Innovation Lab

• Diagnose when an organization's current strategy has stopped working and explain why.
• Apply core frameworks (Christensen disruption, Schein culture, Kotter change) to specific decisions, not just essays.
• Make tradeoffs across innovation, stakeholder trust, and financial discipline under real time pressure.
• Lead through ambiguity — defending decisions when outcomes are messy or politically loaded.
• Reflect on team dynamics: how decisions emerged, who got heard, what would change next time.

1. 1
   Year briefing

   The year opens with a recap of last year's outcomes and a fresh set of decisions to make.
2. 2
   Decisions

   The team answers the year's decisions: 2 to 6 a year, depending on the company. Most are multiple choice; Year 2's budget split is the innovation portfolio, its slider sets the transformational bet's scope, and Year 4 makes a scale, pivot or kill call on each bet. The SaaS track adds written memos, and every track writes a short portfolio rationale in Years 2 and 4; these earn no points themselves (the professor and the Innovation Committee read them). Point values stay hidden until the year is scored.
3. 3
   Events

   The events the professor schedules land, usually at the decisions deadline (a new competitor, regulation, a key hire leaving), which also opens the Adapt window; nothing is scored yet. Every event comes with two or three responses written for the company, and any teammate can choose the team's response, with an optional why. An event's harm always lands. Choose one of its responses before the year is scored: each relieves a different share of the harm and captures a different share of the event's upside, on different pillars, and some cost something. With no response the harm lands in full and the upside is lost; Results shows what each response would have done.
4. 4
   Adapt

   With the events in view, the team may change up to 2 answers. There is no bonus for changing: the final answer is the one scored. The team can also keep an answer with a one-line reason: a reason of at least 15 words that names one of the year's events earns +1.5 Adaptability, for up to 2 answers a year. The window closes at the professor's Adapt deadline (24 hours after the decisions deadline if none is set), and the year is scored then.
5. 5
   Results

   The engine scores the year's answers and events; gains shrink once a pillar is above 50, so balanced play pays more than piling onto one pillar. Pillar scores and business results (revenue against a ghost that funded nothing, the vitality index, adoption) move, the leaderboard updates, and Results lists what moved each number.
6. 6
   Critic feedback

   The Innovation Committee, an AI review in the voices of the company's CFO, its innovation sponsor and an independent director, writes 2-4 paragraphs about the scored year, remembers last year's advice, and scores the team's written work 0-10. That score moves next year's Stakeholder Trust by at most 3 points.
7. 7
   Reflection

   Each student writes a reflection before and after the decisions. Each one of at least 40 words, submitted on time, counts toward the student's personal score (a late one counts half).

One team number everywhere: the Innovation Score of the latest scored year, 60% the pillar composite (by default 30% Innovation, 25% Adaptability, 25% Stakeholder Trust, 20% Financial Health) and 40% Business Results (revenue against the ghost plus the vitality index; 50 = no better than funding nothing). Pillars start at 50 and carry forward, so that number reflects every year played. The leaderboard and the team part of each grade use it.

• Innovation portfolio

  Three authored bets per company (core, adjacent, transformational) with hidden odds drawn once per class, delayed payoffs, a disruptor clock, pilots that buy an honest read, a scale / pivot / kill call in Year 4, and an adoption meter: a bet that works pays only as far as people adopt it. The model is built so that, on average, a mostly-core mix with real adjacent and transformational slices beats all-core and all-moonshot, and testing then deciding beats deciding blind.

• Leaders who remember, and one shared challenger

  Four named leaders per company (the innovation sponsor, the CFO, the line leaders and the head of people) remember how the team handled the Year 1 and Year 3 buy-in calls, each with a one-line reason, and their trust drives half of the adoption meter, so winning people over pays off in adoption years later. One challenger per class grows faster when the class as a whole under-invests in new bets (up to 1.6 times its usual pace if every team stays on the core business), so the leaderboard shows the cost of collective caution.

• LLM Critic

  After each scored year, an AI Innovation Committee reviews the team's portfolio, pilot evidence, calls, event responses and written reasoning, remembers last year's advice, and writes 2-4 paragraphs. It scores the written work 0-10, which moves next year's Stakeholder Trust by up to 3 points. One AI call per team per year.

• Industry tracks

  Same simulation engine, five companies, each with its own briefings, decisions and events: SaaS (NovaTech), Healthcare (Meridian), Financial Services (Crestline), Retail (Harbor & Main), Manufacturing (Apex Industrial).

• Personal score and team roles

  Each student's grade (before the class curve) is 70% team score and 30% personal score. The personal score, averaged over the scored years, is half the student's own reflections (at least 40 words, on time; a late one counts half) and half their own share of the team's submissions and event responses. Optional light team roles (CEO, Chief Innovation Officer, Finance Lead, People Lead) each own the decision questions of set years; the post-decision reflection, written from the role, is the role reflection. Once any team claims a role, the 30% is split: 20% role work (half answering the role's questions, half those reflections) and 10% personal score.