She sought proximity to the networks making technology bets, including the gatherings around investor Bill Tai. Lars Rasmussen, known for co-creating Google Maps, became an adviser after seeing the product vision. His involvement supplied interrogation, credibility, and introductions. A pitch deck designed to raise money was also being used to recruit the people who could make the company fundable.

One of those people was Cameron Adams, a designer and engineer with deep product experience. He joined Perkins and Obrecht as co-founder. The trio's roles were complementary rather than ornamental: Perkins held the sweeping product and company vision, Obrecht helped drive commercial and operational execution, and Adams could lead the technical realization of a browser-based visual platform. Canva emerged when the thesis, evidence, and team finally aligned.

Canva opened to the public in 2013. Its decisive choice was not merely putting design in a browser. It changed the user's starting condition. Instead of an empty canvas surrounded by professional controls, people could begin with a template that already contained hierarchy, proportion, and visual logic. They could replace words and images, search for assets, move objects directly, and produce something credible before understanding the full system.

Templates are sometimes described as creative limitations. For Canva's users they were compressed expertise. A restaurant owner did not need to study typography before making a menu; a teacher did not need to become an illustrator before making a worksheet. Good defaults converted the intimidating question—how do I design?—into a sequence of manageable decisions about this design.

A freemium model removed the purchasing decision from the first experience, but the created work carried distribution further. Presentations were shown to teams. Invitations were sent to communities. Shared designs brought collaborators into the editor. Teachers introduced students, marketers introduced colleagues, and brand teams created reusable templates for entire organizations. The unit of adoption was often not an individual account but a piece of work that needed other people.

This explains why the product crossed categories without a conventional top-down sale for every user. Visual communication appears across departments and life events. Once the editor became familiar, the next format was easier to attempt. Canva could add print, video, websites, whiteboards, documents, and enterprise controls around an interaction model people already knew.

Ease at the surface required complexity underneath: rendering, search, licensing, storage, collaboration, internationalization, payments, moderation, and reliable output across devices and formats. Canva opened a Manila office early in its expansion and localized the product across languages. By 2018 it reported profitability, a US$1 billion valuation, support for 100 languages, and reach across 190 countries. The simple editor had become a globally operated content system.

Capital helped the company accelerate, but its history was not a straight line from rejection to a US$40 billion private valuation in 2021. Valuation marked an investor price at a moment of unusually exuberant software markets; it did not settle the quality of every later decision. The stronger record is operating adoption: people returned because the product repeatedly shortened a task they already needed to complete.

By 2024 and 2025, Canva was moving beyond its original positioning. It acquired Affinity, expanded enterprise controls, and embedded generative AI across its suite. The company said it reached 260 million monthly users and US$3.5 billion in revenue in 2025. It described its integrated product direction as a Creative Operating System and relaunched Affinity as a free professional application.

The expansion is strategically coherent—more of the creative workflow can happen in one environment—but it reopens the problem Perkins first identified. Comprehensive software tends to accumulate friction. Serving professional designers, global brand teams, classrooms, and first-time creators inside one ecosystem requires boundaries as much as features. Canva's next test is whether it can become more capable without making the user's first hour belong to the software again.

The easiest lesson from Perkins's story is to persist after rejection. That lesson is true and incomplete. Persistence without improving evidence is repetition. Fusion Books changed the quality of the claim. The move to Sydney changed access. Adviser relationships changed the network. Adams changed the founding team's capability. Templates changed the user's starting point. Each step removed a specific reason the larger vision might fail.

Canva was not built because one founder believed harder than everyone else. It was built because Perkins protected a clear human problem while changing nearly everything required to solve it at scale. The useful founder pattern is not refusal to listen. It is learning to separate criticism of the route from abandonment of the destination.

Canva's acquisition of Affinity in 2024 complicated the old division between accessible design and professional software. Affinity's photo, vector, and publishing tools served creators who wanted precision and local performance without Adobe's subscription model. Bringing that product into Canva expanded the mission from helping the 99% without formal design training toward supporting the professionals whose work often defines design standards.

In 2025 Canva relaunched Affinity as a free unified application. The move can enlarge the professional community and create a route into Canva's broader ecosystem, but it also asks whether a free tool can sustain the depth and trust professionals require. The answer will depend on roadmap, file compatibility, performance, and whether Canva respects workflows that should not be simplified into templates.

The archive cannot end at the triumphant valuation. Canva is now an employer of thousands and a critical tool for schools, nonprofits, companies, and creators. Product changes can alter established workflows; AI features raise questions about creative labor and training data; organizational restructuring affects people who built the platform. Scale converts founder values into policies with material consequences.

That does not invalidate the origin. It completes it. Perkins wanted to remove unnecessary difficulty from visual communication. The mature company's responsibility is to keep identifying which complexity protects power and which merely transfers cost to users, contributors, or employees. Democratization is not a launch achievement. It is a standard Canva has to meet again whenever the technology changes.