// Gamification for work teams
Five cooperative games for teams in the browser, with a live facilitator. Each person sees a different part of the problem, so the only way forward is to talk, coordinate and decide together. Afterward, a debrief built on observable facts: who had what information, and when they shared it.
Match in progress · Blind Operations
// What it is
Gamifying a team isn't about slapping a score on the work. Here it means one thing: building a situation where the group truly has to cooperate for real to achieve something concrete, with information split the way it's split in a real organization. What gets practiced during the game — asking, briefing, confirming, prioritizing, deciding — is exactly what's needed on Monday.
There are no teams competing against each other for a score. The result belongs to the group: either the mission gets solved together, or it doesn't.
The server shows each person only part of the problem. Whoever has the procedure can't see the controls; whoever reads the readings doesn't have the manual.
No single piece works alone. Before anyone can act, someone else has to have said something first: conversation is the game's core mechanic.
Everyone knows what has to be achieved and watches it move on the same screen. It's not about looking good: it's about bringing a system back online, or getting the submarine out of the trench.
The clock belongs to the server, and it's visible. It forces decisions with whatever's on hand — exactly what happens at work when the full picture doesn't arrive in time.
A wrong move doesn't eliminate anyone: it leaves a logged incident, with the time and what happened right before it. The conversation happens afterward; during the match, the team recovers and keeps going.
The part that turns a match into learning: the timeline of what happened, errors grouped by type, and scenario questions to talk it through as a group.
Scores, timers and wins are decided on the server, not in the browser. No one can get ahead by having a better machine or by jumping the rules.
// How a session works
People, areas and job roles: added by hand, imported from an Excel or CSV spreadsheet, or synced through the REST API. Each organization gets its own data space, with its own branding and its own address.
Pick the game, the scenario, the length, the difficulty and how many teams play in parallel. Assign who facilitates as Game Master and, if you want, schedule the session for a date.
Each person gets a personal, single-use magic link by email. No passwords and nothing to install: they enter the lobby, pick an avatar and color, and mark themselves ready.
Roles are dealt out in secret, each screen shows a different part of the problem, and the team talks. The Game Master follows the live progress, pauses, adds minutes, and sends hints and messages.
Everyone sees the result at the same time. The facilitator opens the debrief with the timeline, errors by type, and the scenario's questions to talk through what happened.
// The golden rule
It's the server that filters what reaches each screen, not a good-faith agreement between people. That's why no single member can solve it alone, and that's why the game doesn't reward whoever knows the most: it rewards the team that best shares what it knows.
For example, in Blind Operations: whoever has their hands on the controls doesn't have the instructions, whoever reads the readings can't see the map, and the final code belongs to a fifth person who can't touch anything.
Can be someone from within the organization, and doesn't play: facilitates. Sees the full state of the game — without the per-role filter — and can step in without breaking the rules, because the server decides everything.
The Game Master panel shows every team's progress and the information players don't see, with no per-role filter.
Pausing freezes the clock for everyone and blocks commands. It can also add minutes, one at a time, up to thirty.
To one person, one team, or everyone. There are nudge hints — which only say which two stations need to talk — and explicit hints.
Global announcements or ones to a single team, which show up in the chat as a message from the facilitator.
If someone has to leave, their panel moves to another person during the match without restarting anything.
Closes the session with the timeline, the incidents and the scenario's questions. This is the part that turns it into learning.
Anyone can jump into Play solo: they pick the game, the spot they want to fill and the level of their teammates, and bots take the remaining spots. The Game Master can also fill a session's empty spots with bots before it starts.
Slow to react, shares little unless asked, and sometimes reads out a wrong number and corrects it a few turns later. Good for practicing the habit of asking and double-checking.
Shares what it sees without being asked, asks for what it's missing, and rarely makes mistakes. The recommended level for getting to know a game.
Fast and precise, and anticipates what's coming. Good for seeing the full arc of a well-played match.
If someone loses their connection, they get a grace period to come back. On reconnecting they recover their role, their screen and the match's state; everyone else sees their connection status in text, not just a color.
The app's notification bell warns 24 hours, 1 hour and 10 minutes before a scheduled session, when the lobby opens, when a hint arrives and when the match ends. 24-hour and 1-hour emails back it up.
