Can AI Run Your Studio's Class Waitlist Automatically?

Mostly yes, and usually without AI. Booking platforms such as Mindbody, Momence and TeamUp already move people off the waitlist automatically, text them, and stop at a cutoff before class. AI adds value around the edges: forecasting late cancellations to decide whether to overbook mat classes, answering "am I in?" messages, and spotting classes that need another slot.

The work that makes a waitlist run itself is choosing the rules: which promotion mode each class uses, when the waitlist stops promoting, how payment is taken, and how that fits your late-cancel policy. Most waitlist complaints ("I was added at midnight and charged a no-show fee") come from settings that clash, not from software failing. Get those right first; the AI is optional.

Follow me on Instagram@sagnikteaches

The three waitlist modes your software probably already has

Different platforms use different names, but they offer versions of the same three approaches. As described in each vendor's own help pages:

Connect on LinkedInSagnik Bhattacharya

Each platform also has a way to stop automatic promotion close to class time: Mindbody calls it the waitlist lock window, Momence has a waitlist cutoff time, and TeamUp's auto-registration applies only if a spot opens more than a set period before class (one day by default), after which people have to claim it themselves. Check the current settings in your own account, since vendors add and rename options regularly.

Subscribe on YouTube@codingliquids

Choosing a mode for each type of class

One setting for every class is the most common shortcut and the source of most complaints. Here's an illustrative filled-in plan for a studio running reformer Pilates, mat yoga and monthly workshops:

CLASS                     MODE              STOP PROMOTING    NOTES
Reformer, 6:30am weekdays Auto-promote      8pm the night     Early risers can't be
(8 reformers)                               before            expected to check phones
                                                              at 5am
Reformer, daytime and     Auto-promote      2 hours before    Matches the 12-hour
evening (8 reformers)                                         late-cancel window below
Mat yoga, Sat 9am         First to claim    45 min before     Always full; people
(25 mats)                                                     nearby grab spots fast
Mat yoga, other classes   Auto-promote      2 hours before
Monthly workshop ($45)    Confirm first,    24 hours before   Nobody should be charged
                          4 hours to reply                    $45 without saying yes

Late-cancel window: 12 hours for all reformer classes, 4 hours for mat.

The reasoning matters more than the exact times. Early-morning classes need promotion to stop the evening before. Expensive sessions need a yes before anyone is charged. And the stop-promoting time has to make sense next to your late-cancel window, which is the next problem.

Walk the workshop row through a real evening to see why the reply window needs thought. A place opens at 10pm on Wednesday for Saturday's $45 workshop. The first person on the list is asked to confirm and has four hours, so the offer lapses at 2am while she's asleep and moves to the second person, who is also asleep, and so on down the list. By morning the place has gone to whoever happened to be awake, which is exactly the unfairness confirm-first was meant to prevent. Either lengthen the window to 12 hours for workshops, or hold requests overnight so the first person is asked at 7am. With a 24-hour stop before a Saturday morning workshop, there's still room for two or three people to be asked in turn.

Cutoffs, late-cancel fees and the midnight promotion

This is the realistic mistake that generates angry emails. A studio sets reformer classes to auto-promote right up to two hours before class, with a 12-hour late-cancel window and a no-show fee. At 11:40pm, someone cancels the 6:30am class. The first person on the waitlist is added automatically and texted. She's asleep. She sees the text at 7am, after the class, and finds she's been charged a no-show fee for a class she didn't know she was in.

Three rules prevent it:

1. Stop auto-promotion before the late-cancel window starts, or at least before most people go to bed for early classes. A promoted student should always have time to cancel without penalty.
2. Waive fees for anyone promoted inside the late-cancel window, if your software can't avoid promoting them. Write this into the policy so staff don't have to decide case by case.
3. Control when messages go out. Momence, for instance, has uptime settings that control when waitlist requests start firing to customers, so nobody gets a text at 2am. If your platform has an equivalent, use it.

Payment timing is the other setting to check. Momence charges customers if they end up making it into class from the waitlist. TeamUp only auto-adds people who hold a valid membership or credits and have signed any required forms and waivers. Make sure your policy page says plainly when a waitlisted student is charged, so the answer never surprises anyone.

