Your First Meetup Isn't the Test. Your Second One Is

A first meetup is a rumour with a date attached. Friends turn up, curiosity does the rest, and the room fills on borrowed energy that has nothing to do with whether the group works. The second one is where the group either exists or quietly stops existing, and almost every decision worth making at the start is really a decision about that second night.

Quick answer

A first meetup measures curiosity. A second measures whether the first one was worth someone's evening. Those are different questions, and only the second one predicts anything.

The published research on event-based social networks puts a hard number on what sustained organising actually looks like: across nearly 100,000 groups, the median group runs roughly one event a month, and a typical event draws around eleven confirmed attendees. Not fifty. Eleven.

So the founding decisions that matter are the ones that make a second event cheap to run: a narrow enough topic that the next one is obvious, a cadence you can hold on a bad week, and a definition of success that counts returning faces rather than sign-ups.

The rumour and the test

Here is what happens at a first meetup, almost universally. You tell people. A few friends come because you asked. A few strangers come because the description sounded interesting and they had a free Tuesday. Someone brings a colleague. The room has twelve people in it, the conversation is good, and you go home feeling like the thing works.

None of that was a test. Every person in that room came for a reason that will not repeat: novelty, loyalty, curiosity, an empty calendar. You spent a one-time budget of goodwill and it bought you one full room.

The second event runs on a completely different currency. Nobody is curious any more - they know what it is. Nobody is doing you a favour twice. The only reason anyone returns is that the first one was actually worth the evening, and that is the first honest data you will ever get about your group.

A first event tells you how good your invitation was. A second event tells you how good your event was. Only one of those is a group.

This is why so many groups have exactly one event in their history and then nothing. It is not that the organiser lost interest. It is that they read the first night as a verdict, planned nothing for the second, and by the time they got around to it the momentum had already gone cold.

What the data says a real group actually looks like

Most advice about starting a group is written from memory by people whose groups worked. It is worth grounding the expectation in something larger than anecdote, because the numbers are much less glamorous than the advice - and much more encouraging.

A 2023 study of event-based social networks presented at the Communities & Technologies conference assembled one of the largest public datasets on this: 3,699,654 events across 97,965 groups in 500 US cities, filtered down to roughly 1.28 million events across 58,356 groups for analysis. It is, as far as public research goes, about as close as anyone has come to measuring what organising actually looks like at scale.

Two findings from it should reset most new organisers' expectations. First, the median group in that dataset holds about one event per month, with a mean of 1.825 and a standard deviation of 3.545 - meaning the average is dragged upward by a small number of very high-frequency groups, while the typical group runs monthly or less. Second, the mean number of RSVPs per event is around 11, with a standard deviation of about 19.

Measure - Typical value - What it means for you

Events per month (median) - ~1 - Monthly is normal, not lazy

Events per month (mean) - 1.825 (SD 3.545) - The average is skewed by a few very active groups

RSVPs per event (mean) - ~11 (SD 19) - A room of a dozen is the norm, not a failure

Groups in dataset - 58,356 analysed - Broad enough that this is not one city's quirk

Sit with that eleven for a second. The mental image most new organisers carry - a packed room, forty people, a real crowd - describes an outlier, not a healthy group. A recurring room of a dozen people who know each other's names is, statistically, what a working group looks like.

That matters enormously for the second-event problem, because the target you set determines whether you read event two as a success or a collapse. Twelve people at your first event and eight at your second is not decline. It is a group forming.

The topic decision is really a cadence decision

Almost every guide tells you to pick a specific topic rather than a broad one, and gives the same reason: specificity attracts the right people. That is true, but it undersells the more important benefit, which is that a narrow topic makes the next event obvious.

A group called "Tech Meetup" has to invent a reason to exist every single month. Someone has to decide what it is about this time, find a speaker for a topic that has not been chosen, and write a description from scratch. That is a creative act, repeated monthly, and creative acts are exactly what stops happening on a busy week.

A group with a narrow remit does not have that problem, because the next event is implied by the last one. Meetup's own guide to starting a group makes this point through naming, with the example of a "Seattle Grandmas Kayak Club" - a name that tells you the city, the people and the activity, and therefore tells the organiser exactly what next month looks like without a single decision being made.

The same guide offers a test that is more useful than it sounds: say the name out loud first, and then type it. Names that survive being spoken aloud tend to be the ones a member can repeat to a friend without explaining, which is the only distribution mechanism most small groups ever get.

Pick a topic narrow enough that you never have to decide what the next event is. You will not always have the energy to decide.

A real example of cadence outlasting size

The clearest illustration of why frequency beats headcount comes from an organiser writing honestly about his own groups. Alexander Reelsen, who has run several technical meetups in Vienna, published a piece arguing that Meetup member counts are a vanity metric - and backed it with his own numbers rather than someone else's.

His Elastic Vienna group had roughly 1,050 members and had held about 30 events on a quarterly cadence. Papers We Love Vienna, by contrast, had 46 members - but had run 46 meetups, monthly, typically drawing four to eight people. A third group, SilverStripe Austria, had managed 7 meetups and had been dormant since 2014.

