How to play

Objective

You control the backlog and running experiments for a small development team. Each day you make decisions that impact product development and overall sales.

Your current progress is displayed at the top of the screen. Once you are done for the day, click proceed to next day to move the simulation forward.

After {{settings.game_duration}} days, the game is over and your final score is computed based on total sales.

Your objective is to maximise total sales.

Backlog

Your teams maintains a simplified backlog. Different tasks can be prioritised or deleted entirely. If nothing is prioritised, your team will pick up a random task when needed.

Each task has visible and invisible properties, depending on the feedback level.

Topics

Each task is related to one of 5 different topics: {{backlog_topics[0]}}, {{backlog_topics[1]}}, {{backlog_topics[2]}}, {{backlog_topics[3]}}, or {{backlog_topics[4]}}. Different topics share different underlying probabilities of success and effect sizes.

This means that all tasks of topic {{backlog_topics[0]}} will tend to show similar results. These settings are randomised each playthrough, so that {{backlog_topics[2]}} might be a promising topic in one playthrough, but disastrous in another.

Once you have learned enough about a topic through experimentation, you can stop new tickets from being generated for it. Use the topic checkboxes at the top of the backlog to control which topics new tickets are added for. Unchecking a topic will not affect tasks already in the backlog, but no new tasks for that topic will be added. This is useful when you have determined that a topic is not worth investing in, and want to focus your team's efforts elsewhere.

Exploration through experimentation in order to figure out which topics to exploit is the core challenge in this game.

Effort

Each task requires a certain amount of effort to implement. There are 3 levels of complexity: {{backlog_complexities[0]}}, {{backlog_complexities[1]}}, and {{backlog_complexities[2]}}.

Increased complexity also results in higher risk or reward. Experiments that result from tasks of {{backlog_complexities[2]}} complexity tend to have larger effect sizes, in either direction, than those of {{backlog_complexities[0]}} complexity.

Experiments

Once a task is complete, a new experiment will be started. Each day, new users will be exposed to the experiment and statistics will be computed.

At any given moment, you can decide to either Stop or Ship an experiment. Stopped experiments are removed from the game. Shipping an experiment will cause all future users to be exposed to this change.

Occasionally, an experiment may contain a bug that causes a very negative impact on sales, regardless of topic. These bugs are rare, but can significantly hurt your score if shipped. Note that leaving a buggy experiment running too long also hurts your score, as users in the experiment are exposed to the negative effect every day it runs.

You can review past experimental results and decisions in the history tab.

This is a game of chance, but not of chance alone. The faster you learn, the more you will earn. Good luck!

You shipped {{message.exp.name}}.

The real effect of this experiment was {{((message.exp.effect[1]-1)*100).toFixed(2)}}%.

This experiment contained a bug.

You stopped {{message.exp.name}}.

The real effect of this experiment was {{((message.exp.effect[1]-1)*100).toFixed(2)}}%.

This experiment contained a bug.

The game is over

{{feedback_levels[settings.feedback_level]}} {{settings.game_duration}} days

Your final score is {{format_percent(overall_performance.get_relative_effect(1)[0])}}.

This is significantly better worse than would have been expected if there had been no product changes made at all. This is not significantly different from what would have been expected if there had been no product changes made at all.