Where AI actually earns its place

Once the rules are right, the waitlist runs without AI. These are the jobs where AI does add something:

• Forecasting late cancellations and no-shows, so you can overbook mat classes by a spot or two with confidence. Reformer classes can't be overbooked, because the equipment is the limit, but the same forecast tells you which reformer classes are worth adding.
• Answering waitlist questions in chat, Instagram or email: "Am I in?", "What number am I on the list?", "Will I be charged?". An assistant connected to your booking system, or one that sends the student to their account page with a clear answer on charging, saves the front desk dozens of messages a week at a busy studio.
• Spotting when a class needs a twin. A waitlist that's longer than five for four weeks running is a timetable signal, not a waitlist problem. Fixing your class timetable with booking data and AI covers the analysis.
• Writing the notifications. Default waitlist texts are often cryptic. AI can draft clear versions that include the class, time, charging rule and how to drop out, which you then paste into your platform's templates.

The notification job is quick to show. A typical default text reads "You have been added to REF MIXED 18:00 from the waitlist." Ask a chat assistant to rewrite it with your class names, the late-cancel rule and a way out, and an illustrative draft comes back: "Good news, a reformer's come free! You're now booked into Mixed Reformer, Thursday 6pm with [teacher]. Plans changed? No problem, you can cancel any time at no charge: [link]." Friendly and clear, and wrong on the one point that matters: "any time at no charge" contradicts a 12-hour late-cancel window. The corrected line is "Cancel before 6am Thursday at no charge; after that the late-cancel fee applies." Paste the fixed version into your platform's template, using its merge fields for the class and time rather than typing them.

The "am I in?" question needs care for a different reason. If your chat assistant isn't connected to the booking system, it can't know. A realistic failure: a student asks on Instagram whether she's off the waitlist for Saturday, and an unconnected assistant replies "You're currently number 2 on the waitlist, so there's a good chance you'll get in!" It made the number up. The rule for an unconnected assistant is to answer the general question and point to the real one: "Your account page shows your live waitlist position: [link]. If a spot opens before 45 minutes to class, you'll get a text straight away, and you're only charged if you're moved in."

Forecasting no-shows for a Saturday mat class

Here's how the numbers work for an illustrative Saturday 9am mat yoga class. The room fits 27 mats comfortably; the class is capped at 25 in the booking system and has had a waitlist every week this term. Export ten weeks of that class's bookings, late cancels and no-shows (most platforms have an attendance report you can download as a spreadsheet), remove student names, and use a prompt like this:

Attached: 10 weeks of attendance for our Saturday 9am mat yoga class,
capacity 25. Columns: date, booked, late_cancel, no_show, attended,
waitlist_at_start.
1. For each week, calculate empty mats at class time
   (25 - attended).
2. Give the average and the range.
3. If we overbooked by 1 or 2 (i.e. let 26 or 27 book), in how many of
   the 10 weeks would more people have turned up than 27 mats?
4. Flag any unusual weeks (holidays, weather) that distort this.
Show your working in a table.

An illustrative answer:

Week   Booked  Late cancel  No-show  Attended  Empty mats
6 Sep    25         2           1        22         3
13 Sep   25         1           2        22         3
20 Sep   25         0           1        24         1
...
Average empty mats: 2.4   Range: 0 to 5
Overbook by 1 (26 booked): no week would exceed 27 attendees.
Overbook by 2 (27 booked): 1 week of 10 would have had 28.
Unusual week: 22 Nov had 5 empty mats, possibly weather.
Recommendation: overbook by 3 to fill the average 2.4 empty spots.

Read it critically. The table and the overbook-by-2 finding are useful. The final recommendation contradicts the assistant's own working: if overbooking by 2 already overflowed the room once in ten weeks, overbooking by 3 would overflow it more often, and the average is pulled up by one bad-weather week. The sensible reading is to overbook by one, because the data shows it never overflows, and to review in a month. Always check the arithmetic on a few rows yourself, and don't let a confident last line override the table above it.

If your booking platform doesn't allow a cap above the class limit, you can get a similar effect by setting the class capacity to 26 and keeping one mat spare in the cupboard. Be open about it with teachers, and never do it for equipment classes.

