Running a Squid Game Event in Singapore

Six Rounds, and What to Do With Everyone Who Is Out

In Singapore the themed elimination format popularised by the Korean survival series is sold by roughly four corporate operators, at published rates running from about $40 to $78 per person, for groups from 30 up to 300, across sessions of 2.5 to 4.5 hours. Local listings promise masked crew, prop rifles, game money, and a named sequence of rounds. What none of them publish is the thing that decides whether the afternoon works: what the rules of each round actually are, and what happens to the person who loses the first one at minute twelve of a three-hour programme.

That second question is the whole design problem, and it is the reason so many in-house versions of this format end with half the room on their phones. This page gives the round-by-round rules, the equipment, the timings and the elimination conditions, then does the arithmetic that shows why straight knockout cannot work in a workplace, then gives four structural fixes. It also gives you the argument about whether a company should run a themed version of a story about people being killed for debt, because that objection will be raised, and it is better raised in your own head than in the leadership meeting where you propose it.

What is the Squid Game format, and what do Singapore operators actually sell?

A themed elimination series is a sequence of short, rule-simple games played in a fixed order by the whole group at once, with a stated condition at the end of each round that removes some proportion of the players, wrapped in a single consistent visual theme. The theme carries the branding. The sequence carries the structure. The elimination condition is the mechanic that makes it a different animal from every other multi-game corporate format.

The three defining properties are worth separating, because operators blur them.

Fixed sequence, not free choice. Everyone plays round one, then everyone plays round two. Nobody chooses their order, nobody routes themselves, and nothing is unlocked by a clue. The organiser owns the running order absolutely, which is both the format's simplicity and the source of every scheduling risk in it.

Rule-simple children's games. Each round has to be explainable in under a minute to a person who has never seen it, because the theatrical pacing collapses if a round needs a five-minute briefing. Every game in the source material is a playground game.

A stated removal condition. At the end of each round somebody is out. Whether "out" means literally out, or costs a life, or costs points, is the single most consequential decision the organiser makes, and it is dealt with at length below.

Locally, published listings converge on a familiar picture. Per-person rates in the published bands sit at roughly $40 to $60 at one end and $72 to $78 at the other, with a minimum event fee from about $2,000 applying to smaller groups, which is the usual pattern in this market. Published group ranges start at 30 and run to 300, with one regional listing quoting a ceiling of 1,000. Published durations run 2.5 to 4.5 hours, and a half-day of roughly four hours is the common shape. Against the general Singapore team building bands, where an in-office single activity sits around $25 to $40 per head and a standard half-day with facilitator, MC and light catering sits around $45 to $90, this format prices at or slightly above a standard half-day, which is what the costume, the props and the higher facilitator count buy you.

The round lists operators publish overlap almost completely: the freeze game, the sugar-cookie carving, tug of war, the marbles round, the glass bridge, and a final. One published set drops tug of war and the marbles round and substitutes a lights-out negotiation segment. Another advertises five signature challenges ending in an arena round. Every local listing, without exception, states somewhere that nobody is really eliminated and everybody goes home together. That statement is doing two jobs: it reassures the buyer about the theme, and it quietly admits that the operators have already solved the arithmetic problem by removing the elimination. If you buy this format from a vendor, ask them exactly what replaced it, because a themed series with no consequence at all is a costume party with games in it.

How do you run each of the six rounds, and which ones are worth running?

This is the part that no local page publishes, so here it is in full. For each round: the corporate-safe adaptation, the rules, the round length, the equipment, and the removal condition. At the end of each, a plain verdict on whether the round is genuine team development or pure theatre, because a format built entirely from theatre rounds will be enjoyed and will teach nothing, and you should know which you are buying.

Two rules govern all six. First, no round may put a colleague in the position of personally removing a named colleague; the game removes people, never a person. Second, every round must have a stated finish time, announced before it starts, after which it ends whether or not it has resolved.

Round 1: The freeze game, commonly called red light, green light

Adaptation. Players advance from a start line towards a finish line while a caller, back turned, chants a phrase of fixed length. On the final syllable the caller turns. Anyone still moving is caught. In the source this is where the story establishes its stakes; in a workplace it is your opening round, and its real job is to get every single person onto the floor in the first five minutes.

Rules, in order.

1. Mark a start line and a finish line 15 to 25 metres apart. Indoors, 12 metres is workable if the floor is not slippery.

2. All players line up on the start line. If you are running the team-based structure below, players line up in team blocks so you can see who belongs to whom.

3. The caller faces away and chants. Vary the chant length deliberately between three and six seconds, because a predictable rhythm turns the round into a metronome and removes the only decision in it.

4. On the turn, the caller and one spotter identify movement. Two officials, not one, and they must agree. A single caller adjudicating 30 people will miss half of them and be accused of favouritism for the half they catch.

5. Anyone identified as moving returns to the start line rather than leaving the game. Full removal in round one is the single worst design decision available to you.

6. Players reaching the finish line stop there and become spotters for the remainder of the round.

Round length. 6 to 10 minutes including the brief. Cap it hard: if a third of the field has not crossed by the cap, call time and score by distance travelled.

Equipment. Floor tape or cones for two lines. A caller with a voice or a microphone. Optionally a phone on a tripod at the finish line filming the field, which settles disputes in four seconds and is worth the setup. No other kit.

Removal condition. Caught moving after the turn. In the recommended version this returns you to the start rather than removing you, and the round scores on who crosses and how many resets each team accumulated.

Verdict: mostly theatre, with a thin genuine layer. It is an individual test of impulse control with no coordination requirement, so it develops nothing on its own. Its virtue is that it takes the entire room into the game simultaneously with no equipment and produces the best photographs of the day. Keep it, and keep it first. One variation adds actual substance: run it in linked pairs, wrists loosely joined by a scarf, and any movement by either partner sends both back. Suddenly a synchronisation problem exists where none did before, and the round earns its place.

Round 2: The sugar-cookie carving round

Adaptation. Each player receives a brittle sugar disc stamped with a shape and must free that shape intact, using a pin or a toothpick, inside a time limit. This is the round most likely to create a real problem in an office, so read the safety note before the rules.

