When everything is a priority, the queue decides.

The acceleration paradox

The pattern is recognizable. Leadership sets ambitious goals, every department gets approval to launch its initiative, and the portfolio grows from six programmes to fourteen. Nobody is working less. Delivery dates land later than before.

The explanations offered are usually about execution: optimistic estimates, loose scope, overstretched teams. None of that is wrong, but all of it describes symptoms. The cause is structural, and it states plainly.

What makes it hard to see is that nothing looks idle. Every initiative has an owner, a plan and a status that reads green. The waiting happens between the visible steps — a specification pending a legal answer, a release pending a security review, an integration pending a decision from a team already committed elsewhere. None of that waiting shows up in a status report, and all of it shows up in the delivery date.

A relationship nothing gets around

In queueing theory, the law John Little proved in 1961 establishes a stable relationship: the average number of items in a system equals the arrival rate multiplied by the average time each item spends there. Applied to a project portfolio, it reads as follows: average delivery time equals the number of initiatives in progress divided by the rate at which they finish.

The consequence is uncomfortable. At a constant completion rate, doubling the number of open initiatives doubles the average time each one takes. No management method, tracking tool or weekly status meeting changes that, because the relationship makes no assumption about how the work gets done.

The table assumes completion capacity holds steady. In practice it drops: every additional initiative consumes coordination, decisions and executive attention. The observed slowdown is usually worse than the arithmetic predicts.

A waiting initiative is not free. It occupies attention, meetings and decisions exactly like an active one.

Cap the work instead of adding capacity

The practical conclusion runs against instinct: to deliver faster, start fewer things. That is the logic behind work-in-process limits, which the DORA research programme lists among the capabilities associated with stronger delivery performance.

At portfolio level the rule is simple: set a maximum number of active initiatives, and start nothing new until something finishes — or is explicitly stopped. The difficulty is not technical, it is political. Someone has to tell an executive that their initiative will wait.

That conversation gets easier when the cap is explained as a delivery commitment rather than a refusal. An initiative placed in the queue is not cancelled: it has a position, a condition for starting, and a realistic date attached to both. Compared with the usual alternative — an immediate green light followed by eighteen months of quiet drift — most sponsors find the queue preferable once they see the arithmetic behind it.

1. Set an explicit cap on active initiatives, lower than feels reasonable.
2. Finish before starting: a new launch requires a closure, not a promise.
3. Make the waiting list visible, with the reason for waiting and the condition to resume.
4. Actually stop what has stopped moving, rather than leaving it paused indefinitely.

What this changes in governance

Prioritization as usually practised means ranking initiatives by importance. The exercise is comfortable because it refuses nothing: everything stays on the programme, in some order. It therefore produces no decision at all.

Prioritization that has an effect answers a different question: what waits, and until when? That is a visible decision, it will be contested, and it has to be explainable. In exchange it gives teams something a ranking never does: confidence that what has been started will be finished.

Organizations that adopt the discipline often notice a useful side effect. Once seats are limited, the conversation about an initiative's value becomes real, because it carries an identifiable opportunity cost. As long as everything can start, no discussion about priorities has any stake — and it is the queue, not the strategy, that sets the order.