Fairness rules students notice

A waitlist that runs automatically still has to feel fair, and students talk to each other. Decide these points before a regular complains, then write them on your booking page and in your chat assistant's answers:

• Members before drop-ins? Many studios want members to get priority. Mindbody has announced a priority waitlist feature for exactly this; on other platforms you may need to handle it with an earlier booking window for members instead.
• How many waitlists can one person join? A student who joins five waitlists for the same evening and takes whichever comes through blocks four other people. A limit of two at a time, or one per day, is common.
• Intro-offer students. A new student on an intro pack who never gets off the waitlist for your best classes won't convert. Some studios hold one spot in popular classes for newcomers; turning Pilates intro offers into memberships explains why the first three classes matter so much.
• Overlapping classes. A student waitlisted for the 6pm reformer and booked into the 6pm mat class as a fallback can be auto-promoted into both, then charged a late-cancel fee on whichever she drops. Check whether your platform blocks overlapping bookings; if it doesn't, add a line to the booking page asking students to drop their fallback when promoted, and waive the first fee when it happens.
• No manual favours. The fastest way to undermine an automatic waitlist is for staff to slot friends in ahead of it. If a teacher needs to add someone, it goes through the same rules.

When your chat assistant answers "why wasn't I moved in?", it should be able to explain the rule in one sentence. If the honest answer is "because a staff member added someone else", fix the process, not the chatbot's wording. Deciding when a chatbot hands over to a person helps for the complaints that do need one.

Six reformers and no room to overbook: a different job for AI

A studio with six reformers and a long waitlist on weekday evenings can't overbook, so AI's job is different. Forecasting there answers two questions: which late-cancel window gives waitlisted students a fair chance of getting in, and whether the evening demand justifies a 7:15pm class after the 6pm. If eight weeks of data show an average of seven people on the 6pm waitlist and almost all of them attend when promoted, that's a second class, not a better waitlist.

The sum for that second class is short. With six reformers, an illustrative drop-in or pack rate of $25 a class, and a realistic five of six filled once the class settles, a 7:15pm class brings in about $125 a week. If the teacher costs $45 for the hour, that's roughly $80 a week of margin before any new memberships it creates. Run it for four weeks before deciding: waitlisted students sometimes say they'd come to a later class and then don't, and the booking data after launch is the only honest answer.

A waitlist settings check, filled in

Run through this list once a term for each class type. An illustrative completed version for the reformer studio above:

• Mode per class type set deliberately? Yes: auto-promote for regular classes, confirm first for workshops.
• Promotion stops before the late-cancel window? Fixed this term: was 2 hours, now 12 hours, matching the window.
• Early-morning classes stop promoting the night before? Yes, 8pm.
• Messages can't go out overnight? Checked: no waitlist texts between 9pm and 6am.
• Charging rule stated on the booking page? Added: "You're only charged if you're moved into the class. We'll text you when that happens."
• Test done? Booked a test student onto a full test class, cancelled a spot, checked the text and the charge.
• Chat assistant knows the rules? Updated its answers page with the new cutoff and charging rule.

A filled-in version of that test, for the 6:30am reformer class with promotion set to stop at 8pm the night before:

Step 3 failing is the kind of thing only a test finds. In this illustration the 8pm stop had been set on the class template, but the recurring 6:30am classes already on the schedule had been created before the change and kept the old two-hour setting. Updating the existing classes, not just the template, fixed it. Re-run the failed step after any change.

The test is the item people skip. Ten minutes with a dummy class and a test account shows you exactly what a student sees, which is the only way to know your settings do what you think. For the reminder side of the same problem, reducing no-shows with AI reminders pairs well with a tidy waitlist.

Further reads

• AI Booking Systems for Appointment Businesses: What to Check — What to check in any booking system before trusting its automations.
• How to Build the FAQ Your AI Chatbot Needs Before Launch — The answers a chatbot needs before it can handle waitlist questions.
• Using AI in a Yoga or Pilates Studio While Keeping Your Voice — Make waitlist and reminder messages sound like your studio.
• One AI Inbox for WhatsApp, Instagram and Facebook Enquiries — Pull waitlist questions from every channel into one inbox.
• Automated Check-Ins for Gym Members: What to Send and When — What to send members between bookings, and when.
• How Salons Use AI to Rebook Clients and Fill Gaps in the Diary — Return intervals from your own history, a chair-side rebooking habit, and a way to match each empty slot to the five clients most likely to take it.
• How Driving Schools Use AI to Keep Instructor Diaries Full — Fill cancelled slots in an hour, replace pupils before they pass, cut dead travel between lessons and spot quiet weeks early, with the data each step needs.
• Can AI Cut Last-Minute Driving Lesson Cancellations? — Why the 72-hour confirm-or-release reminder matters more than the bot, how to spot pupils likely to cancel late, and how to refill the hour.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: Mindbody waitlist feature update page; Momence help centre, waitlist FAQs for classes; TeamUp help centre, waitlist overview.