In 2018, Microsoft News leadership gave me a directive with almost no shape to it: use our content to do some good. There was no budget, no team and no existing playbook to follow. What existed was a news platform with enormous reach and zero infrastructure for turning that reach into charitable action. I was asked to figure out what “doing good” could actually mean for a news product, then build it.
Building the infrastructure
Before any campaign could run, I had to build the scaffolding that would support them: an editorial calendar mapping planned activation moments, relationships with content syndication partners, vetted nonprofit beneficiary relationships alongside a partnership with Microsoft Philanthropies, a donation processing integration (Benevity) that made giving possible directly from an article, and the Local Hero of the Month program, which spotlighted grassroots causes every month and rewarded the honoree with a Surface Go. I wasn’t running one campaign. From the first campaign, the underlying structure was built with scale in mind.
The craft behind the system
The system only worked if the words did. A donation widget with the wrong ask got ignored; a Local Hero feature with a flat headline didn’t get shared. Each form factor needed its own customized strategy.
01/
Donation widget copy. Short and specific, leading with a match incentive or a concrete outcome rather than a generic ask — and once a campaign crossed a threshold, showing the live fundraising total for its own motivating effect.
02/
Social copy. Paired a factual hook with human interest and a direct donation link, written for a general news audience rather than a cause-marketing one.
03/
Blog writing. Campaign recap posts used a longer narrative and a more personal tone than a standard article, creating a throughline across a campaign’s life and a place to celebrate what readers had made possible together.
04/
Content curation. Historical engagement data informed which story angles would resonate for a given cause, and which content partner was the right fit to tell it.
First proof points
Notre Dame Cathedral restoration raised $45K, including $40K in a single day, our fastest fundraising moment to date. Earth Day 2019 was our first matching campaign, with Microsoft matching $50K in donations to fund 50,000 trees planted. These moments mattered less for the dollar totals than for what they proved: readers would act on a donation ask embedded directly in editorial content, which was the whole bet behind the infrastructure.
Scaling the playbook globally
Once the model was proven, the work became about repeatability: turning what worked in one market into something any market could run. I built a global playbook and campaign calendar template, including editorial voice and CTA-writing guidance alongside the operational logistics, then trained a virtual team of market leads who used it to launch localized campaigns in their own languages and with their own local nonprofit partners. Within a year, the program grew from a single US market to 40 giving campaigns across 15 or more global markets.
The donation infrastructure problem: what I could not standardize was the money itself. Every market’s donation processing depended on Benevity, our third-party platform, and its global support was inconsistent in ways that varied by country: missing currencies, formatting bugs, mixed-language widgets and no country-level reporting. In markets without Benevity support at all, including Brazil, Mexico, France and India, we ran donation campaigns as an untracked “direct model,” linking out to a partner’s own site instead. I wanted a single global solution to this. I never found one, and by the end of the program, donation processing was still being solved market by market.
Crisis at scale: Australia bushfires
The infrastructure met its first real stress test in the 2019–2020 Australia bushfires. What started as a single-market disaster response scaled to 24 participating markets within weeks, raising $774K worldwide: the clearest evidence that the playbook and the market-lead network I’d built could mobilize fast, under pressure, without me personally running each market’s response.
Adapting to an unprecedented crisis
COVID-19 broke the assumptions the whole program had been built on. Campaigns had always been calendar-driven, tied to a known date or a bounded disaster; COVID had no endpoint and no fixed shape. Donation appeals had to shift as the virus itself became better understood and spread to new regions, multiple markets needed to respond simultaneously rather than in sequence, and the content cadence moved from monthly and seasonal to constant. The team adapted in real time: parallel international, national and local giving campaigns in the US; the UK’s “Kindness in Crisis” appeal; India’s first-ever local market campaign; and, as the crisis wore on, a shift toward mental health content and partnerships once acute donation fatigue set in.
Impact
100+
Campaigns launched from a standing start with no existing team or infrastructure
100+
Benefiting charities across the life of the program
$3M+
Raised across the program’s run
15+
Global markets running localized campaigns off the playbook I built
“I just wanted to thank you for featuring Ella on MSN. I knew something had happened because we did see a spike in donations on that day… how rewarding to see their little faces as the photos are just now starting to come in.”
— Meg Casano, mother of Local Hero honoree Ella Casano