The safety note, which is not optional. Do not have participants make the discs. Casting them requires melting sugar to roughly 150 degrees Celsius and pouring it, and molten sugar sticks to skin and burns far worse than boiling water. Order the discs pre-cast from a caterer or a specialist supplier, or run the non-food substitute below. Then handle it as a catered item: a written allergen list on the table, since the discs are sugar and bicarbonate but the kitchen may not be nut-free or gluten-free, a stated alternative for anyone fasting or diabetic, and a bin plus wet wipes at every table.

The non-food substitute, if catering is a problem. Laser-cut thin acrylic or dense card with a scored outline works identically and removes the allergen question entirely. It is less photogenic and nobody minds by the second minute.

Rules, in order.

1. Each player draws a disc blind from a covered tray, so nobody chooses an easy shape. Use four shape difficulties and distribute them evenly across teams, not randomly, or one team gets four hard shapes and the round is decided by the draw.

2. Issue one pin per player and state clearly that the pin stays pointed down at the table at all times.

3. Start the clock. 10 minutes is right for a mixed group; 6 minutes is brutal and produces almost no successes.

4. A player who frees the shape intact raises it and a marshal records the time to the second.

5. A player whose shape cracks may claim one replacement disc at a stated cost, which is where the team structure enters: the replacement is charged against the team, not the person.

Round length. 12 to 15 minutes including the brief and the tray distribution. Budget two discs per person plus 20 percent, because breakage is the point of the round.

Equipment. Pre-cast discs, two per player minimum. One pin or toothpick per player. Trays, one per table. Wet wipes. Allergen sheet. A bin. Tables and chairs, which is the only round in the set that needs furniture.

Removal condition. Shape not freed intact when the buzzer sounds.

Verdict: pure theatre, and the weakest round in the set for team development. It is performed alone, in silence, at a table, and the only inter-personal content is that your colleagues can see you failing. It is also the most recognisable image from the source material, which is exactly why operators lead with it. Two fixes make it defensible. Run it as a relay, where one disc passes down a team of five and each player has 90 seconds before handing it on, so the shape is a shared object and the handover is the skill. Or set a team target, where a team of five must produce three intact shapes from five discs and may allocate its pins and its time however it likes, which forces an explicit conversation about who is steady-handed and who should not be trusted with the fine edges.

Round 3: Tug of war

Adaptation. Almost none needed. This is the one round from the source material that transfers to a corporate setting essentially unchanged, and it is also the round most likely to injure somebody, so the adaptation is entirely about safety and seeding.

Rules, in order.

1. Lay a centre line. Mark two lines 2 metres either side of it. A team wins when the centre marker on the rope crosses its line.

2. Teams line up alternating heavier and lighter pullers rather than front-loading the strong ones, and the anchor is the last person, with the rope running along the outside of the body, never wrapped.

3. Officials check hands and stance before the whistle. Rope wrapped around a hand, wrist or waist is the injury that ends your event, and it is checked, not requested.

4. Pull on the whistle. A pull is capped at 90 seconds. If neither line has been crossed at 90 seconds, the pull is scored on which side of the centre the marker sits.

5. Best of three pulls, with a 60-second rest between them. Three pulls, not five.

Round length. 20 to 30 minutes for six teams in a knockout, because each pull is short but the resets and rotations are not. This is the longest round in the set and the one that overruns.

Equipment. A proper tug of war rope, 30 to 36 metres, at least 30mm diameter, with no loops or knots at any point. Rope of the wrong diameter is the second most common cause of hand injury after wrapping. Centre marker and three floor markers. Gloves are a genuine question: they reduce rope burn and reduce grip, and the common practice is to allow them and require them to be the same type across both teams. Flat, non-slip ground. Do not run this on a polished indoor floor, a wet field, or anywhere near a wall.

Screening. State in the calendar invitation that this round is not suitable for anyone with back, shoulder, elbow or hand conditions, anyone pregnant, or anyone who has had recent surgery, and give those people a named role instead, which the guard structure below supplies neatly.

Removal condition. Losing team is out, or in the recommended structure, loses the round's points.

Verdict: genuinely good, with one large caveat. The physics of a rope means that eight people pulling in rhythm beat eight stronger people pulling out of rhythm, and teams discover this inside one pull. That is a real, immediate, physically undeniable lesson in coordination beating individual effort, delivered in 90 seconds without a facilitator saying a word. The caveat is that it is still substantially a strength and mass contest, so seed it. Either equalise approximate combined team mass when you form teams, or run it as a handicap where the lighter team's line sits half a metre closer to its winning line. Unseeded, the round is decided before the whistle and everybody knows it.

Round 4: The marbles round

Adaptation. This is the round that requires the most careful handling, because in the source material its entire dramatic function is that partners must betray each other, and that is precisely the content you cannot import into a workplace. The mechanic underneath it, wagering under uncertainty against a person you have to keep working with, is genuinely valuable. The pairing is what you change.

The rule that makes it safe. Never pair colleagues from the same team, and never let the outcome be framed as one named person removing another named person. Pair across teams, so the loss lands on a team, not on a relationship. If your group has any existing tension between two individuals, do not seed those two against each other, and do not leave the pairing to chance in a room where you know that risk exists.

Rules, in order.

1. Every player receives 10 tokens. Marbles, poker chips or coins all work. Tokens are individual property for the round but count into the team bank at the end.

2. Players pair off across team lines. The marshal announces the pairings; players do not choose.

3. Each exchange is a single wager. The holder conceals between one and three tokens in a closed hand. The guesser calls odd or even. Correct, the guesser takes the wagered tokens. Incorrect, the guesser pays that number.

4. Roles alternate every exchange. Exchanges continue until the clock stops.

5. A player reduced to zero tokens joins their partner as an observer for the remainder of the round and their team's bank takes the loss. They do not leave the room and they do not stand alone.

Round length. 8 to 12 minutes. It is short, it is quiet, and it is the round where the noise level in the room drops noticeably, which is worth watching for.

Equipment. Ten tokens per player, in team colours if you can, plus a small pouch each so tokens are not on the floor. A visible clock. Nothing else.

Removal condition. Reduced to zero tokens. In the team structure, the round scores on the team's total token bank rather than on survivors.

