TBM 348: Shared Understanding At Scale

You build a beautiful goal cascade five levels deep. North Star down to weekly tasks, everything connected. Looks great in your goal-tracking tool. But are you actually using all five levels? No. You're focusing on things you can control—the daily run, the water intake—while watching simple indicators like resting heart rate. Those middle layers? Just sitting there, stale and useless. They helped you think through the structure initially, but without dedicated rituals or attention, they're dead weight. This is cascade flattening: deep hierarchies satisfy the intellectual urge to categorize and the political urge to align with org levels, but they become a drag over time.

Information has to move through organizations somehow. Four patterns: the email blast where the CEO asks everyone "tell me what you're working on" creates immediate overload—high volume with no translation layer. The focused blast ("tell me specifically about Project Voltron") works because specificity wins even with messy format. The hierarchical telephone game compresses and reshapes every message through layers—by the time leadership gets the signal it's too late to act, and what gets passed up is shaped by what people think leaders want to hear, not what they need to hear. The flat weekly review (Amazon WBR-style) with 200 initiatives works because of consistent presentation, choreographed rituals, and the expectation that someone in the room knows the details. Scale requires shared interfaces and lightweight structure. The goal isn't to eliminate complexity, it's to make complexity legible.

Teams waste endless cycles searching for "the big picture to rule them all"—that magical dashboard with everything at once. But there's no universal view, only role-specific, moment-specific views good for a particular purpose. Views must be coupled to the task. Air traffic controllers have big-picture dashboards, but they're hyper-tuned for one specific, time-sensitive job. The dream of a single unified evergreen dashboard is seductive but misleading.

Everything's a loop—knowledge work is inherently loopy. The trap is assuming all loops are neatly nested and synchronized. Strategic opportunities linked to sub-opportunities linked to team initiatives, years to quarters to months to weeks. Looks great on slides but reality doesn't conform. Sometimes a big project is just a big project that doesn't need to fit into another loop.

All models compress complexity and leave things out. A product marketing tier system (Tier 1 major release, Tier 3 quiet changelog) works great for highly visible customer-facing work but misses developer community engagement, partner ecosystem effects, and high-leverage internal changes. Some models invite depth and flexibility (Team Topologies). Others flatten nuance and lock it out (MoSCoW prioritization). What makes them wrong and what they suppress matters.