A voice channel per team plus a general channel with push-to-talk. If voice isn't enabled, team chat is enough: the games are designed to be playable entirely by typing.
If people are missing, the Game Master fills the empty spots with bots before starting. They're rules with memory, not generative artificial intelligence: they decide from the same filtered screen a person would see.
// Where it applies
Not every game fits every situation. Each one puts the focus on a different problem, so the choice starts with what the team needs to talk about afterward.
Someone who just joined doesn't yet know who to ask for what. A match has them requesting data from other areas from minute one, without the cost of a real mistake.
Best-fit game
Because each area holds part of the story: reconstructing it means asking where each piece of data comes from. It's a map of how information moves through a company.
Areas that hand off work to each other all day and almost never sit down together. The game puts the dependencies on the table, with numbers.
Best-fit game
Because each team manages different areas and only sees a summary of the rest: the bottleneck always shows up at the border, never inside a single area.
Practicing the protocol of saying a number, repeating it back, and confirming it before acting. It's a habit you train, not a talk you sit through.
Best-fit game
Because the procedure is written at one station and the hands are at another. There's no way forward without reading it out loud and waiting for confirmation.
Before kicking off, it helps to agree on what the team is willing to give up. A simulated crisis gets that conversation out of the way.
Best-fit game
Because every vote moves six indicators at once — client, cash, morale, quality, time and reputation — and forces the team to say out loud which one matters most.
Everything happens in the browser, in real time, with team chat and a general chat. It works the same whether the group is scattered or everyone's in the same room.
Best-fit game
Any of the six
Because the server already hands out the information: nobody has to deal out cards, and the group doesn't have to be physically together.
Whoever leads isn't always whoever has the data. These games split the two apart on purpose.
Best-fit game
In Crisis Room there's a mode where whoever leads confirms the committee's decision. On the submarine, the Commander gives orders without touching a single control: all they can do is talk.
The boundaries between areas just changed, and nobody knows yet where work is going to get stuck. Better to find out in a simulation.
Best-fit game
Because the game splits people and budget across connected areas and shows, round by round, what's piling up and where. New dependencies get tested without real clients on the other end.
An activity the group can actually talk about afterward: something that really happened, not just another talk.
Best-fit game
Because it's the catalog's most real-time-pressure game: five chained phases, a simulation that runs on its own, and a concrete ending — the submarine gets out of the trench, or it doesn't.
// What it improves and reinforces
A skill isn't trained by naming it on a slide: it's trained by putting the team in a situation that calls for it. This table shows, for each game, how each skill shows up during the game. The empty cells are honest: that game just doesn't exercise that particular skill.
| Skill | Blind Operations | Knowledge Files | Dependency Network | Crisis Room | Silent Depth |
|---|---|---|---|---|---|
| Clear communication Saying the complete piece of data, with its unit and its proper name. | Reading out exact numbers and waiting for confirmation | Naming the evidence, not the conclusion | The forecast is only useful if it's said in time | Holding a position for two and a half minutes | Heading, depth and speed, always with a unit |
| Active listening Repeating back what you heard before acting on it. | Whoever acts repeats out loud what they were told | Connecting your own evidence with another area's | Listening to the analyst before confirming the plan | Changing your vote because of data another role brought | One voice at a time, and confirming before executing |
| Coordination Sharing the conversation when everyone has something to say. | One system at a time: someone has to order the turns | Agreeing on who reads, who connects, and who organizes | Five areas and a single pool of people | Reaching a single decision with five different dossiers | Five stations, one hull |
| Systems thinking Seeing how a decision affects the rest of the system. | — | Reconstructing the full causal chain, not a culprit | What one area decides changes the next one's load | Every vote moves six indicators at once | How power gets distributed affects the whole submarine |
| Prioritization Choosing what gets solved first when there isn't enough time. | Which system to tackle first, with five open at once | Which thread to pull with the time that's left | Priority and reinforcements, round by round | Which indicator gets sacrificed to prop up another | Which breakdown gets fixed before the others |
| Decisions under pressure Deciding with incomplete information while the clock runs. | The satellite window closes either way | Submitting the report even if a link in the chain is missing | Confirming the plan without seeing the full forecast | Voting in 150 seconds, every round | The simulation runs on its own: not waiting is part of the game |
| Resource management Allocating what's scarce and doesn't come back. | — | — | People and budget shared across areas | Cash, morale and quality act as a budget | Power, decoys and torpedoes, all limited |
| Trust and psychological safety Being able to say "I got it wrong" or "I didn't understand" at no cost. | A mistake leaves an incident, never a culprit | A mismarked lead gets corrected, and the team moves on | The bottleneck belongs to the system, not to one area | There's no single stored correct answer | Flagging a mistake early costs less hull |
| Organizational knowledge Understanding who knows what, and how the work chains together. | — | What each area holds onto, and what won't come out without asking | How Marketing, Sales, Legal, Operations and Billing chain together | What each area looks at when there's a crisis | — |
Scroll the table to see all six games.
