When capable people keep missing,
the problem is usually the demand,
not the people.
A team that used to deliver starts slipping. Headcount did not drop. The work did not look harder on paper. But decisions take longer, the same three questions keep routing through one person, and good performers start leaving for reasons that sound vague in the exit interview.
The easy read is that the team lost a step. Most of the time, it didn't. What dropped was not their skill. It was their access to it. Work demand started consuming more capacity than the system protected, and that gap compounded quietly until it landed in the numbers.
That is design debt. It is what the five capacity taxes are actually measuring.
The Core Model
Demand doesn't go straight to the numbers. It goes through state first.
Capacity is the cognitive, emotional, and physiological resources a person actually has on hand right now. It moves through the day. When demand climbs, it doesn't subtract skill from anyone. It drops their access to the skill they already have, and that access runs through four observable states.
Demand
Time, decisions, meetings, emotional load, change, routing, no recovery
Capacity state
● ● ● ●
Green → Yellow → Red → Can't-Even
Cost
The five capacity taxes the organization pays
Four things hold that state up: control over how the work gets sequenced, the resources to do it, recovery between pushes, and someone with your back when it spikes. Let demand keep climbing while those four stay flat and state slides down the chain. The skill is all still there. The access isn't.
Demand is not the enemy. Intolerable demand is. The trouble starts when the work needs more capacity than the system gives back, and people spend their days a zone or two below where the work assumes they are.
The Four States
Green, Yellow, Red, Can't-Even.
These are the states access runs through. They are observable, so a manager can read them without a survey. And most of a workforce lives in Yellow.
Green — full access. Strategy, hard problems, the high-stakes call. Rarer than the calendar assumes.
Yellow — compensating, and good at it. Quality slips slowly enough that nobody flags it. Where most of the workforce sits, and where the cost hides.
Red — overloaded. Where the expensive mistakes get made. "Just push through" makes it worse.
Can't-Even — shutdown. Even simple tasks stall. It reads as checked-out. It is biological depletion, not attitude.
What Creates the Load
Demand comes in seven kinds. Most teams carry several and name none.
"The team is overloaded" is not something you can act on. Which load, coming from where, is. These are the seven that show up most.
Time pressure
Volume and deadlines outrunning the hours, plus the catch-up that bleeds into nights and weekends until that becomes the baseline.
Decision complexity
Hard calls and ambiguity, made worse by the context switching that spends attention before the decision is even on the table.
Meeting and handoff load
Coordination overhead landing in the morning hours, when capacity is highest and deep work should be happening.
Emotional load
Difficult customers, conflict, and holding steady when things are tense. A work demand you can measure, not a wellness concern.
Change load
Reorgs, reprioritization, and tool churn that resets the work faster than anyone can settle into it.
Unclear routing
Work nobody clearly owns, bouncing between people and coming back as rework.
Recovery limits
Demand running flat-out with no reset built into the week, so there is never a point where the tank refills.
Cause Into Cost
Demand is what you can change. The tax is the bill if you don't.
Quick warning before the map: the five taxes are not five separate buckets you total up. They are five lenses on the same drag, and the demand types feed them messily — one demand can show up in three taxes at once. What is worth noticing is that each tax tends to surface once a team is parked in a particular state. That is the trajectory the diagnostic follows: which demands are concentrating, which state they have pushed people into, and where the cost lands.
Meeting Tax (coordination cost)
Driven by meeting and handoff load, time pressure. The calendar fills with coordination and the work that needed real focus gets pushed to 7pm, or doesn't happen.
Surfaces in ● Yellow — the slow slide as the best hours get eaten
Decision Density Tax (quality cost)
Driven by decision complexity, unclear routing, recovery limits. The reversed call and the missed flaw aren't people getting worse. They're people deciding on empty.
Surfaces in ● Red — where the expensive mistakes get made
Manager Load Tax (delay cost)
Driven by unclear routing, change load, everything funneling through one person. Approvals sit in an inbox during back-to-backs. The team stalls behind them, and after a while stops bringing new ideas because nothing moves anyway.
