Codie Sanchez spent her authority building a paid curriculum out of a free newsletter. Alex Hormozi’s got formally separated from the company whose name he built. Whitney Wolfe Herd is spending hers tearing down the two features that made Bumble famous, in public, while the stock keeps falling. Melanie Perkins is spending her authority on something none of them tried: asking strangers what she should do with it.
In July 2026, Canva’s billionaire co-founder and CEO launched a global survey asking ordinary people — in Wollongong, Australia, and across the United States — to name the one goal they’d most want achieved in their lifetime, for their community, their country, or the world. The stated purpose was to help guide where she and her husband direct a multibillion-dollar fortune they’ve already pledged away. It’s a genuinely unusual mechanism: most philanthropists decide privately, then announce. Perkins is trying to crowdsource the decision itself, in public, before making it. The problem is the fortune in question is denominated in private-company stock — and in the same several months, that stock’s own backers marked it down by billions, and the company that issues it got fined for failing to do the regulatory paperwork that would let anyone outside Canva actually check the numbers.
The pledge
Perkins and Cliff Obrecht, her husband and Canva’s co-founder, signed Bill Gates and Warren Buffett’s Giving Pledge in 2021, committing — per Forbes’s reporting on the signing — to donate more than 80% of their wealth through the Canva Foundation, which they’d established the previous year. At the time, that wealth was substantial on paper: Startup Daily reported that, following Canva’s 2021 funding round, the couple held a 30% stake in a company then valued at roughly $55 billion, putting their paper stake at more than $16 billion.
The founding story behind that fortune is itself well-documented and often repeated in Perkins’s own telling: she and Obrecht, together with Cameron Adams, built an earlier product called Fusion Books — a simpler yearbook-design tool — before founding Canva in 2013. Fortune’s longform profile of Perkins describes her being turned down by roughly 100 investors while pitching Canva with no existing Silicon Valley network, a number Perkins herself has repeated in interviews. Canva went on to become, by Forbes Australia’s July 2026 accounting, a $42 billion private company — one of the most valuable startups to come out of Australia.
The question
The first concrete step toward spending down that pledge arrived five years after the couple signed it. On July 16, 2026, in an exclusive interview with Forbes Australia, Perkins and Obrecht announced the Global Goals Platform — an initiative funded through the Canva Foundation that asks anyone, anywhere, to answer a single question: what is the one goal you’d like to see achieved in your lifetime, for your community, your country, or the world. “We have a very unifying objective, which is to empower people,” Perkins told Forbes. The idea, as Forbes described it, is to let the answers — not a private decision inside the foundation — help determine where the couple’s giving goes next.
Forbes’s July 2026 figures put real numbers on the pledge for the first time: Perkins and Obrecht each hold an estimated 18% stake in Canva, worth $7.6 billion apiece at the company’s $42 billion valuation, and the couple’s commitment specifically covers eventually giving away a 30% stake — worth roughly $12.6 billion at that same valuation. The foundation’s existing giving, separate from that eventual equity transfer, already runs into nine figures: more than $80 million donated through 2025, including a $150 million, multi-year commitment to GiveDirectly’s cash-transfer program in Malawi (more than $50 million of it already disbursed to over 130,000 people), plus more than $2.5 billion in Canva product donated to students, teachers and nonprofits, per Forbes’s reporting.
The Global Goals Platform launched with two pilots. In Wollongong — a city of roughly 220,000 people about 90 minutes south of Canva’s Sydney headquarters — the council partnered with Canva to invite residents to submit goals online or by postcard starting July 24, according to regional outlet Region Illawarra; Wollongong’s Lord Mayor, Tania Brown, framed the goal as reaching people who don’t usually speak up at public meetings, including young people and newer migrant communities. A parallel US pilot, called USA Goals, paired open public participation with a nationally representative research sample run with the Ad Council. Both were due to feed into a findings report later in 2026.
What the data actually showed
Participatory initiatives like this invite an obvious question: participatory among whom? Canva’s own team answered it candidly, if quietly. In a September 6, 2026 op-ed for the Illawarra Flame, Canva’s global chief marketing and communications officer, Zach Kitschke, wrote that “so far, 80 per cent of goals have come from 10 suburbs” — with Wollongong, Thirroul and Dapto leading in raw submissions, and Mount Kembla, Otford and Haywards Bay trailing. “Too often,” Kitschke wrote, “the future of a city is shaped by the people who have the time, confidence or connections to speak up.” The Illawarra Mercury’s own news coverage reported the same split independently. It’s a real admission, made by the company itself rather than extracted by a critic: the platform built to surface voices that don’t usually get heard was, months into its pilot, dominated by the same kind of well-connected participation it was designed to counteract. Kitschke’s framing treated that as a reason to keep building outreach rather than a reason for doubt — but the number is Canva’s own, not a rival’s.
