About our methodology

Decision-making is the most important skill nobody is taught directly. We teach math, language, science, history, music, sport. We don’t teach the actual mechanics of choosing well when time is short, information is partial, and the stakes are real.

Children grow up learning to follow the stream. Then one day in their twenties or thirties, they have to make a choice nobody prepared them for, with consequences they cannot reverse. We wanted a place where they could practice the thinking before the consequences are real.

Most decision-skill products teach a checklist. "Here are the five steps to a good decision." Children memorize the checklist, pass a quiz, and forget. Nothing transfers to the moments that matter, because real decisions don’t announce themselves as quiz questions.

We do the opposite. Every mission looks completely different on the surface — a town through a storm, a delivery business through a tough month, a research ship across the open ocean. But underneath, the same structure is being practiced: choosing well when you cannot have everything, when time is short, and when you do not have all the facts.

Vary the surface story; preserve the deep structure. The child sees three different stories and one repeated skill. After enough repetition across enough surfaces, the skill stops belonging to any one story and becomes the child’s.

Harvard Business School built its reputation on the Case Method: rather than telling students the right answer, give them a hard situation and make them choose. Discuss. Defend. See what happens. Then a new situation, similar in structure but different in surface.

It works because real judgment is built through repeated honest attempts at hard choices — not through reading about them. The case is the medium because the case is the closest classroom analog to real decisions: ambiguous, time-pressured, with no clean answer key.

Heurister adapts the Case Method for children aged 10 to 18. We take the proven core of "decide, defend, see consequence, calibrate" and apply it through age-appropriate story-based missions a child can play in 15 minutes.

Right now we have three pilot missions, each built to be felt as a complete crisis a leader walks through — not a quiz with right and wrong answers.

Storm Response Mission: a town faces an approaching storm. You allocate medical readiness, infrastructure, and information across competing needs. Every round changes the situation; every decision compounds. There is no perfect path.

Drone Delivery Mission: a small logistics operation hits a hard month. Cash is tight, drivers are stretched, customers are unhappy. You decide what to protect and what to let go. The wrong answer is real; so is the right one.

Ocean Research Expedition: a vessel crossing open ocean encounters a sequence of unexpected conditions. Resources are finite. Crew morale matters. You navigate to the destination without compromising the mission.

Each mission has multiple rounds, multiple phases per round, and autonomous agents whose behavior reacts to your choices. No two plays unfold the same way — but the skill being practiced is the same in every one.

After each mission, we look at how the child thought — not whether they "won." We measure six layers of decision quality: how clearly they perceived the situation, what they were optimizing for, how their plan held up, how robust it was to surprises, whether their reasoning was internally consistent, and whether they kept options open or closed doors early.

These aren’t school grades. We don’t rank children, post leaderboards, or surface scores to the child. The measurement exists to make the parent-facing weekly narrative honest, and to help us tune which mission to recommend next. Children play. Parents see growth over time.

Heurister is in Early Access. We have three pilot missions live and a roadmap of more — including arc-less missions (open-ended scenarios with no built-in dramatic arc) and meta-skill missions that teach specific decision sub-skills (priority-setting, constraint-handling, judgment under partial information).

We are deliberately not racing to a hundred shallow missions. Each mission takes months of design, simulation tuning, and play-testing because the underlying mechanics — autonomous agents, world-state evolution, multi-round consequence chains — are computationally and pedagogically heavy. We would rather ship six excellent missions than sixty mediocre ones.

If you want a quick win, this isn’t it. If you want real growth, you’re in the right place.