Surfaces in ● Yellow ● Red — one person's drop becomes the whole team's ceiling
Recovery Debt Tax (attrition cost)
Driven by time pressure, emotional load, recovery limits. It compounds in stages. By the time the resignation lands, the people who weren't watching the earlier stages are the ones who get surprised.
Surfaces in ● Red ● Can't-Even — depletion that no longer resets on its own
Forfeited Upside Tax (missed signals, connections, and innovation)
Driven by all of it, sustained, once there's no slack left to absorb anything. The visible work still ships. The customer signal nobody had the bandwidth to chase just quietly goes unchased.
Surfaces in ● chronic Yellow — the state nobody flags, because the work still gets done
Slack Is Not Waste
Zero slack looks efficient right up until it isn't.
Run a team at full utilization and the staffing model looks lean and the org chart looks disciplined. Then someone is out for a week, or a deal comes in hot, and there is no give anywhere to absorb it.
The work doesn't stop. It comes out slower and rougher, and nobody can point to the line item that explains why. Some slack isn't waste. It's the buffer that keeps judgment sharp, decisions fast, and someone free to notice the thing worth noticing. The research that tracked stable scheduling and adequate staffing found the people and the business improving together, which a pure efficiency lens says shouldn't be possible.
It's also why the fix runs in a particular order. You rebuild the buffer first — recovery back into the week — before you touch the load. Cut work off a team that has no slack and the freed hours just get swallowed by the next demand. Nothing moves. Restore the margin, then redesign what flows through it. Recovery first, load second.
The guardrail. Lean, kaizen, and the rest aren't the problem. Pointing them at "get more out of the same people" is. Use them to take demand off the system, not to claw back the buffer and book it as productivity.
Upstream Demand Debt
The team didn't lose discipline. The demand changed around them.
Most of the demand a team carries was not created by the team. Marketing pulls a launch forward two weeks. A board-deck request lands Friday at five. A reorg reshuffles priorities mid-quarter. Approvals back up two levels above the people waiting on them. None of it was the team's call. All of it ends up on the team's plate.
So the frontline gets the post-mortem for instability that entered through the planning calendar and the approval chain. The analyst is told to tighten up. The account lead is told to communicate better. Meanwhile the demand keeps arriving from upstream and lands downstream as Manager Load Tax and Recovery Debt, pushing teams into chronic Yellow and Red faster than recovery brings them back. No amount of effort at the bottom closes a gap that keeps reopening at the top.
The read for whoever holds the lever: before you ask a team for more, find out what demand the design is adding and where it starts. The fix is almost never where the cost shows up.
Load Transfer
The reliable move is usually not "add people."
It's taking the low-value admin and coordination off your highest-value people so their hours go to the work only they can do. You hired the senior engineer to solve hard problems, then handed her ticket triage. The same swap shows up everywhere once you look for it.
Manager → ops coordinator
Route the organizational noise somewhere else so decisions stop queuing behind one person. Pulls down Manager Load Tax.
Closer → deal desk
Take pricing, paperwork, and approvals off the person who should be selling. Protects the capacity that actually books revenue.
Engineer → clean intake
Fix how work arrives so the build doesn't open with triage. Cuts context switching and Decision Density Tax.
Executive → decision pre-brief
Stage the options before the room so judgment goes to deciding, not to rebuilding context from scratch. Cuts Decision Density Tax.
Same pattern every time: the most expensive thinking in the building is getting spent on work a form, a coordinator, or a pre-brief could have absorbed. Stop spending it there.
Where It Connects
A model is only worth anything when it meets your actual mess.
This page is the demand side of the picture. Underneath it sits the measurement layer: The Zones Framework, which is how you read what state a team is actually in. The diagnostic is where both get pointed at your organization — a half-day session that maps your patterns to the taxes, finds the demand concentrating underneath, reads the state it has pushed people into, and builds a cost floor from your own numbers in the room.
And it is a chain you can move, not just name. Redesign the demand and the first thing that shifts is the state distribution: the share of the week a team spends in Green instead of Yellow or Red. The pilot measures that shift against a baseline. That is what turns the model into evidence your sponsor can hold, on your team, with your numbers.
Find the demand before it becomes debt.
Tell us the execution pattern you have been watching and the size of the leadership team. We will tell you whether the diagnostic is the right starting point and what a realistic output looks like.