The same months, a different kind of disclosure
While Perkins was publicly asking the world to help decide the fate of her fortune, the company that fortune is built on was failing a much more basic public obligation. On May 6, 2026, Australia’s corporate regulator, ASIC, issued four infringement notices totaling $792,000 — $198,000 each to Canva Pty Ltd, Canva Operations, Canva Trading and Fusion Books — after Canva Pty Ltd filed its consolidated FY24 financial report on March 27, 2026, roughly 11 months after its April 30, 2025 deadline, according to The Nightly’s reporting. ASIC Commissioner Kate O’Rourke said in the announcement that non-lodgment and late lodgement “prevents creditors and other users of the reports from making timely and informed” decisions — part of what ASIC described as a 2026 enforcement priority that produced 21 infringement notices worth more than $4 million across multiple companies. Canva’s payment of the fines is not, per ASIC’s own notice, an admission of guilt or liability, and the company has said it has since strengthened its reporting processes.
Three months later, the number itself started moving. Per the Australian Financial Review’s reporting on August 3 and 14, 2026 — summarized by The Next Web and independently corroborated by Startup Daily and SmartCompany — Canva cut its own revenue-growth forecast by a third, to 20%, after discovering the AI features it had aggressively rolled out cost far more to run than management had projected. Canva’s backers Blackbird and AirTree marked their stakes down 17%, from $42 billion to $34.9 billion — a roughly $7.1 billion cut — while Canva’s own internal valuation fell even further, from $38.9 billion to about $31 billion, a bigger write-down than its investors took. The company’s revenue kept growing through all of it — the June quarter brought in $921.9 million, up 25.2% year-over-year — but missed Canva’s own internal guidance, per the AFR’s reporting.
Perkins addressed the AI cost problem directly in emailed comments to Fortune, saying the rollout “validated the demand, but also showed us we needed to reduce the cost of completing an AI task to support a broad rollout,” and that cost per task had fallen nearly 90% since Canva AI 2.0 launched that April. PitchBook analyst Derek Hernandez told Fortune the pressure wasn’t unique to Canva — “Canva and Figma both hit the same wall about five days apart,” he said — and pointed to the same dynamic pushing Canva’s long-rumored IPO toward 2027 rather than 2026: “I’m sure they’re trying to protect their profitability, especially if they want to go to public investors.” That lines up with what Canva’s own COO, Cliff Obrecht, told Capital Brief back in April 2026 — “from all metric accounts, we are fully IPO ready,” he said, while adding “I assume it will be next year” and that the company wouldn’t list until its shift from subscription tiers to an AI-credits business model was “really bedded in.”
Why the sequence is the story
None of these are hidden facts — ASIC’s fine, the AFR’s valuation reporting, and Forbes’s interview all ran in mainstream, on-the-record coverage, several of them with Canva’s own cooperation. The sequencing is what makes it worth noticing. In roughly four months in 2026, Melanie Perkins’s company was fined for failing a basic statutory filing obligation, had its valuation marked down by close to a fifth by the investors who’d have to eventually cash out of it, and pushed its public listing to a year most analysts weren’t expecting — while Perkins herself spent that same window building and launching an elaborate, deliberately public mechanism for deciding how to give away a fortune denominated in exactly that stock.
It isn’t a contradiction in the sense of dishonesty; nothing here suggests Perkins or Canva misrepresented anything. It’s a contradiction of register. The Global Goals Platform is built on the premise that a careful, transparent, data-driven process — survey pilots, a published report, an acknowledged participation gap — produces a better answer than one person deciding alone. The ASIC fine exists because Canva, for nearly a year, didn’t clear the much lower bar of filing a report on a legally mandated schedule. Perkins is asking the public to help calibrate the spending of a $12.6 billion pledge with more process than her own company applied to telling its shareholders what it was worth. Whether that’s a founder building unusually durable trust before the market gets to test her stock price directly, or a founder’s authority running slightly ahead of the paperwork behind it, is exactly the kind of question the Global Goals Platform’s own report — due before the end of 2026 — won’t be able to answer about itself.