Verdict: the strongest genuine exercise in the whole set, and the most dangerous socially if you get the pairing wrong. Wagering under uncertainty against a person whose incentives oppose yours, repeatedly, with the ability to read them across the table, is exactly the negotiation skill most office roles use and almost never practise. It also produces the best debrief material of any round, because people can articulate what they were reading and how wrong they were. Get the cross-team pairing right and this is the round to build the event around. Get it wrong and you have staged a betrayal exercise between two people who share a manager.

Round 5: The glass bridge

Adaptation. Two parallel paths of tiles run from one end of the room to the other. At each step, one of the two tiles holds and the other does not. In the source, the wrong tile kills you. In a workplace version, the wrong tile is a marked floor tile that a marshal declares broken, and the player steps off the path and joins the queue as an observer or, better, hands the run to the next player on their team.

Why this round is more interesting than it looks. The first player has no information at all and a 50 percent chance at every step. The fifth player crosses on the accumulated corpses of the first four and needs to be lucky only on the tiles nobody has reached. The round is therefore not a test of luck: it is a test of who volunteers to go first, in what order the team sends people, and whether the information gained is actually transmitted. Almost no other corporate format models the cost of being the person who goes first.

Rules, in order.

1. Lay two rows of floor markers, paired, 10 to 16 pairs long. Paper plates, gym spots, floor tape squares or A4 sheets all work. Spacing is one comfortable stride, not a leap.

2. Before the round, a marshal privately fixes the safe side at each pair on a key card and hands a sealed copy to a second marshal. Two copies, because a single key card that goes missing ends the round.

3. Teams take a published order. Randomise it visibly, by drawing lots in front of everyone, so the sequence is not a management decision.

4. A player steps onto one tile of the current pair. The marshal calls hold or broken. Broken means the player leaves the path and the team's run passes to the next player, who restarts from the beginning but retains everything the room has learnt.

5. Broken tiles stay marked with a visible cross for the rest of the round. This is the mechanism that transmits the information, and if you skip it the round becomes pure chance.

6. The round ends when a team crosses, or on the clock.

Round length. 12 to 18 minutes for six teams if you run two bridges in parallel, which you should. On one bridge with six teams it takes 30 minutes and five teams spend it watching.

Equipment. Twenty to 32 floor markers, two key cards, cross markers or red cards for broken tiles, one marshal per bridge and preferably a second, and a clear floor with nothing hard within a stride of the path.

Removal condition. Stepping on a broken tile. Team-level scoring is total pairs crossed, so a team that reaches pair 11 out of 14 scores 11 and is not empty-handed.

Verdict: genuinely good, and the round that most rewards a real debrief. The behaviours it exposes, who volunteers, who is silently relieved when someone else volunteers, whether the team argues about order or defaults to seniority, and whether the fourth player actually used what the third one learnt, are all directly observable and all directly relevant to how the team handles risk at work. Watch the order in which teams send people; it is frequently the exact order of internal status, and pointing that out afterwards lands hard.

Round 6: The final

Adaptation. The source's final round is a fight between two people. Do not stage a fight, do not stage anything with physical contact, and do not stage anything where one person can be humiliated in front of the entire company as the closing image of the day. Substitute a head-to-head skill duel between champions nominated by the two surviving teams, at a station nobody has played yet.

What works as a final. Three properties: it must be watchable by everyone from one position, it must resolve within a few minutes, and it must be something neither champion has practised, so the round is not decided by who happens to own the relevant hobby. Reaction duels, accuracy duels and precision-under-time duels all satisfy this. Anything requiring skill built over years does not.

Rules, in order.

1. Each surviving team nominates one champion. Nomination happens before the discipline is announced, which turns it into a genuine allocation decision under uncertainty rather than a casting call.

2. The discipline is announced. Both champions get 60 seconds of practice, and the practice is public so nothing is hidden.

3. Best of three, alternating first attempts.

4. The rest of the room is arranged behind their team's champion, and the two eliminated-round guard teams run the scoring, which gives everybody a physical position in the finale.

Round length. 10 to 15 minutes including the nomination, the practice and the ceremony.

Equipment. Whatever the chosen duel needs, plus a space where 30 people can see one lane.

Removal condition. Losing two of three.

Verdict: theatre, and that is correct. A finale's job is not development, it is a visible ending that gives the event a shape and a photograph. The design error to avoid is making the finale worth so much that the previous five rounds were decorative, which the weighting rule below handles.

The honest summary of the six

Two rounds are genuine team development: tug of war, for the immediate physical proof that rhythm beats strength, and the glass bridge, for order, volunteering and information transfer. One round is genuine individual development that produces excellent group conversation: the marbles round, provided the pairing is cross-team. Three rounds are theatre: the freeze game, the sugar-cookie carving and the final. Theatre is not a criticism. A themed event needs its recognisable images, and a programme of nothing but development exercises is a training course. Just do not tell your leadership team that the sugar-cookie round builds collaboration, because it does not, and somebody in the room will have watched the show.

Why does straight elimination break, and what is the arithmetic?

Here is the structural failure that almost every write-up of this format skips, stated as numbers.

Take 30 participants and a six-round programme. Apply the obvious reading of the source material: each round removes half the field.

By round four, four people are playing and 26 are watching. By round five, two people are playing a game in front of an audience of 28 colleagues who paid the same price per head and took the same afternoon off.

Now put a clock on it. At an average of 12 minutes per round, rounds two through six occupy 60 minutes. The watching column across those five rounds totals 119 person-rounds, which is 1,428 spectator-minutes. Divide by 30 people and the average participant spends roughly 48 minutes of a 72-minute game block watching other people play. The median experience is worse than the average: the 15 people knocked out in round one, half the room, watch all 60 remaining minutes without touching a game again.

Fifteen people, one hour, nothing to do. That is the event you have booked if you take the format literally. It also gets worse rather than better with scale, because a 60-person group halving over six rounds leaves two players and 58 spectators, and the last three rounds of a 100-person event are effectively a stage show.

There is a second, quieter problem in the same arithmetic. Elimination sorts your organisation by whatever the early rounds happen to measure. If round one is the freeze game and round three is tug of war, then by round four you have selected for reflexes and mass, and the four people still playing are statistically likely to be the four youngest and fittest people in the company. The finale of your inclusive team event is then two of them, in front of everybody else. Nobody planned that. The arithmetic did it.

