How GiveDirectly increased donations by over $3 million/year through experimentation

A guest post from GiveDirectly's team, shared alongside a Giving Tuesday fundraising appeal, walks through the experiments that raised roughly $3 million more in donations per year. It covers winning tests like faster payment options, email opt-in copy, smart defaults, and homepage redesigns, plus a failed donor-recipient matching product that was later replaced by a cheaper village-level approach.

01Key takeaways

• Small checkout friction removals, such as one-click wallets, can produce large revenue lifts.
• Clearer opt-in copy can meaningfully raise email subscription rates, which compound through donor retention.
• Defaults matter: a failed nudge can succeed when reframed, so iterate on a test before abandoning it.
• Anchor experiments to a single North Star metric so you can accept decreases in sub-metrics.
• Validate costly product ideas with cheap tests, like village-level matching, before building engineering features.

02Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.