6 July 2026 · Insight

Why Your Best People Should Do One Thing at a Time

Your most capable specialist is assigned to the three most important projects in the portfolio.

On paper, this is what good prioritisation looks like. The work that matters most has your best person on it. Nobody is idle, and no priority project is left without the skill it needs.

Here is the problem.

All three of those projects will finish later than they would have if the same person had taken them one at a time. And because the value of finished work is time-sensitive, all three will also be worth less when they finally arrive.

This is the cost of multitasking across projects, and almost no organisation puts a number on it. This piece is about what it actually costs, and why the fix is the opposite of what most managers reach for.

Why splitting a scarce resource feels like progress

When you spread your specialist across three projects, all three light up as active. Each one can report movement. Each status line goes green. From the outside it looks as though the most important work is advancing on every front at once.

Movement is not the same as flow.

A project delivers its value when it is finished, not while it is in motion. Three projects that are each partly done have delivered nothing. The activity is real, the effort is real, and the value realised so far is zero. What looks like progress on three fronts is often three unfinished things and a person paying a tax you cannot see on a report.

What multitasking across projects actually costs

There are two separate costs here, and both are invisible on a status update.

The first is the switching cost. Every time your specialist moves from one project to another, she pays a setup cost that has nothing to do with the work itself. She has to put down one problem, pick up another, remember where it was, and rebuild the context she had before she can produce anything. The constraint - the specific person, skill or asset that actually limits how much the organisation can deliver - spends its scarcest hours reorienting rather than producing.

Eliyahu Goldratt named this bad multitasking in his work on Critical Chain. The insight was simple and uncomfortable: the switching itself consumes real capacity, and the more things a person holds open at once, the more of their time disappears into moving between them. It is one of the most reliable ways to make everything slower at the same time.

The second cost is that unfinished work holds value hostage. Work in progress - anything that has been started but not yet completed - sits on the books producing nothing until it crosses the line. Three projects at seventy percent complete are not seventy percent valuable. They are worth what finished projects are worth, which is to say nothing yet, and they are all waiting behind the same person.

This is where drag cost comes in. Drag cost is the money value of the time a delay adds to a project. If finishing a month later costs a hundred thousand pounds in delayed revenue or a missed window, then a month of delay is worth a hundred thousand pounds. While your specialist switches between the three, each project accumulates drag cost as it waits for attention that is being spent on re-immersion rather than output. You are not running three projects in parallel. You are running three meters, all of them charging.

This is a property of the system, not a lack of effort

The instinct at this point is to treat the delay as a discipline problem. If the specialist would concentrate, or work a longer week, the switching cost would go away.

It would not. The switching tax is not a sign that she is disorganised. It is a property of what happens when a constrained resource is shared across too much concurrent work, and it is the same mechanism that produces the coordination ceiling at portfolio level, where adding more concurrent projects starts to reduce total throughput rather than increase it. Working harder inside an overloaded system does not remove the overload. It just runs the same losing arithmetic faster.

What managers usually do, and why it makes it worse

When the three projects start to slip, the common response is to spread the specialist wider still.

A fourth priority appears, and she is made available to that one too. Coordination meetings are added so that each project can check on her progress, which takes more of the very hours that are already scarce. Pressure goes up across all four. Sometimes the answer is to ask her to context-switch faster, as though the problem were the speed of the handover rather than the number of handovers.

None of this removes the queue for her time. It lengthens it. Every extra open project adds another setup cost, another meeting, another claim on the one person who cannot be substituted. The system is being asked to recover from a cost it created by design, and the recovery method is more of the thing that caused it.

The better question to ask

The fix is not a tool and it is not a productivity technique. It is a change to the question.

Stop asking how to get your best person onto all of the priorities. Start asking which one she should finish first.

That question forces a sequence, and a sequence is what a constrained resource needs. To choose the order, rank the work by value per constrained resource hour: the economic value a piece of work returns for each hour of the scarce resource it consumes. The work that returns the most per scarce hour goes first. Give the specialist that one thing, protect it from interruption until it is done, release its value, and only then move her to the next.

This is the discipline of flow. Fewer things open at once, more things actually finished. It feels slower, because at any given moment you are visibly working on less. It is faster, because the value arrives sooner and the switching tax stops eating your scarcest capacity. The result is counter-intuitive only until you have watched it happen: limit what your best people work on at once, and they finish more, not less.

A portfolio that keeps its best people on everything is not backing its priorities. It is diluting them. The projects that matter most deserve undivided access to the resource that makes them possible, one at a time, in the order that returns the most value soonest.

---