Fix 1: Eliminate teams, not individuals

Form six teams of five. Each round scores at team level, and the lowest-scoring team is eliminated rather than half the field.

Same six rounds, same 12 minutes each. Spectator-minutes fall from 1,428 to 840, a 41 percent reduction, and the shape changes even more than the total does. Nobody is out before minute 12, the last full round still has 10 people on the floor rather than two, and elimination is something that happens to a group of five who then have each other for company rather than to an individual who stands alone. That last point is not sentimental; a person eliminated alone at minute 12 disengages, while five people eliminated together immediately start commentating.

This is the single highest-value change in this document, and it costs you nothing.

Fix 2: Lives and buy-backs

Give every team three lives. A round loss costs a life rather than the game. Teams are only out at zero lives, which cannot happen before round four, and with six rounds a team that loses three can still be present for the finale as a scoring crew.

The buy-back sharpens it. A team on zero lives may re-enter once by winning a side challenge run in parallel with the next main round, at a cost: they re-enter with half their accumulated points. Now elimination is reversible, the eliminated have something to play for immediately, and there is a genuine decision about whether the buy-back is worth the point penalty. Run the side challenge in a corner of the same room so the buy-back is visible; the moment a team fights its way back in is the best moment in the format.

Fix 3: Parallel brackets

Split 30 people into two brackets of 15 in separate zones, running the same rounds simultaneously with two sets of marshals. Each bracket runs its own elimination down to one surviving team, and the two survivors meet in the finale.

This roughly halves the spectator load, because at any moment two rounds are live instead of one, and it produces a genuinely charged final because the two finalists have never faced each other. The cost is real and should be priced honestly: double the marshals, double the equipment for every round, and a room that can be divided so that two callers are not shouting over one another. Below about 40 people it is not worth it. Above about 60 it becomes the only sane option, and above 80 you should be thinking about three brackets or heats.

Fix 4: The eliminated become the guards

This is the fix that is both structurally sound and thematically perfect, and it is the one operators do not publish because it costs facilitator hours to design rather than to staff.

The source material already contains a second class of people: the masked staff who run the games. So build them into the programme. As each team is eliminated, it takes on an operational role for the remainder of the event, with a mask or a coloured jacket and an actual job:

• Round callers and spotters. The freeze game needs at least two spotters, and eliminated players are more attentive spotters than paid staff because they know exactly how the cheating works.

• Bridge marshals. The glass bridge needs somebody holding the key card and calling hold or broken. An eliminated team runs the second bridge in the parallel-bridge setup.

• Scorekeepers. One eliminated player per surviving team, shadowing them and recording their round scores on the master sheet.

• The tribunal. Disputes go to a panel of eliminated players rather than to the organiser. This is more useful than it sounds: it converts every argument about a marginal call from a complaint about management into a negotiation between peers.

• The finale crew. The last two eliminated teams set up, run and adjudicate the final round.

Three things follow. Eliminated players stay physically in the game with a role and a costume rather than sitting down. Your facilitator count drops, because by round four you have 15 unpaid, motivated officials. And the theme stops being decoration and becomes the structure, which is the difference between a themed event and an event with a theme applied to it.

The one thing to watch: give guards authority, not busywork. A guard who is genuinely allowed to call a step broken is engaged. A guard holding a clipboard nobody reads knows exactly what has happened to them.

The combination worth running

For 30 people, the recommended structure is fix 1 plus fix 2 plus fix 4: six teams of five, three lives each, one buy-back available, and every eliminated team taking an operational role. Nobody leaves the floor, nobody is out before minute 12, the finale has 10 players and 20 officials, and the spectator count at any moment is zero because everybody has a job. For 60 people and up, add fix 3.

How many people does it need, and how do you cut the teams?

Minimum. Four teams of four, so 16 people, is the practical floor. Below four teams, eliminating one team per round exhausts the field in two rounds and the arithmetic problem returns in a smaller form. Published local minimums for this format start at 30, which is higher than most formats in this market and reflects the fact that operators are pricing a staffed, costumed production rather than a set of games.

The comfortable band. 30 to 60 people, in six to twelve teams of five. Six teams is the cleanest number because it maps to a six-round programme with one elimination per round and a two-team finale.

Maximum. Published ceilings run to 300, with one regional listing at 1,000, and those numbers are achievable only with parallel brackets and a large crew. Run one bracket per 30 to 40 people and accept that above roughly 120 the event is a stage-managed spectacle with participation, not a workshop.

Team size. Five. It matches the tug of war minimum for a decent pull, it gives the sugar-cookie relay five 90-second segments inside a 10-minute round, and it gives the glass bridge five attempts before a team is out of players.

How to form the teams. Three rules, and they matter more here than in most formats because the elimination structure will amplify whatever imbalance you build in.

• Equalise approximate mass across teams, not for fairness in the abstract but because tug of war is in the programme and an unseeded pull is decided before the whistle.

• Mix departments, because an elimination format run along departmental lines produces a losing department rather than a losing team, and that is a genuinely worse outcome than doing nothing.

• Name a rotating captain per round. The captain is the only person who may nominate the champion, claim the buy-back or accept a marginal call. Rotating it means five different people have to make one consequential decision each.

Who cannot play, and what they do instead. Tug of war rules out more people than any other round: back, shoulder, elbow and hand conditions, pregnancy, recent surgery. The freeze game and the bridge involve sudden stopping and lateral stepping, which is a problem for knee, ankle and balance conditions. Do not handle this on the day. Ask in the invitation, and staff the guard roles deliberately with anyone who opts out, so opting out is a role rather than an absence.

How long does it run, and what can you cut?

A round is 12 minutes on average, but the average conceals a wide spread that will wreck your schedule if you plan on the average.

Tug of war and the glass bridge are twice the length of the freeze game, and both are the ones organisers under-budget. A six-round programme for 30 people therefore runs like this:

• Arrival, teams, theme brief, safety brief: 20 minutes

• Six rounds as tabled: 95 to 115 minutes

• One break, because the sugar-cookie round needs tables set and the tug of war needs the floor cleared: 15 minutes

• Buffer: 10 minutes

• Results, ceremony, debrief: 25 minutes

Total: 165 to 185 minutes, or roughly three hours. That sits inside the published local band of 2.5 to 4.5 hours and towards its lower half, which tells you that the operators quoting 4.5 hours are including a meal, a longer ceremony, or more rounds.

