2.2 · Registration ops: the no-show math

IntermediateDuration ~10 min readTools Your master sheet, Runway tab, Your Luma event page, The venue capacity number from 1.6

The big idea: registration is a forecast, not a form. Two numbers make the forecast, the room number and the no-show rate you’re assuming, and once you write both down the cap, the waitlist and the whole check-in setup fall out of them.

The analogy

Think about the day the script called for a crowd. The 1st AD needs fifty bodies in frame at nine in the morning, and casting does not book fifty. They book seventy, sometimes more, because a known share of booked background never turns up, and everybody in that chain has known it for about a century. Nobody treats it as betrayal. It’s arithmetic, and it’s why the extras tent has a sign-in sheet, why there’s a confirmation call the night before, and why a few standby people are sitting at basecamp with a coffee waiting to be pulled in.

Look at where that decision gets made. Not at six in the morning when someone counts heads and panics. In pre-production, when a person chose the booking number, and chose it out loud so the budget could see it.

Your registration page is the booking number. Everything else in this lesson is the sign-in sheet, the confirmation call, and the people on standby.

How it works

Registration looks like a form. It is really a forecast. Every time somebody hits register you learn one small thing about a room that doesn’t exist yet, and your job is to translate those clicks into an honest guess at how many humans will physically be standing in that space. Get it wrong in either direction and you have either turned away people you had room for, or bought food for a room that arrives half empty.

Start with the room number. That comes from capacity, which you scored in 1.6, and it is the honest count of how many people fit in that space in the setup you’re actually running, not the number you wish it held. Write it at the top of the page. It is the only number in this lesson that the room itself decides.

Then the number you choose. Your no-show rate is the share of registrants who never arrive, and for free in-person events the planning range published across the industry is roughly 40 to 60 percent. That range is wide because it is really a range of communities, so you have to pick a figure inside it and commit. The important part is not being right on the first attempt. It is writing the figure down next to the room number, labelled as an assumption, so that six months from now you can look at the two numbers together and audit which one was off.

Now the arithmetic that turns them into your cap. Overbooking just means letting more people register than the room holds, on purpose and by a chosen multiple. Divide the room number by the share you expect to actually show, or use the shorthand the industry works with and take 1.5 to 2 times the room. Say August is built for a room of 120, and swap in the real venue number the moment it is confirmed. Assume half don’t show and your registration cap is 240. That is not optimism. It is the same booking number the AD would have chosen, and it is the number that goes into Luma.

Worth being straight about why free carries the highest no-show, because the reason is not what most people assume. It isn’t that signing up is too easy. Platform data actually shows free events completing registration at a higher rate than paid ones and still losing far more people on the day. What price buys is commitment, not friction. Which leaves you two honest levers. A small ticket price, or a deposit you refund at the door, moves the number hard, and it also puts a cost on a door your whole mission says should stay open. That tension is real and you get to weigh it. Or you manufacture commitment without a price: ask one registration question that takes actual thought, ask people to name what they’re working on, ask them to bring someone. Every small act at sign-up deepens the feeling of having committed, and committed people turn up.

Then the number that replaces your guess. Attendance rate is checked in divided by registered, and you only ever get it if you check people in properly. That is the sign-in sheet at the extras tent. Skip it and you will be assuming again next event, and the event after that, forever.

The Luma side is short, and one detail is worth getting right the first time. Set an event capacity, not just a ticket capacity, because the waitlist switches on when the whole event is full, and a ticket type that fills on its own just reads Sold Out with nowhere for the overflow to go. With the cap in place the room refills itself. Cancellations and released seats become approvals off the waitlist, and waitlisted people don’t count against your capacity until you approve them. Approve in batches, with a message, and remember that people keen enough to wait in line are usually your most committed guests.

The rest is cadence and the door. Luma sends its own reminders a day and an hour before, which is the floor, not the plan, so add your own blasts. Learn the guest statuses too, because Going, Pending, Waitlist and Invited are four genuinely different situations and only one of them is a person in your room. And build the check-in setup before load-in day rather than at the door: choose your scan mode, add your check-in volunteers, and have someone scan a test ticket while there is still time to fix it.

