Product Owner
Clarifies needs, value and priorities. Does not impose estimates or overload the team.
ARIAVEO FOLLOWS / SIMULATION
Live the project.
Understand his decisions.
A virtual team, a client, a budget and unexpected. A mission to learn how to deliver a useful site without forgetting the people who build it.
An individual game, several roles to play, a debriefing possible with a coach.
HORIZON WORKSHOP · YOUR TEAM
"What can we really end with the available capacity? »
EXPERIENCE
A concrete project. Separate responsibilities.
Horizon workshop wants to present its workshops and facilitate appointments requests. The learner clarifies the need, estimates the work, prepares a sprint, responds to the situations encountered and inspects the result. The completed features change the site preview.
Developers
Collectively estimate, organize their work and evaluate what they can take care of.
Scrum Master
Facilitates exchanges and resolution of disagreements. Do not decide in place of others.
Changing roles serves learning. This is not a multiplayer game: colleagues, the client and the sponsor are simulated characters.
FROM NEED TO RESULT
Each feature puts arbitration into practice.
Collective Poker Planning
Different estimates, arguments and clarifications about scope, dependencies and risks. The team votes again; disagreement may lead to a short investigation or splitting the work.
What this implies: no numbers imposed by PO, Scrum Master or lead. Points compare the size of needs; They don't become hours.
Capacity per person
Tasks show their time ranges, skills and proposed contributions. The schedule compares the load to the availability of each member.
What this implies: a developer's free time does not automatically compensate for a verification overload. An investigation also consumes capacity.
Daily and mutual assistance
The player chooses a priority, explains his or her connection to the objective and may agree on a task recovery or a pair with an available and competent person.
What this implies: help mobilizes two people; the usual work of the colleague can wait. The pair does not automatically double the speed.
A board that follows the work
The cards change from "To do" to "On" and then to "Checked" and "Completed". Dependencies, remaining hours, managers and blockages remain visible.
What this implies: a card is not delivered because it is moved or because its estimate was low. The work and controls must actually be completed in the model.
Variable situations and absences
The new parties retain a draw of situations and absences. A planned absence reduces the capacity of the schedule; an unexpected event is announced at the relevant Daily.
What this implies: Reloading does not redraw events or give extra hours to other people. Losing a skill can delay dependent work.
A new demand during the sprint
From the second Daily, Léa can offer a business need. The player examines the contributions and chooses an integration agreement, a scope exchange or a backlog deferral.
What this implies: integration creates real tasks. It shall be refused if the charge remains higher than available. A deferral keeps the work already done and dependencies are controlled.
Additional technical work
Developers can retain a technical check required for an ongoing item. The task appears on the board, with a proposed load and contribution.
What this implies: the checks await this task. The load increases, without making trade points; The team may have to review the scope with the PO.
A team budget per sprint
The pedagogical hypothesis is CHF 10,000 for a five-day sprint, or CHF 30,000 for the mission. Funding, commitments and spending are distinguished.
What this implies: the cost follows simulated time, not points or productive hours. An absence does not reduce the package. An extension request does not automatically grant any budget.
Burn-up, burn-down and reviews
The burn-up compares the points delivered to the sprint scope. The burn-down follows the remaining hours in front of an ideal landmark. The reviews bring delivery, value and financing closer together.
What this implies: adding work or re-evaluating a difficulty can cause the curve to rise. An ideal trajectory is not a promise. Future days remain empty.
Retrospective and historical
At each cycle, the player retains an improvement and can review his decisions. Results and reflections by role provide a concrete basis for debriefing.
What this implies: explaining a compromise counts in learning. The score provides a point of discussion; he does not replace the coach's analysis.
UNDERSTANDING THE CONSEQUENCES
A choice can help the client
and weaken the team.
The consequences are defined by the scenario. The example below compares several responses to the same customer return from an identical state.
A return changes priorities
The first tests show that the majority of visitors come from a telephone. The customer prefers a simple and legible route.
Discussion helps to arbitrate; the next schedule may reflect this return.
Team energy70/100
Customer confidence70/100
Quality78/100
Technical debt0/100
Example from the scenario. Click compares answers; This does not alter any part. Without JavaScript, the first answer and its effects remain legible.
What indicators measure — and what they do not.
Team energy
From 70/100. Excessive pressure and promises can reduce it; cooperation and some retrospective improvements can support it.
Effect verified: it enters into the "collaboration" axis of the balance sheet, with confidence. It does not directly alter the available hours or the speed of work. An energy of 50 does not mean "twice as much production".