Compressing it. The floor is about 100 minutes: four rounds instead of six, dropping tug of war (the longest) and the sugar-cookie round (the one needing furniture), a single 10-minute brief and a 15-minute close. Cutting to three rounds is not compression, it is a different event. The one thing never to cut is the debrief, for the reason given below.

Extending it. Add rounds rather than lengthening them, because the format's energy comes from the reset between games. Good additional rounds that fit the theme and the constraints are a shape-memory round, a silent-coordination round and a rope-free strength-balance round. Beyond about eight rounds the theme wears out and people start asking when lunch is.

What kit and what kind of space does it take?

Split it three ways.

Cannot be improvised, and determines whether the event happens:

• A tug of war rope of the correct diameter and length, from a sports supplier. This is the one item where the wrong purchase causes injuries, and it is a rental rather than a stationery-cupboard item.

• A space with 25 metres of clear run for the freeze game and the pull, on non-slip ground, with nothing hard inside the run-off.

• Pre-cast sugar discs from a caterer, or laser-cut substitutes, with an allergen sheet.

• Two people running the event who are not participating, minimum, and more once brackets are involved.

Genuinely improvisable in-house:

• Floor markers for both bridges. Paper plates, gym spots or tape squares.

• Tokens for the marbles round. Poker chips, coins, bottle caps.

• Team identifiers. Coloured lanyards or wristbands.

• The guard costumes. A coloured jacket and a plain half-mask does the job, and the point is that the guards are visibly a class, not that they are screen-accurate.

• Scoring. A printed master sheet per round plus one spreadsheet.

• The key cards for the bridge, written on index cards and sealed in envelopes.

Rented or bought if the theme matters:

• Costume and prop hire, which is the visible line item that separates this format's per-head price from a generic multi-game afternoon.

• Sound. A caller without a microphone loses their voice by round three, and a room of 30 excited people will not hear an unamplified brief.

• Photography, because the recognisable images are half the reason this format is chosen.

• Additional facilitators, if you are not using the eliminated-as-guards structure. With that structure, the crew requirement drops sharply after round two.

The venue test. One clear space at least 25 metres long, non-slip, with a separate area or an adjacent room that can hold tables for the carving round. Function rooms usually fail the 25-metre test. Sports halls, futsal courts, covered outdoor areas and large event spaces pass. Outdoor venues pass on space and fail on weather, which in Singapore is a scheduling risk you must actually plan for rather than hope about: have a covered fallback for the pull and the run, or accept that a downpour ends your event at round three.

How do you score it so the last round does not decide everything?

Two scoring architectures work, and mixing them badly is the most common in-house error.

Architecture A: pure survival. The last team standing wins and there are no points at all. It is the most faithful to the source, it is the simplest to run, and it has one fatal property in a corporate setting: a team eliminated in round two finishes joint-fifth with a team eliminated in round two, having played 20 minutes of a three-hour event. There is no consolation, no ranking and nothing to talk about. Use this only for short, light versions of under 90 minutes.

Architecture B: points plus lives, recommended. Every round awards points to every team that plays it, and separately costs the lowest-placed team a life. Points and survival are two different currencies, which is what makes the format interesting: a team can be surviving and losing, or eliminated and second overall.

Score each round on placings rather than raw output, because the rounds produce wildly incompatible units. Six teams, first place 10, then 8, 6, 4, 2, 1. Raw units never leave the round.

Weighting the final. The finale must be able to change the winner and must not be able to erase five rounds. The workable rule: set the finale above whatever separates the top two teams on entry, and below whatever separates first from third. With the 10-8-6-4-2-1 ladder across five rounds, typical leader gaps are 4 to 8 points, so a finale worth 12 points to the winner and 6 to the runner-up sits in the right band. Announce this number to the teams before round one. A finale whose value is announced afterwards will be read, correctly, as the organiser choosing the winner.

Two variants worth knowing.

The no-elimination variant. Every round is scored, nothing is ever removed, and the theme is carried entirely by the costumes and the round names. This is what most local operators appear to be selling, and it is a legitimate product: everybody plays everything, the arithmetic problem never arises, and you have a themed Mini Olympics. Be clear-eyed that you have given up the one mechanic that makes the format distinctive. If you want the theatre without the risk, this is the right choice, and you should say so internally rather than letting people expect eliminations that never come.

The debt variant. Teams start with a notional debt and each round win pays some of it down. The last team in debt loses. It maps the source material's actual premise more closely than anything else, which is either the most thematically coherent option available or exactly the thing you should not do in a workplace, depending entirely on the argument in the next section.

Stopping one person carrying a team. The structural defence in this format is different from a station-rotation format, because rounds here are whole-team by construction. Three rules do the work. Every player must take at least one scoring turn per round where turns exist, which covers the carving relay and the bridge. No player may be the nominated champion more than once across the programme, which forces five different people to be exposed. And the captaincy rotates by round, so the person deciding the bridge order in round five is not the person who decided the pull order in round three.

Which single rule do you move to make it harder?

Shift one rule per tier and the difficulty tracks predictably.

Base version. Six teams, three lives each, buy-back available, no round ever removes an individual, the freeze game resets players rather than eliminating them, the bridge scores on tiles crossed, and the final is a light duel. Every eliminated team becomes guards. Nobody is ever standing alone and the theme is visual only. This is the correct version for a first attempt, for a mixed-age group, and for any organisation where the theme is already a slightly sensitive subject.

Advanced version. Three changes. Lives drop from three to two, so elimination becomes real by round three. The buy-back costs half the team's accumulated points instead of being free, which makes it a genuine decision. And the bridge runs with an information penalty: broken tiles are marked for 60 seconds only, then the marker is removed, so a team must actively remember rather than read the floor. Run this tier with a group that already competes happily and has been through a themed format before.

Hell version. Four changes, and be careful with the third. Lives drop to one, so every round is knockout. The freeze game eliminates rather than resets. The marbles round becomes cross-team with an open market, where any player may propose a wager to any player from any other team and the round becomes a trading floor for eight chaotic minutes, which is the best version of that round for a sales team and the worst for a quiet one. And a hidden reversal: announce before round five that the current last-placed surviving team may swap positions with the leader if it wins the bridge outright. That last rule is the only one in this document that reliably produces genuine resentment, so use it only where the group already competes without grudge, and never in a group where the leadership team is playing.