Last, the pulse check. A registration health check at one month out asks whether you are sitting at 40 to 50 percent of your target, counted without sponsors, guests, organizers and volunteers. For our 240, that is roughly a hundred real registrations. Under it, you don’t cut the cap, you pull convert phase work forward in the calendar you built in 2.1. And know that a genuine spike in the last two weeks is normal, so the check is a steering signal, not a verdict.

One thing this reframes. The first Filmmakers Connect event drew 120 plus registrants against an expected 20, which felt like a lovely surprise at the time. Through this lens it was never a surprise. It was a forecast nobody had written down yet.

The pattern

Two numbers, then let the room refill itself. Write the room number and the no-show rate you’re assuming, side by side, labelled as an assumption. Divide to get your registration cap. Put that cap in Luma as event capacity so the waitlist turns on, then check in every single person, so next time the second number is measured instead of guessed.

The receipts (evidence, if you want it)

• Free in-person events routinely lose 40 to 60 percent of registrants to no-shows, while paid in-person events land around 70 to 90 percent attendance (eventcloud, how to cut event no-shows; knvi labs).
• DevOpsDays’ own operating number for free tickets: “about 30-40% of the people ‘grabbing’ a free ticket don’t show up in the end,” which they say “makes it harder to plan logistically” and means “people who could have attended are left out because the event appears ‘full’” (devopsdays.org/organizing). They also define the measure the way this lesson does: “actual attendance rate (i.e., checked-in people vs tickets sold).”
• Paid events typically see 10 to 20 percent no-show, with some sources reporting 3 to 5 percent at meaningful price points (AirLST).
• The price ladder, from one organizer’s own data: free runs about 40 percent no-show, a $5 ticket drops it to around 15 percent, and above $10 it falls under 5 percent (Localist, should I charge for my event). The course’s industry map also records a $3 refundable deposit cutting no-shows from 38 percent to 14 percent (eventcloud), and the same piece recommends manufacturing commitment without a charge: ask registrants to pick a session, submit a question for the speaker, or RSVP to a specific table.
• Commitment, not friction: Pheedloop’s platform data shows free events with higher registration completion, around 96 percent median, but higher no-shows at around 28 percent, against roughly 86 percent completion and 17 percent no-shows for paid events. Their summary is that payment rather than friction is the commitment filter (Pheedloop Event Data Lab report 06).
• The 1.5 to 2 times overbooking figure is the planning implication practitioners work from for free community events, not a published study. It is arithmetic on the no-show ranges above, which is exactly why this course teaches you to write your assumed rate down rather than adopt a multiplier on faith.
• The health check, verbatim: “Did we hit about 40-50% of expected ticket sales minus sponsors, speakers, organizers, and volunteers by a month or so out?” with the note that if you announced late or skipped emails “you probably are closer to 25-30% at this point” (devopsdays.org/organizing). The same guide lists where registration spikes come from (announcement, early bird close, speaker announcement, agenda announcement, sponsor and speaker codes, reminder emails) and confirms that “it is normal to get a spike in ticket sales in the last two weeks.”
• Luma waitlist mechanics: set the event capacity and toggle Over-Capacity Waitlist under Registration. The waitlist is event wide, not per ticket, so “a ticket capacity on its own won’t turn on the waitlist.” Waitlisted guests don’t count toward capacity, hosts must approve them manually, and for paid tickets payment is authorized when they join and only captured on approval (Luma help, waitlist).
• Luma registration flow: the confirmation email carries a calendar invite that adds itself automatically, default reminders go out 1 day and 1 hour before an in-person event, and registration can be closed immediately, on a sales end time, or by hitting capacity (Luma help, event registration process).
• Luma guest statuses are Going, Pending, Waitlist, Invited, Not Going, Checked In and Not Checked In, and check-in is tracked separately from approval status through a per-ticket checked_in_at timestamp. Bulk status updates accept a CSV or a pasted list of emails (Luma help, managing your guest list).
• Check-in setup: Standard Mode opens each guest’s details for review, Express Mode checks people in automatically on scan with colour coded feedback and is “ideal for large events where speed matters.” You can lock the scan mode so every volunteer behaves the same way, and add check-in staff who can scan tickets and view the guest list without full event management access. Check-in staff requires Luma Plus (Luma help, check in guests; adding hosts and managers, which also caps hosts at 5 per event, rising to 70 on a verified calendar).
• Luma Insights tracks the funnel from page views to registration to showing up, plus top referrers and custom utm_source links, which is how you find out which announcement actually produced registrations (Luma help, event insights).
• Platform choice: free events are free on Luma, paid tickets carry a 5 percent platform fee plus Stripe processing, and payouts land as tickets sell rather than after the event. Eventbrite trades a higher blended fee for marketplace discovery and features like seat maps (Luma help, Luma vs Eventbrite).
• Honest gap. There is no credible published cost per registrant benchmark for community scale events, so this course never quotes a CAC figure (industry map §8, flag 2). If you run paid promotion, measure your own and treat it as your number, not the industry’s.

