1️⃣ AI returns are becoming an operating model test
BCG argued on March 26 that most companies still miss meaningful returns from AI because they treat it as a tool rollout. The firms getting value are redesigning end-to-end processes, narrowing the use-case list and tracking P&L impact.
💡 Why it matters
This puts the management job back on sequencing, workflow design and financial discipline. If AI is not changing how work actually moves, it is still a pilot.
☕ Coffee talk
Which AI project in your company still has a demo but no owner for the P&L?
2️⃣ Strategy-culture fit is won or lost in the middle layer
INSEAD wrote on April 9 that companies usually attack culture from the top with new values or from the bottom with behaviour checklists. What gets missed is the middle layer: decision rights, KPIs, incentives, training and routines that teach people how work really gets done.
💡 Why it matters
If those levers point one way and the strategy deck points another, the deck loses. Leaders do not fix that with another values refresh. They fix it by rewiring the context managers run every day.
☕ Coffee talk
Which KPI or approval rule in your company is quietly teaching the opposite of the strategy slide?
3️⃣ Better strategy comes from learning across bets, not grading one bet at a time
Stanford GSB highlighted on March 25 the idea of correlated learning: one decision should teach you not just whether that move worked, but what it says about the next market, product or structure you should try.
💡 Why it matters
That shifts strategy from postmortems to search design. Teams need experiments that generate useful information across options, not just a pass-fail verdict on one initiative.
☕ Coffee talk
After your last failed bet, what did the team actually learn about the next two options?