Where does an in-house version fall over?

Half the room is out at minute 12. Cause: individual halving elimination taken literally from the source. Fix: team-level elimination, lives, and guard roles, as set out above. This is the failure mode of the format, not a peripheral risk, and every other problem here is smaller.

Tug of war injures somebody. Cause: rope wrapped around a hand or waist, undersized rope, wet or polished ground, or somebody who should not have been pulling. Fix: physically inspect grips before every whistle, buy the correct rope, refuse the surface if it is wrong, and screen in the invitation rather than at the start line. This is the highest-consequence risk in the programme and the checks take 20 seconds per pull.

Somebody is burnt or has an allergic reaction at the carving round. Cause: participants making the sugar discs, or a catered item with no allergen information. Fix: never melt sugar with participants, order the discs pre-made, put the allergen sheet on the table, and have a named alternative ready for anyone who cannot eat them. Nobody should have to ask.

The event overruns and the finale is rushed. Cause: tug of war and the glass bridge budgeted at the same 12 minutes as the freeze game. Fix: use the round-length table above, run two bridges in parallel, cap every pull at 90 seconds, and protect the last 25 minutes before you finalise the round count.

The theme lands badly with part of the group. Cause: nobody thought about it before the invitation went out. Fix: the section below.

A marginal call turns into a real argument. Cause: one official adjudicating 30 people, with no published dispute process. Fix: two spotters who must agree, a phone filming the finish line, and the eliminated-player tribunal. Announce the process to everyone before round one so that the first dispute has somewhere to go.

The winners are the four fittest people and everyone can see it coming. Cause: an early-rounds mix that selects for reflexes and mass. Fix: alternate what the rounds measure. Physical, then fine-motor, then physical, then negotiation, then deduction, then a duel nobody has practised. The published order in this page is deliberately alternating and is not the order in which the source material presents the games.

The costumes do all the work and nothing is learnt. Cause: buying the theatre and skipping the two rounds that develop anything. Fix: keep tug of war and the bridge even when you are short of time, and run the debrief.

Should a company run a Squid Game theme at all?

This objection is real, it will be raised, and an HR lead who has not thought it through before walking into the leadership meeting will lose the room to the first person who says the obvious thing. Here is the argument on both sides, stated properly rather than dismissed.

The case against. The source material is a story about people so deep in debt that they consent to a lethal contest, and its explicit subject is that economic desperation makes people disposable to those above them. The games are the mechanism; the point is the hierarchy between the players and the masked staff who count them. A company, which is itself an economic hierarchy, restaging that story with its own employees as the players and its own managers or hired crew as the masked staff, is making a joke whose punchline is the relationship in the room. Some of your colleagues will find that tasteless, and they will be responding accurately to what the source actually says rather than being humourless. The objection sharpens where the workforce includes people with real financial precarity, where there have been recent redundancies, or where the word elimination has been used in a company communication in the past year. It sharpens again if anyone in the group has personal experience of debt or gambling harm, which is not visible to you and which nobody will disclose in order to opt out of an afternoon activity. And it is at its very worst in the debt-scoring variant described above, which converts the objection from subtext into the actual rules.

The case for. The games themselves predate the show by generations and belong to nobody: a freeze game, a rope pull, a wagering game with tokens, a stepping puzzle and a sugar sweet are childhood playground and street games across East and Southeast Asia, and running them is not a re-enactment of anything. Cultural properties with dark premises are routinely used as harmless aesthetics, and adults are generally able to hold both facts at once. The show is enormously well known, which is precisely what makes attendance easy and the invitation self-explanatory, and that is a real operational benefit when you are trying to get 30 people to turn up to something on a Friday afternoon. Most importantly, the corporate versions on offer invert the premise rather than restaging it: nobody is removed permanently, the eliminated become the crew, and the structural point of the format as designed here is that everybody stays in the room. An event whose entire design effort goes into ensuring nobody is left out is a strange thing to call a celebration of disposability.

How to decide, practically. Four questions, and if you answer yes to any of the first three, pick a different theme rather than defending this one.

1. Has this organisation made redundancies, restructured, or run a performance-management exercise in the last 12 months? If so, the word elimination is not available to you as entertainment, whatever the costumes say.

2. Is attendance mandatory? A themed event that some people find distasteful is a much larger problem when they cannot decline. If you keep the theme, make attendance genuinely optional with a visible, unpenalised alternative, and make the guard roles a real option rather than a consolation.

3. Would you be comfortable if a photograph of your senior leadership in masks, standing over eliminated staff, appeared on a public feed? That photograph will be taken.

4. Can you run it with the elimination inverted, so that being out means gaining a role rather than losing a place? If yes, most of the objection dissolves, and it dissolves because you actually changed the thing rather than because you argued about it.

What to do if you keep it. Say the quiet part in the invitation, in one line, in plain language: the theme is the games, not the premise, nobody is out of the event at any point, and anyone who would rather not can take a crew role. That sentence costs nothing and pre-empts almost the entire objection. Then avoid the two things that make it indefensible: do not have managers play the masked staff while their own reports play the contestants, and do not use debt, money owed or financial language in the scoring. Use lives, points, or the token bank instead.

What does it develop, and what does it not?

Two things, with the mechanism named rather than the virtue.

Sequencing decisions under acknowledged risk. The mechanism is the glass bridge order. A team of five must decide, publicly and in about 40 seconds, who steps onto a tile with a 50 percent chance of being wrong, knowing that the information their failure produces is what lets a colleague cross. That is an unusually clean model of a real organisational problem: the first person to try something absorbs a cost so that later attempts are cheaper, and organisations are chronically bad at recognising and rewarding that. Teams reveal their actual answer within seconds. Some send the most junior person, some send a volunteer, some send the person who is worst at the game in order to preserve their strongest player for the far end, which is the sophisticated answer and is arrived at by roughly one team in six unprompted.

Reading intent across the table. The mechanism is the marbles round. Repeated wagering against a person whose gain is your loss, with the ability to see their face, is the closest a team building format gets to a negotiation drill, and it produces specific, articulable observations rather than general impressions. People come out of that round able to say what they were reading and how consistently they were wrong, which is a debrief that writes itself.

