538 people, 105 meetups: what our 2024 campaign taught us
By Dev, founder of AroundMe · August 3, 2026 · 7 min read · AroundMe · Behind the scenes
There's a version of this post every startup writes, and it's a highlight reel. This isn't that.
In 2024, before the AroundMe you see today existed, we ran a one-quarter campaign at UC Riverside that onboarded 538 people and facilitated 105 meetups. Then we did something that needs explaining: we retired the product that ran it and rebuilt AroundMe from the ground up.
Both halves of that sentence are the story. The numbers are the receipts; the rebuild is what the numbers taught us. And because what didn't work turned out to be more useful than what did, that's where we'll spend most of this post.
What was the 2024 campaign?
In one quarter at UC Riverside, a student team took AroundMe from zero to a campus-wide network (538 users onboarded), testing three use cases in the real world: academic, social, and professional.
It was run like an actual operation, not a vibe. We defined the objective up front (500 users, three use cases), then built a team around four functions: Partnerships & Ops, In-Class Marketing, Content Marketing running a LinkedIn and Instagram blitz, and a real-time Data team watching the numbers as they happened. Partners gave a no-name product borrowed credibility: Microsoft for Startups, Blackstone Launchpad, the UCR School of Business, AGSM Women in Business, and the AGSM Product Club.
The testing itself was physical. For the academic use case, we ran in-class networking activities across 30 classes. For the social one, we set up on campus and talked with students face-to-face. For the professional one, we hosted a networking event with product marketing experts from Atlassian, Microsoft, and Tyler Technologies. We documented the whole thing in a case-study documentary. The full campaign page, video and photos included, lives at /campaign1.
What do the numbers actually say?
The headline stats: 538 people onboarded, 105 meetups facilitated, 65 hotspots created, a 58.35% conversion rate, and reach of 922 people offline and 9,348 online, all from one quarter on one campus.
The conversion number deserves decoding, because it's the one we still think about. 538 is 58.35% of 922: the recorded conversion rate is exactly the share of people we reached in person who became users. We reached ten times as many people online, but the campaign's defining number measured the face-to-face funnel. Make of that what you will; we made a product philosophy out of it.
One more thing about receipts, because it matters for trust. When we rebuilt this website in 2026, we discovered the old campaign page had been showing template placeholder numbers ($40,000, $215,000, 50%) that leaked from the old site builder's mock data into production. Nobody invented them; nobody caught them either. The numbers in this post are the real ones from the campaign's own tracking sheet, and fixing that page is part of why this post exists. If you call something a case study, the numbers have to be the ones that happened.
What did the 2024 product look like?
Nothing like today's. The 2024 MVP was a browse-and-request app: you joined or created a hotspot, saw who else had joined, sent meetup requests, and coordinated the actual meetup in chat.
The 2024 flow: browse-and-request
- OnboardA quick signup, the one step that survives unchanged today.
- Create or join a hotspotAnyone could spin one up; 65 existed by quarter's end.
- See who joinedBrowse the other users in your hotspot.
- Send a meetup requestPick people and ask to meet.
- Chat and coordinateWork out the where and when yourselves.
To its credit, the loop worked end to end, and it produced 105 real meetups, which is more than most MVPs ever manage. But look at where the flow terminates: chat and coordinate a meetup. After everything the app did, the hardest part of adult social life was still the user's job: turning mutual interest into a confirmed plan with a time and a place.
What did 105 meetups teach us?
Six lessons, and every one became a design decision in today's AroundMe. Where the campaign gave us direct evidence, we'll say so; where a lesson is a conviction the campaign made unavoidable rather than a stat it produced, we'll say that too.
1. Coordination is the product, not a step you can leave to chat. The meetups happened where we did the work: 30 in-class activities, a staffed spot on campus, a hosted event with invited speakers. A four-function team was operating all quarter. We came to believe the app's job is to hand you a finished plan (group, activity, time, meeting spot) so the only work left is showing up. That's why today an invite arrives as a complete plan, and final logistics land the day before.
2. Browsing people is the wrong primitive. See-who-joined, send-a-request has an asymmetry built in: someone requests, someone gets requested, and everyone is quietly evaluating everyone from a screen. We came to believe you should express what you want to do, and never order up other people. That rule now runs through the whole product: no profile browsing, no requests, no swiping. You opt in for the week; a group is composed around you.
3. A hotspot has to be a commitment, not a label. Under the 2024 model, creating a hotspot took a minute, and 65 existed by the end of the quarter. Creation was the easy part. What makes a hotspot produce meetups, again and again, is everything after creation: a real venue, a recurring time, someone accountable for it. We came to believe a hotspot should be a promise the company keeps, not a label a user makes. Today every hotspot is company-created and verified: a theme, a real public meeting spot, and a recurring time slot, with a backup venue defined before the first invite goes out.
4. Groups need composing, not accumulating. In 2024, a meetup was however many people answered a request. We came to believe group size is most of the outcome (4-6 people, small enough for one conversation, large enough to survive a no-show) and that who's in the group should be a decision, not an accident of who clicked. Today matching composes every group deliberately, and 4-6 is a rule, not an average.
5. An invite has to be precious. When requests are unlimited, every request is cheap: easy to send, easy to ignore, easy to flake on. We came to believe one invite at a time is what makes a yes mean something: when you confirm, everyone else in your group is actually coming. And since friendship runs on repeat hours, the goal was never many meetups this week. It's a meetup that actually happens this week, and again after that.
6. Activity-first was in the campaign before we had words for it. The three use cases we tested (academic, social, professional) were all built around doing something together, and every meetup formed inside a themed hotspot, not out of browsing strangers at random. Today that's the entire thesis. The meetup forms around what you want to do: hike, study, co-work, play, build. Dinner is one format on the list, not the list.
What 2024 proved, and what we kept
- Demand is real: 538 signups in one quarter, with no famous brand behind us
- People show up: 105 meetups actually happened
- Concentration works: one campus was enough to form a real network
- In-person trust converts: over half the people we reached face-to-face joined
- Shared context beats browsing: every meetup formed inside a themed hotspot
How different is today's AroundMe?
Different enough that we retired the old app instead of updating it. The App Store listing is gone on purpose. This wasn't iteration; it was a rebuild around the lessons above.
| 2024: browse-and-request | Today: curated invite | |
|---|---|---|
| Who forms the group | You do: browse users, send requests | Matching does: you opt in for the week |
| Hotspots | Anyone could create one | Company-verified: theme + venue + recurring time |
| The venue | Whatever the chat settled on | A real, verified public meeting spot, backup included |
| Group size | However many said yes | 4-6, composed on purpose |
| Invites | Unlimited requests | One at a time (a yes means everyone's coming) |
| Coordination | Your job, in a chat | The plan arrives finished; logistics land the day before |
If you want the step-by-step of how the current version works from signup onward, we wrote the whole path, waiting included, in What happens after you join the waitlist.
Where does this go next?
Southern California, region by region: Irvine / Orange County, Riverside / Inland Empire, Los Angeles, and San Diego. The waitlist is open to anyone anywhere; activation is deliberately narrow, because a meetup only works when enough people in one place are in the pool at the same time. That's the same concentration lesson the campaign taught us, applied to a bigger map. And there's a detail we like: Riverside, where 538 people taught us everything in this post, is one of the four regions.
No launch-date theater, no fake momentum. The 2024 campaign is the reason we get to say "we've done this before." The rebuild is the reason we'd rather show you the receipts than the highlight reel. 538 people and 105 meetups taught us what this product had to become, and the only test that matters now is the next meetup that actually happens.