Your one move

Open the Runway tab on your master sheet and put two labelled cells at the top: “Room number” and “Assumed no-show rate.” Fill both, using the venue capacity from 1.6 and a figure you picked on purpose. In the next cell, divide them into your registration cap. Then open the Luma event, go to Registration, set Event Capacity to that cap, and toggle on the Over-Capacity Waitlist.

Twenty minutes. Done when both numbers are visible on the sheet, the cap is derived from them rather than guessed, and Luma’s capacity matches.

Want to go further?

Add one registration question that takes real thought to answer. What are you working on right now, or what do you want to walk out of the room with. It gives you something to read before the event, it gives your host something to open with, and every second a person spends answering it makes them fractionally more likely to walk through the door.

The needle: an event is won before doors open, and a half empty room on the day was decided the afternoon you typed a number into the capacity field. Change the number now, while changing it costs nothing.

Terms introduced

Terms introduced in this lesson

• Overbooking
• No-show rate
• Attendance rate
• Waitlist
• Registration health check
• Capacity

Each term links to the glossary, where it is defined by the way you just used it.

Check

Check yourself

1. Your venue confirms the room holds 120. The event is free, so you set the Luma capacity to 120 to be safe. It fills in a week, you turn people away, and on the day 71 people walk through the door. What actually went wrong?

   The promotion calendar underperformed and did not reach enough peopleNothing, no-shows are unpredictable and this is just how free events goThe venue gave you the wrong capacity numberThe cap was set to the room number instead of the room number adjusted for a no-show rate, so a half empty room was decided the day you typed 120 into the capacity field

2. It is one month out. Your registration target is 240. You have 62 registrations, and 14 of those are sponsor reps, podcast guests and volunteers. What does the health check tell you?

   You are on pace, because the biggest spike always arrives in the last two weeks anywayYou are well under pace, so promotion work from the convert phase gets pulled forward now instead of counting on the final two week spikeCut the cap down to match the current pace so the forecast stops looking optimisticNothing yet, the health check is a number you read after the event

3. You put a capacity on your general admission ticket type but left the event capacity blank. The ticket hits Sold Out, people start messaging you asking how to get in, and nobody is landing on your waitlist. Why?

   The waitlist is event wide, and it only switches on when the event reaches its event capacity, so a full ticket type just reads Sold OutWaitlists have to be enabled by Luma support before your first eventWaitlists on Luma only work for paid ticketsThe event page was set to private, so the waitlist never appeared

Check answers

You can move on when you can… say your room number and your assumed no-show rate out loud, do the division that turns them into a registration cap, and name what you would change in the promotion calendar if you were sitting under half your target a month out.

Mark this lesson complete

Go deeper

• Next up: 2.3 · Community-facing marketing: hook, story, belonging, where the registrations this lesson forecasts actually get earned, in the Filmmakers Connect voice.
• Optional reading, about ten minutes total: Luma’s own docs on the waitlist and checking in guests. Read them once before you set the cap, not on the morning of the event.
• Optional reading: the tickets and registration sections of the DevOpsDays organizing guide. It is written for a paid conference, so the pricing advice needs translating, but its registration pace signals and free ticket trade-offs are the clearest published version of the math in this lesson.