What it does not develop. Collaboration between teams, in any form, because the architecture is competitive by construction and elimination makes it zero-sum. Where the underlying complaint is that two functions never speak, an elimination architecture can widen that gap rather than close it, and a collaborative build or a shared-objective format is the correct instrument instead. It does not develop trust: trust requires depending on someone whose behaviour you cannot verify, and every round here is adjudicated and observed. It does not develop communication under complexity, because every round is deliberately rule-simple. And it does not develop psychological safety, which is worth stating explicitly because the format arguably works against it: an event whose structure is public failure in front of colleagues asks people to take a visible risk in exactly the conditions where a hesitant person will not. Whether that is a feature or a fault depends on the group in front of you.

It also does not fix anything. It shows you, quickly and legibly, how a team decides who goes first, and that is diagnostically useful to a manager who intends to do something with the information afterwards.

Which debrief questions matter once the last round ends?

Twenty minutes, immediately after the ceremony, in teams first and then to the room. The guards join their original teams for this. Keep the round scoreboard visible, because the gap between how a team felt it did and where it actually placed is itself a prompt.

1. On the bridge, who stepped first, and how was that decided? Was it decided at all, or did somebody just move while the rest of you were still talking?

2. If you were the first to step, what did you think you were buying for the team? Did anyone acknowledge it afterwards?

3. When your team lost its first life, what changed in how you played the next round? Name a specific behaviour, not a mood.

4. In the marbles round, what were you actually reading in the other person? Give one concrete tell you thought you had spotted. Were you right?

5. In tug of war, what did your team do differently between pull one and pull three? Who suggested it, and how long did it take for the team to accept the suggestion?

6. Who did your team nominate as champion for the final, and why that person, given you did not know what the duel would be?

7. For the teams eliminated early: what did you see as a guard that you could not see as a player? Name something specific a team was doing wrong that they did not know they were doing.

8. Which round did you personally contribute least to? What would have had to be different for you to contribute more, and was that within your team's control?

9. Was there a moment when you disagreed with your team's decision and did not say so? What was the cost of not saying it, and what stopped you?

10. The scoring rewarded consistency across six rounds rather than winning any single one. Look at the board. Does that match how your team was actually playing, and if not, which round did you over-invest in?

11. If we ran the same six rounds tomorrow with the same teams, name the one rule of your own you would change. Not three. One.

12. What did you learn about a colleague in the last three hours that you would not have learnt in three hours of meetings?

Question 7 is the one to protect. It is the only question here that a format with a guard structure can ask, it is the payoff for having designed the eliminated back into the event, and the answers are consistently the sharpest of the session, because a person who has just spent 40 minutes watching a team from the outside sees things that team cannot.

What else is this format called, and what is it not?

The generic descriptor for what you are buying is a themed elimination series. Locally, the same product is sold under the show's name where the operator is willing to use it, and under a set of deliberate near-misses where the operator is not: survival games, a themed challenge series, a regional-childhood-games variant renamed after a local sea creature, and simply a themed team challenge. In this market a single format routinely carries ten different names, most of them coined by whoever is selling it, and the name tells you nothing about the mechanics. Read the round list, not the heading.

Five questions cut through the naming and get you comparable quotes, and a proposal rarely answers any of them unprompted:

1. Is anybody actually eliminated? If yes, at what level, individual or team, and what does an eliminated person do for the rest of the session?

2. How many rounds, and what is the running order? Ask for the order, not just the list, because the order is where the design lives.

3. Which rounds involve food, and how are allergens and dietary requirements handled?

4. Is tug of war included, and what rope, what surface, and what pre-pull safety check?

5. How is the finale weighted against the earlier rounds, and is that number published to participants before round one?

The adjacent formats it gets quoted against, and how to tell them apart:

Mini Olympics or telematch. Parallel, not sequential. Every station is live at once, teams are distributed across them and rotate on a signal, and nothing is ever removed. The design work there is the balance of the station mix and the weighting of a finale. Here it is the running order and the handling of the eliminated. If a quote describes teams rotating simultaneously between stations, you are being sold a Mini Olympics in a costume, which may be exactly what you want, but should be priced as one.

Amazing Race. Route-based and clue-gated. Teams move between checkpoints and unlock the next by finishing the current one, so each team follows its own path and no two teams are doing the same thing at the same moment. This format is the opposite: everyone does the same thing at the same time in an order fixed by the organiser, and there are no clues because nothing is locked.

Running Man. Also elimination-based, and the closest relative, but the mechanic is different in kind. There, elimination happens through name tags, concealed identities and betrayal, players are removed by other players continuously rather than by rounds, and there is no fixed sequence of games. Here, the game removes people at stated moments, never another player directly, which is exactly what makes it usable in a workplace where the other one needs careful handling.

Minute To Win It. Sixty-second props-based units in a single room, run for laughs, with no elimination and no theme beyond the timer. If your venue is a function room and you have no 25-metre run, this is what you can actually run there.

Escape room. Collaborative rather than competitive, small groups, and puzzle-gated. Nothing in this page transfers.

Can you run a Squid Game format at PLAYON?

Not the whole thing, and the honest answer is worth more to you than a hopeful one. PLAYON is an indoor attraction floor, and the format described on this page needs a 25-metre run, a rope, a caterer and a set of floor markers. Say plainly what is not there:

• No tug of war. There is no rope, no pull lane and no surface set up for it. This is the round the venue cannot supply in any form, and it is one of the two genuinely developmental rounds.

• No sugar-cookie carving. No catering of that kind, no tables set for it, and nothing that substitutes.

• No marbles round. It needs tokens and a quiet table, which is not what an attraction floor provides, and the ambient noise works directly against the round's only real quality.

What the venue does supply is two, at most three, of the six rounds, and it supplies them better than a rented function room would, because the scoring is automated and the equipment is permanent.