Customer confidence
It starts from 65/100 and evolves according to the answers to the customer or sponsor, transparency and compromises proposed.
Effect verified: it shall contribute to the collaborative assessment. It does not automatically trigger a client departure or a reduction in funding.
Quality
It starts from 75/100. Shortcuts can degrade it; early checks and quality decisions can improve it.
Effect verified: it contributes to the stocktaking and the on-line readiness indicator. Verification tasks remain to be completed separately: a high gauge does not complete them.
Technical debt
She's starting from 0. Some compromises are increasing; improved verification may reduce it.
Effect verified: from 12 onwards, the model withdraws one hour of simulated day availability. Debt also reduces the balance sheet quality axis. It is a teaching rule, not a universal law of software development.
How to read the final score?
The score combines four axes: delivered value (30%), quality (25%), adaptation (20%) and collaboration (25%). Collaboration is the average of confidence and energy. The quality axis subtracts half of the technical debt, rounded. Adaptation is based on responses to situations recognized as adaptation practices in the scenario.
Free explanations, the relevance of arbitration and all management actions are not semantically assessed by this score. They remain in history for debriefing. This score is not a Scrum certification.
FUNCTIONAL VALIDATION / 2 NOVEMBER 2026
20 full missions.
Searchable results.
Parcours automatisés sur cinq politiques et quatre tirages partagés. Ce ne sont pas des essais utilisateurs. Each policy was executed on four identical runs of situations and absences, with the exception of the witness case without absence. Script choices follow a fixed rule: they are neither people nor learning agents.
20three sprints2 443assertions in this campaignCHF 30,000final cost of each route
| Automated strategy | Delivered Features | Energy /100 | Debt /100 | Applications accepted |
|---|---|---|---|---|
| Priorities and capacity | 5–8 | 94 | 0 | 0 |
| Broad scope, Dialogue | 7–10 | 94 | 0 | 0 |
| Wide scope, pressure | 7–10 | 36 | 24 | 0 |
| Negotiated requests | 7–9 | 94 | 0 | 3 |
| Dialogue, without absence | 11–13 | 94 | 0 | 0 |
The pressure is not just in the flow.
In these tests, runs with broad scope and pressure finish with energy at 36/100 and debt at 24/100, compared with 94 and 0 for comparable runs that favour dialogue. The number of delivered features is sometimes similar.
The debriefing must therefore look at several dimensions, rather than reward only the number of cards completed.
Reducing the scope does not guarantee the best score.
The "Priority and Capacity" script is deliberately prudent; he delivers less elements on some prints. The absence-free version has more capacity. New requests can also change the number and value of items in the backlog.
These tests verify the operation of the model. They do not prove a learning gain or the superiority of a strategy in a real project.
What did we check in addition to the 20 routes?
- Collective agreement at Poker, refusal of an imposed estimate, investigation and cutting.
- Individual capacity, absences, mutual assistance, dependencies and checks before delivery.
- Negotiated addition, overload refusal, deferral to next Poker and technical task blocking checks.
- Fixed budget, work conservation, graphics and history of sprints.
- Backup, reload, double submission, revision conflicts and access separation.
- Browse the browser, read-only coach and display at 390, 768 and 1,440 pixels.
Test games use isolated bases. No part of a learner has been completed or reset for this campaign.
FOR TRAINING BODIES
Prepare, observe, debrief.
Administrator
Enables the module for its establishment and assigns a mission to an existing participant or group. Sessions remain separate between institutions.
Learning
May open several parts of the same scenario in parallel, repeat each separately and consult its decisions. Progress and old results remain independent.
Coach
Consults the sessions allowed for reading alone. The justifications, the evolution of the indicators and the results make it possible to discuss concrete compromises.
Four questions for debriefing
- What did you protect when capacity decreased?
- What promise had a cost to the team or quality?
- What new request was worth replacing?
- What would you change by replaying the same situations?
An explicit framework for learning with discernment.
- Game time: 60–90 minutes is a pedagogical estimate, not a measure from user tests. Plan more for an animated debriefing.
- Written reactions: characters, votes and consequences follow the pattern. This is not a free dialogue with human colleagues.
- Separate effects: some responses change the indicators; explicit scope and addition agreements create or remove work on the board.
- Funding: the package is an exercise assumption. An extension can be prepared, but the mission remains limited to three sprints.
- Background: an old part keeps its script. Graph values that were not recorded are not invented.
- Scope: a mission ready to play, without script editor. The findings published here relate to the V6 engine tested on the date indicated.
People, business, budgets and fictitious data. Presentation and functional validation report · November 2, 2026.