The closing conversation isn't about people: it's about what's visible in the game's timeline. Three things the timeline reveals, and all three can be discussed without singling anyone out.
// The games
All of them are cooperative and all of them split the information between the stations. What changes is the problem: a control room, a cross-department investigation, a dependency network, a crisis committee and a submarine.
No one has the full picture.
Aurora-7 is a weather station at 3,200 meters. Last night a lightning storm burned out the link and left it mute: no one knows if the people at the shelter are okay. The satellite passes overhead once a day, and that window has already started. There are five systems to bring online — Power, Frequency, Routing, Pressure and Transmission — and none of them can be solved alone.
Each of you walks into the room through a different door and sees a different slice of it. Whoever has their hands on the controls has no instructions; whoever has the manual can't see a single light. The readings belong to someone else, the antenna map to another, and the authorization code to a fifth person. On its own, every screen is an incomplete puzzle.
The room's console: the power breakers, the frequency dial, the patch panel with its ports and nodes, "VALVE B" and "VALVE C" with their state spelled out (OPEN or CLOSED), Pump A (ON or OFF), the vent button and the transmission keypad. You also see which systems are already online and the Game Master's hints, which show up inside the system they belong to.
The live instruments: the "BUS PHASE" screen with the phase's name and color, the load gauge, the overload warning, the signal oscilloscope, the current dial reading, routing integrity (how many links are correct out of how many), the pressure gauge with its safe range and the temperature, and the alerts.
The map with the eight antennas — each one with its sector and channel; the active one has a pulsing ring and a "▶" in front of its name, the rest are decoys — and the network diagram that says which node belongs to which port.
The manual, split into tabs: the table that translates each bus phase into breakers, the frequency formula by sector (base and step in MHz), the safe pressure range with its rules — more or fewer depending on difficulty — and a notes page. You can't see a single instrument.
The objectives board with the breakdown of who contributes what to each system, the four-digit authorization code, the number of transmission attempts, and a facilitation guide. You can't touch controls: your only command is "Request attention on…"
Exact numbers and names. An outdated reading, a vague "the one on top" or a lever pulled before the table arrives are three typical reasons a room gets stuck.
Every area holds part of the truth.
On Monday, March 3rd, Grupo Almendra launched Portal Faro for Distribuidora Río Claro. Within hours the client suspended its use. No one knows why: each area holds part of the story, and none of them has the full picture.
Sales, Technology, Operations, Legal and Finance receive emails, tickets, memos, photos, invoices and credit notes. Fifteen pieces of evidence in total, split differently every match. Some explain what happened; others are true but unrelated.
The emails and commercial documents for the Río Claro account: what was promised, to whom, and by when. You see your own evidence in full and, on the board, whatever the rest of the team has already shared.
This quarter's roadmap, the deployment notices, and the platform team's chat. You know what shipped, at what time, and in what state.
The day-to-day tickets and rollout plans, plus the client's complaint. You know what got stuck, since when, and what it was waiting on.
Policies, memos, and clauses: what could be done, what needed review, and what happens if it isn't followed.
Purchase orders, invoices, and credit notes for the client. You know what was signed, when, and what stopped being paid.
What evidence each area has, and what date it carries. The useful discussion isn't who's to blame, but what happened before what.
Your decision changes everyone else's work.
The busiest season of the year has arrived. A network of five connected areas — Marketing, Sales, Legal, Operations and Billing — has to keep the flow going round after round. The areas are split between teams: each team sees its own in detail and only a summary of the rest.