The freeze round, on the lit grid or a camera-tracked attraction. Colorful Grid is a floor of independently lit tiles: a colour and position cue flashes, and the player must move to that tile before the timer expires. Correct tiles inside the window score, wrong tiles and timeouts both deduct, consecutive correct hits build a streak multiplier that any error resets, and the result posts automatically to a live shared leaderboard with no human judge. That maps onto the freeze round's underlying demand, which is respond correctly on the signal and hold your discipline when the tempo changes, and it does so with automated adjudication, which removes the two-spotter problem and the arguments entirely. Multiple players can be on the grid at once and the page references battling a group of six, though an exact maximum is not published. The camera-tracked attractions carry the freeze idea more literally: the Interactive Motion-Sensing Game and the Body-Sensing Arena both read a real-time skeleton map through overhead and depth sensors and score accuracy and timing against on-screen cues, with no controller, so a cue that says hold still is registered as reliably as a cue that says move. Details sit on the individual pages for Colorful Grid, Motion-Sensing Game and Body-Sensing Arena.

The bridge round, as a points stand-in. This is the most interesting partial fit and it needs an honest caveat. The grid has no bridge mode and the venue does not publish one. What it has is the mechanical property the bridge round needs: a floor where landing on the wrong tile costs you, immediately and automatically, and where a run of correct choices compounds through the streak multiplier while a single error wipes it. A tile that deducts points is a workable stand-in for a pane that gives way, with the crucial difference that the cost is a score rather than removal from the game, which is the corporate adaptation you wanted anyway. Layer your own rule on top: the team's run continues to the next player when the multiplier resets, the room watches the pattern of errors accumulate on the shared board, and the fifth player plays with everything the first four learnt. The machine supplies the floor, the timing and the scoring. You supply the sequencing rule, and you should tell participants that, because it is your design, not a mode on the attraction.

The final duel, head to head. Reaction Testing suspends padded rods on a release mechanism above the player, drops them at random intervals, and scores a point for each catch, with missed rods costing nothing and intervals shortening as the round goes on. It runs head-to-head between two players, the score posts to a leaderboard automatically, and nobody in your group will have practised it, which is the whole requirement for a finale. Details sit on the Reaction Testing page.

What this adds up to. Two rounds of six, or three if you count the freeze and the bridge stand-in as separate rounds run on the same floor, which they can be with different rules. That is a partial fit, and it means one of two things for your plan. Either you run a shortened, venue-native version of three or four rounds inside a two-hour booking and accept that it is a themed elimination series without the rope or the sugar, or you run the full six-round format somewhere with a field and treat the venue as a different event entirely. What you should not do is book the floor expecting the six-round programme in the show, because half of it is not in the building.

Practical notes if you do run the venue rounds. Grip socks are required on Colorful Grid, Reaction Testing, the Motion-Sensing Game and the Body-Sensing Arena, and bare feet are not permitted on any of them, so put that in the calendar invitation rather than discovering it at the door. Colorful Grid carries a health caution from the venue that belongs in your invitation word for word: "It is not recommended for anyone with knee, ankle or balance conditions that make quick lateral steps difficult." The camera-tracked attractions are separately not advised during pregnancy or with heart, back, knee or balance conditions, so the guard and scorer roles are the right destination for anyone screened out. Rings, watches and bracelets come off for Reaction Testing and the Body-Sensing Arena, which means a bag drop. Maximum simultaneous players are not published for any of the four, so confirm at booking if your round structure depends on a specific number, and note that the scoring, while automated per attraction, does not aggregate across attractions: somebody on your side still transcribes the round result onto a master sheet.

And the plain version, for a brief that says field, rope, costumes and 100 people: that brief belongs outdoors, with a crew, and this is not the venue for it.

Squid Game format questions, answered

How much does a Squid Game team building event cost in Singapore? Published per-person rates run from about $40 to $78, with minimum event fees from around $2,000 for smaller groups. That sits at or slightly above a standard half-day rate, and the premium buys costumes, props and a higher crew count.

How many people do you need for a Squid Game format? Sixteen people in four teams of four is the practical floor, though published local minimums start at 30. The comfortable band is 30 to 60 in six teams of five, and published ceilings reach 300 with parallel brackets and a large crew.

How long does a six-round Squid Game event take? About three hours for 30 people, including briefs, one break, the ceremony and a debrief. Published local durations run 2.5 to 4.5 hours. The playable floor is roughly 100 minutes if you cut to four rounds and drop tug of war.

What happens to people who are eliminated in round one? They become the crew. Eliminated teams take real operational roles as callers, spotters, bridge marshals, scorekeepers and a disputes tribunal, which is thematically apt and keeps everybody physically in the game rather than sitting down watching.

Is a Squid Game theme appropriate for a corporate event? It depends on your organisation. If there have been redundancies or a performance exercise in the last year, choose another theme. Otherwise, invert the elimination so being out means gaining a role, keep financial language out of scoring, and make attendance optional.

Can you run the dalgona round safely in an office? Yes, if you order the sugar discs pre-cast. Never have participants melt sugar, which reaches around 150 degrees Celsius and burns badly. Provide a written allergen sheet, an alternative for anyone who cannot eat them, and wet wipes at every table.

Which Squid Game rounds are actually good team building? Tug of war and the glass bridge. The rope proves within one pull that coordinated rhythm beats raw strength, and the bridge forces a public decision about who goes first and whether information is transmitted. The marbles round is excellent if pairs are cross-team.

How do you run the glass bridge round without a bridge? Lay two parallel rows of floor markers, ten to sixteen pairs long, with one safe tile per pair fixed on a sealed key card. A marshal calls hold or broken, broken tiles stay visibly crossed, and teams score on pairs crossed.

What is the difference between a Squid Game format and a Mini Olympics? Sequence and consequence. A Mini Olympics runs every station in parallel with teams rotating on a signal and nothing removed. This format runs one game at a time in an order the organiser fixes, with a stated removal condition after each.

Do you need professional facilitators to run it? Two non-playing organisers is the minimum, plus a second pair if you run parallel brackets. Beyond that the eliminated-as-guards structure supplies your officials for free, and by round four you have fifteen motivated, unpaid, attentive spotters and scorers.

How do you stop the fittest person winning everything? Alternate what each round measures, and score at team level. Physical, then fine motor, then physical, then negotiation, then deduction, then an unpractised duel. Also cap champion nominations at one per person, so five different people must be exposed.

Can the format run indoors in a single venue? Partly. The freeze round, a bridge stand-in and a head-to-head final all run indoors on reactive floors and camera-tracked attractions with automated scoring. Tug of war needs 25 metres and a proper surface, and the sugar round needs tables and catering.