Group - Members - Events held - Cadence - Typical attendance

Elastic Vienna - ~1,050 - ~30 - Quarterly - Varies

Papers We Love Vienna - 46 - 46 - Monthly - 4–8

SilverStripe Austria - - - 7 - Dormant since 2014 - -

The group with 46 members had run more events than the group with 1,050. By any measure that involves actual gatherings happening, the small group was the healthier one, and its member count made it look like the weakest of the three.

Reelsen's conclusion is worth adopting wholesale: the number to watch is not how many people have joined, but a rolling count of who has actually shown up recently. A membership number only ever goes up, which makes it useless as a signal - a group can be dead for two years and still be gaining members.

The mechanics of getting the group live

The platform-level steps are unglamorous and worth knowing precisely, because a surprising number of first groups stall on a detail nobody mentioned.

On Meetup specifically, a newly created group is not immediately visible. It goes through a review, and the platform's own help documentation states that a decision arrives within 24 hours - during which the group cannot be found by anyone. Organisers who create a group and immediately start telling people to search for it tend to discover this the hard way.

The same documentation recommends creating your first event immediately after the group is approved, and this is better advice than it appears. A group page with no event on it is an abstraction: someone can join, but there is nothing to actually attend, and joining a group is a much weaker commitment than putting a date in a calendar.

• Write the name out loud before you type it. If you cannot say it to a stranger without adding a sentence of explanation, it is not the name yet.
• Create the group and expect to wait. Meetup's documentation puts the review decision within 24 hours, and the group is invisible until it clears.
• Schedule the first event before you announce anything. An event is something to attend; a group is something to think about attending.
• Schedule the second event at the same time. This is the entire argument of this article compressed into one action, and almost nobody does it.
• Decide what you will count. Pick returning attendees over total members now, before the member number starts flattering you.

Why scheduling the second event first changes everything

There is a practical reason and a psychological one, and the psychological one is stronger.

Practically, at the end of a good first event you have a room full of people who are, at that exact moment, the most positively disposed toward your group that they will ever be. Telling them the date of the next one while they are still in the room converts that feeling into a calendar entry. Telling them three weeks later, by message, converts it into nothing, because the feeling has expired.

Psychologically, having a second date on the calendar changes what a mediocre first event means. If event one is a disappointment and nothing is scheduled, the rational move is to quietly not do it again, and that is what most people do. If event two is already booked, the disappointing first event becomes a data point rather than a verdict - something to adjust, not something to abandon.

A scheduled second event turns a bad first event from a reason to stop into a reason to change something.

What experienced organisers do differently

Dave Kiss ran a tech meetup for five years, from 2018 to 2023, and wrote up what he learned with unusual specificity about the boring parts - which are, predictably, the parts that determine whether a group survives.

His first event drew around forty people at a brewery, which by the numbers above is an exceptional start. What kept the group running for five years was not that night; it was a set of operational habits that removed decisions from his weekly workload.

He plans three to four months of events ahead - the cadence problem solved structurally rather than through willpower. He runs a five-touch reminder sequence: the initial announcement, then two weeks out, one week out, three days out, and the platform's automatic reminder. And he budgets around $250 to $300 per event, ordering food for 75 to 80 percent of the RSVP count, because he measured roughly a 30 percent drop-off between RSVPs and actual attendance.

Decision - His answer - Why it matters

Planning horizon - 3–4 months of events booked ahead - Removes the monthly "what now?" decision

Reminder cadence - 5 touches: announce, 2wk, 1wk, 3day, auto - Attendance is a memory problem as much as an interest one

Catering quantity - 75–80% of RSVP count - Accounts for a measured ~30% no-show rate

Budget per event - ~$250–$300 - Predictable enough to sustain, small enough to self-fund

One detail from his write-up is worth repeating because it is the kind of mistake that only shows up on the invoice: switching from pizza to sandwich catering cost roughly two and a half times as much for the same headcount. Food is usually the largest variable in a small event budget, and the difference between a sustainable group and an expensive hobby often sits entirely in that line.

His most quotable line is also his most useful: "two attendees is a meetup." A group that only runs when fifteen people confirm is a group that will cancel itself out of existence. A group that runs with three is a group that still exists next month.

The topic-overlap question nobody asks early enough

There is a counterintuitive finding in that same 2023 event-based social network study that is directly relevant to anyone choosing a topic in a city that already has groups.

The researchers examined how much a group's topics overlap with other groups nearby, and found the relationship to activity is not linear. Organisers create more events when topic overlap with neighbours is very low or very high, and fewer when it sits in the middle.

The interpretation is intuitive once stated. Very low overlap means you are the only group doing this thing, and the people who want it have nowhere else to go. Very high overlap means you are in a genuine scene, where multiple groups reinforce each other's audiences and a rising tide lifts everyone. The dead zone is the middle: similar enough to compete for the same people, not similar enough to build a scene together.