Every round brings an event — a campaign that goes viral, an audit, a transport strike — and teams split a shared pool of people and a shared budget: what one team takes, the other doesn't have. Each area gets people, money and a priority: quality, balance or speed. Once everyone confirms the plan, a simulation runs the flow through the network and updates seven shared variables.
The panel for the areas under you: fixed staff, productivity with the event's effect applied, quality, backlog, how many reinforcements each area accepts at most, the event note for that area, and the inflow and outflow. You also see how many people and how much budget are still free. You can't see the forecast.
Everything the Lead sees, plus the forecast for their areas during planning: demand, capacity, how much will get processed, how much will be left waiting, the load percentage, the band (stable, strained or saturated), and the projected quality.
Where work is going to pile up next round, and how to split a pool of people that isn't enough for every area.
Deciding under pressure, together.
Nimbo Sistemas is six weeks from launching Faro, its logistics platform. This morning, Transportes Ceibo — the strategic client, the success story they were going to show the market — canceled the project. The crisis committee convenes.
Each role arrives with a different confidential dossier: Leadership knows what the board expects, Finance has the cash flow, Sales knows what clients really think, Operations knows the technical state and how tired the teams are, Legal has the clauses. No one sees the full picture, and the options on the table arrive without saying what their consequences are.
A dossier with the strategy, what the board expects, and what's said behind closed doors. In your permanent file: your vote breaks ties, and in "Leader Decides" mode you confirm the final decision. This year's goal is to launch Faro without losing key people.
The cash flow, the costs, and the margin on every path, with amounts expressed in "cash points." Your file reminds you that the board wants the crisis closed with Cash at 30 or above, and that the costs are this week's estimates.
What clients say and what they really value, the market rumors, and what the prospects expect. Your file reminds you that Ceibo was the success story and that two big prospects are watching.
The team's real capacity, Faro's technical state, and how much each squad can take. Your file reminds you there are three active squads, that the routing module is the product's technical heart, and that every change in priorities costs days.
The contracts: exit clauses, penalties, warranties and commitments no one else sees. Your file makes clear your job is to report risks, not decide for the team.
Which indicator the committee is willing to sacrifice, and which confidential data has to go on the table for the vote to make sense.
One crew, one hull.
A mesh of autonomous sonar buoys — the Red Cazador — seals off the Fosa Nahuel. The submarine S-71 Abismo has to cross the trench undetected, find the control buoy hidden among decoys, take it out with a torpedo, and get out before the network resets.
Five stations, one hull. The Helmsman can't see the chart. Sonar doesn't touch the rudder. Engineering doesn't have the damage control manual. Weapons doesn't know the launch code. The Commander has the navigation chart but no instruments. The only way to operate the submarine is to talk.
The mission chart with the three legs — each with its heading and a recommended depth note — the acoustic signature book (which band and pulse belong to the control buoy), the launch code, and the available orders.
A large compass with the current heading and the ordered heading, a depth gauge with the current needle and the ordered mark, and the speed telegraph with four positions. Nothing else: no chart, no seafloor, no contacts.
The circular sweep display with contacts as blips — bearing, approximate distance, band and pulse — a vertical profile with the thermocline and seafloor lines, the clear corridor between the mines shaded onto the compass, your own noise, and the exact detection meter (0 to 100).
The reactor with its output, its temperature, and its scram status; the five power-allocation bars with the available budget; the compartment grid with each one's flooding level and casualty; and the progress of the repair in hand ("step 2 of 3"). You don't have the procedure.
The two tubes with their status, the bearing and depth controls for the firing solution, the code keypad, the decoys, and the attempts. On the right, the damage control manual: the procedure for every casualty type, plus restarting the reactor and the rudder.
Headings, depths and breakdowns, in that order and with complete numbers. Here a misunderstanding doesn't get discussed later: it gets paid for on the spot, in hull or in detection.
You're the last line of defense.
It's 11:40 PM at Cardal Logística S.A., and the night shift is in charge of the company's shared inbox (guardia@cardal.com.ar). Emails, internal chat messages, texts and voice notes arrive in three waves: legitimate, urgent requests — an invoice due tomorrow, a vendor with a delivery, a coworker who needs a signature — mixed in with phishing, fraud and impersonation attempts. Nobody can classify a message alone: Inbox sees the message the way any employee would, Analyst can inspect the technical layer Inbox can't see, and Context knows the organization — the org chart, who's on vacation, vendors and their registered account, company policy — and can call the registered number to confirm an unusual request.
