Twelve years of Opportunity Hack, in numbers.
Across 21 hackathons on three continents, 4,076 hackers registered to build technology for nonprofits — and 1,080 of them shipped. This is what their data tells us about how social-good hackathons actually work.
From one San Jose weekend to a multi-format, year-round program.
Opportunity Hack started as a single hackathon in October 2014. It grew, stalled, pivoted through COVID, and emerged with a more flexible event design. The history breaks cleanly into four eras.
Three of every four registrants never ship a project.
Across the full history, the registration → submission rate sits at 26.5%. That's the universal hackathon problem: getting people to sign up is easy, getting them to ship is hard. The good news is that completion has been climbing meaningfully since 2022 — recent events run 30–40%, double the historical baseline.
Caveat: 2026's 71% rate is inflated because the dataset for that year is filtered to invited/accepted-only flows. Treat post-2023 trend as directional.
Team status at registration is the strongest predictor of who ships.
Hackers who register with a team submit at 39.7%. Those still looking for one drop to 17.8%. Solo hackers crater at 9.8% — a 4× gap. The same pattern shows up cleanly on the project side: solo projects ship 46% of the time, while three-or-more-person teams ship 93%+ of the time. This is the closest thing to a universal law in the dataset.
Listing platforms bring volume. Friends and direct outreach bring finishers.
Four channels do almost all the work: friend referrals (606), DevPost (463), school/club (331), and direct from organizer (278). But quality is wildly uneven. Hackers who heard about the event on DevPost finish at 11.9%. Hackers who heard from a friend finish at 34%. Hackers who heard directly from the organizer finish at 39.2%. Returning alumni finish at 75%.
An anchor school, a developer-heavy crowd, two strong country bases.
Geography is heavily concentrated. Arizona State University alone supplies 985 registrants — about 24% of the entire history. ASU + Arizona cities account for over a quarter of all signups. Outside Arizona, San Jose, Chicago, and Atlanta form a secondary cluster, each tied to a specific corporate-sponsored event series. This "anchor school" pattern is common in nonprofit hackathons: a single committed academic partner generates most of the volume.
Late registrants finish. Repeat hackers compound.
The closer to the event a hacker registers, the more likely they are to ship. Same-day registrants finish at 44%. Hackers who signed up 31–60 days early finish at 18%. Long lead times signal weak intent. On the loyalty side, 219 hackers have shown up to two or more events — and they convert at 42%, nearly double the first-timer rate. But the multi-time finisher tail is small: only 51 hackers have ever submitted projects at more than one event in twelve years.
JavaScript-heavy, modern, increasingly leaning on managed backends.
Across 13 years of submitted projects, JavaScript / Python / React dominate the long-tail counts. But the modern additions — TypeScript, Next.js, Tailwind, Supabase, Firebase — show projects increasingly leaning on managed services that compress time-to-MVP. This stack evolution probably contributes to the rising completion rate in §02: it's measurably easier to ship a working app today than it was a decade ago.
Five patterns the data argues for — broadly applicable beyond Opportunity Hack.
These aren't OHack-specific recommendations. They're patterns that emerge from twelve years of registration and submission data, and they're likely to repeat at any hackathon with a similar structure (multi-day, project-based, open registration, social-good or technical theme).
Treat solo registration as a problem to solve, not a state to allow.
Solo hackers finish at 9.8%. Teamed hackers finish at 39.7%. The single highest-impact change a hackathon can make is to put real product effort into team formation before the event starts — not as an afterthought during the kickoff. Matchmaking flows, "looking for a team" boards, and forced team-join steps at registration would all raise the completion floor.
Don't confuse listing platforms with recruitment channels.
DevPost-sourced registrants finish at 11.9% — about a third of the rate from friend referrals (34%) or direct organizer outreach (39%). Public listing platforms are useful for awareness but should not be where most marketing energy goes. Word-of-mouth and direct outreach are 3× more efficient on a per-finisher basis.
Keep registration open until the event starts.
Same-day registrants finish at 44%. Hackers who signed up over a month early finish at 18%. Many hackathons close registration two weeks early "for planning purposes" — that decision systematically cuts off the highest-intent segment of the audience. Late registration is high-intent registration. Plan for a day-zero surge instead of fighting it.
An anchor school is a strength — but track the concentration risk.
OHack's relationship with Arizona State produces 24% of all registrations at a 35% completion rate. That's an excellent partnership. But it's also a single point of failure: a policy change, a curriculum shift, or a key contact leaving could halve the funnel overnight. The lesson for any hackathon: name your anchor partnerships explicitly, measure the concentration, and proactively cultivate two or three more before you have to.
Multi-time finishers are rare. Treat the ones you have like family.
Across twelve years, only 51 hackers have ever shipped projects at more than one Opportunity Hack event. Five have shipped three times. Zero have shipped four. That's a small number — and it argues against mass-email "alumni newsletters" as the right tactic. The right tactic is high-touch, named-relationship outreach: invitations to mentor, advise nonprofits, judge, or recruit teammates from new cohorts. Help your alumni stay in the ecosystem after they age out of competing.
Three ways to put twelve years of social-good hackathons to work.
We've spent over a decade learning what works (and what doesn't) when you run a hackathon for nonprofits. If any of the patterns above are useful for your team, here's how to put them into practice — whether that's hosting your own event, partnering with us, or just staying in touch.
Run a hackathon for your nonprofit or company
Tell us about the problem you want solved or the team you want to put through a hackathon. We've co-hosted with corporate sponsors, university partners, and nonprofits since 2014.
Fund a social-good hackathon
Sponsorship keeps OHack free for nonprofits and developers. Tiers include dedicated mentor and judge slots for your team — a popular professional-development perk that counts toward most ESG and corporate-volunteer programs.
What a hackathon for social good actually is
The companion page to this report — what makes a hackathon a "hackathon for social good," how the OHack model works in detail, and details for the next flagship event.
To empower students, professionals, and nonprofits to collaboratively create sustainable tech solutions that drive social impact and foster learning.
To build a global community where individuals can accelerate their career growth while making a lasting impact for nonprofits.
- What's Opportunity Hack?
- Sponsor Social Good
- FAQ
- Our History
- Join Us on Slack
- Privacy Policy
- This website is a public good.
Fork this on GitHub 6261594