The study also found that groups with high topic overlap that were geographically clustered showed roughly 1.35 more events per month - a substantial difference, and a strong argument for starting a group in a neighbourhood that already has adjacent activity rather than deliberately picking an underserved area.

Being the only group of your kind works. Being one of many of your kind works. Being the second-most-similar group in town is the hard place to be.

What to count instead of members

Every platform hands you a member number on the front page of your own group, and every organiser checks it. It is the wrong number, for a simple structural reason: it only ever increases. People join and almost never leave, so the figure measures cumulative historical interest rather than current health.

Three numbers are more useful, and none of them are displayed by default.

• Returning attendee count. Of the people at this event, how many were at the last one? This is the single best health signal a small group has, and it is the number the second-event framing is built around.
• Rolling twelve-month attendance. How many distinct humans have physically attended in the last year? This is the group's real size, and it is usually a small fraction of the member count.
• RSVP-to-attendance ratio. A stable ratio means you can plan catering and space. A worsening one usually means your reminders, not your content, need work.

That last one connects to a broader pattern worth knowing: on Eventbrite's data, free events can see no-show rates as high as 50 percent while paid events run as low as 10 percent. If your ratio is drifting badly, the fix is often structural - a token price, a firmer reminder cadence - rather than anything to do with the quality of the event itself.

The platform is doing more work than you think

It is easy to assume a new group is entirely dependent on the organiser's own promotion, and for the first event that is largely true. Over a longer horizon, discovery mechanics matter more than most organisers account for.

Meetup's 2025 progress report describes a platform of more than 300,000 groups across 10,000-plus cities, and reports specific effects from discovery changes: a 20 percent increase in RSVPs following an Explore page redesign, a 10 percent increase in clicks on search recommendations, and a 9 percent increase in group joins originating from search.

None of those numbers will save a group that has nothing on its calendar. But they do mean that a group with a clear name, a specific topic and a consistently scheduled next event accumulates discovery over time in a way that a vaguely named, sporadically active group does not. The founding decisions compound quietly.

Three questions before you create anything

• If eight people come, is this still worth doing? Eight is close to the statistical norm. If the honest answer is no, the format needs to change before the group launches, not after.
• Do I know what the second event is, right now? Not roughly - specifically. If the answer requires thinking, the topic is too broad to sustain a cadence.
• What cadence could I hold during a bad month? Whatever that is, that is your real cadence. Anything faster is a plan that depends on nothing going wrong.

What to take from this

A first meetup is easy and means very little. The energy that fills it - novelty, friendship, curiosity - is a one-time grant, and reading it as evidence is the most common way new groups end up with exactly one event in their history.

Build for the second one instead. Choose a topic narrow enough that the next event never needs to be invented. Set a cadence you could hold on your worst week rather than your best. Book event two before event one has finished. And count returning faces rather than members, because one of those numbers can tell you the truth and the other one is designed to flatter you.

A room of a dozen people who keep coming back is not a modest version of a real group. According to the largest public dataset we have on the subject, it is what a real group looks like.

Frequently asked questions

How many people should come to a first meetup?

Fewer than most organisers expect. In a dataset of nearly 100,000 groups, the mean number of RSVPs per event was around eleven, so a first event drawing a dozen people is squarely normal rather than disappointing.

How often should a new meetup group meet?

Monthly is the median across the same dataset, and it is a realistic default. The more important question is what cadence you could sustain during a bad month, because that is the cadence the group will actually settle into.

Should I schedule the second event before the first one happens?

Yes, and it is one of the highest-leverage things a new organiser can do. It converts the goodwill in the room at event one into a calendar entry, and it turns a disappointing first event into something to adjust rather than a reason to stop.

Is member count a useful measure of a group's health?

Not really. It only ever increases, so it measures cumulative historical interest rather than current activity - a group can be dormant for years and still be gaining members. Returning attendance is the more honest signal.

How long does it take for a new Meetup group to go live?

Meetup's help documentation states that a review decision arrives within 24 hours, and the group cannot be found by anyone until it is approved, so it is worth creating the group before you start telling people to look for it.

Should I start a group on a topic that already has groups in my city?

Often yes. Research on event-based social networks found organisers run more events when topic overlap with nearby groups is either very low or very high, and fewer when it sits in the middle - a genuine scene tends to help more than it competes.

How much should a small meetup cost to run?

One organiser who ran a tech meetup for five years budgeted roughly $250 to $300 per event including food, and noted that switching from pizza to sandwich catering cost about two and a half times as much for the same headcount.

What should I do if only two or three people show up?

Run it. An organiser with five years of experience put it plainly: two attendees is a meetup. A group that only runs when fifteen people confirm will eventually cancel itself out of existence.

Ontaym Editorial Team

Ontaym builds tools for organising real-world gatherings, so the team spends its days on the coordination problems this article describes. Articles are researched against primary sources, reviewed before publication, and revised when the underlying facts change rather than on a schedule.

Give your next meetup one address instead of one more thread.