There's no answer you can memorize: every match draws its messages from a seed, so the second attempt is never the same exam. The team keeps two meters, Security and Operations, and closes the shift in one of six endings depending on how it resolved each message. Reporting everything stalls the business (Operations drops); complying with everything opens the door (Security drops) and, in the most advanced cases, can leave an internal account compromised that keeps sending messages in the following waves — now clean across every technical control.
Every message the way any employee would: the sender's visible name and address, subject, body, the link's text (not its real destination), the attachment's name (not its real type), the remaining deadline, and whether the message allows verification.
The same as Inbox, plus one technical layer per message that gets revealed on inspection: authentication headers (SPF/DKIM/DMARC and the sender's real address), a link's real destination, a domain's age, and an attachment's real file type.
The same as Inbox, plus the full organization, always visible at no cost: the org chart, who's on vacation, vendors with their registered bank account and phone number, and company policy.
// Measurement and debrief
When it ends, everyone sees the same result at the same time. What gets recorded is what actually happened during the game: objectives, time, incidents and hints. None of it measures a person — it measures a team solving a concrete problem — and that's on purpose.
Seen by the players and the facilitator, with the same content.
A separate view, for guiding the final conversation.
A design decision, not a technical limitation.
Exporting the debrief to PDF for the People team is on the list of possible improvements, not implemented. Today, results and the debrief are viewed from the organization's panel.
// Who it's for and how to start
Built for People teams, team leaders and anyone who has to put together an activity that leaves something behind. People sign in with a link; whoever organizes it manages everything from a panel.
Isolated data and your own branding. Each organization is reached from an address like /your-company, with its own logo and colors.
Were you invited to a session? Use the link you received by email. Questions before deciding? Write to us at larp.latam@gmail.com.
Each organization lives in its own data space, with its own branding and an address like /your-company. Actions are kept in an audit log.
Add people by hand, import an Excel or CSV spreadsheet with areas and job roles, or load them through the employee REST API.
Email invitations with a magic link, scheduled reminders, and automatic notices before every session.
Admins, managers who set up and facilitate sessions, and participants who only see their own.
To sync the roster of people from the system the organization already uses, without loading anything twice.
Anyone in the organization can play solo with bots before the session, to show up already knowing how to play.
// Frequently asked questions
Missing one? Write to us at larp.latam@gmail.com.
Between 4 and 20 per session, depending on the game. Blind Operations and Silent Depth run from 4 to 10; Knowledge Files, Dependency Network and Crisis Room go up to 20, split across several teams playing the same mission in parallel.
A match runs between 20 and 90 minutes depending on the game and the settings. On top of that there are a few minutes of lobby and briefing at the start, and the debrief at the end, which is where most of the value shows up.
Yes, it's the case it was built for. Everything happens in real time in the browser, with team chat and a general chat, and voice if the organization has it enabled. It also works with the whole group in the same room.
A computer with a browser and the link they get by email. There's nothing to install and no password to create: the magic link leads straight to the session's lobby.
A Game Master, who can be someone from within the organization. They don't play: they set up the session, start it, follow each team's progress, pause, add minutes, send hints and lead the final debrief.
They get a grace period to come back, and on reconnecting they recover their role and the match's state. If they don't come back, the Game Master can hand that station to someone else without restarting anything.
Yes. Anyone can play solo with novice, competent or expert bot teammates, and the Game Master can fill a session's empty spots with bots. The bots are rules with memory, not generative artificial intelligence: they decide from the same filtered screen a person would see at that station.
No, and that's a design decision. The games are cooperative: the result belongs to the team. There are no individual scores, no comparisons between people, and no personality labels.
With observable facts from the match: objectives completed, time, incidents logged by the game, hints used, and the timeline of what happened. They're material for the debrief conversation, not for evaluating anyone.
Always on a computer. Five of the six games also work on a tablet in landscape; Dependency Network needs a computer. None of them work on a phone: the game screens need 1280 × 720 or more.