The Big Three Behind Late Deliveries

When something ships late, the first explanation is usually the least useful one. It was harder than we thought. The scope grew. We hit a snag. All true, and all beside the point, because those are descriptions of the delay, not causes of it. Dig underneath almost any late delivery and you find one of three things. Process, capacity, or incentives. Naming which one is the whole job, because the fix for each is different, and using the wrong fix makes it worse.

Process

A process problem means the work is waiting on the way the work moves, not on the work itself. The classic sign is that people are busy but the thing is not progressing. Steps are blocked on other steps. Approvals sit. The one person who has to touch three stages is a bottleneck that no one drew on the plan. Handoffs pile up and each one adds delay that has nothing to do with difficulty.

You fix a process problem by removing steps and seams, not by adding effort. If you respond to a process problem by pushing people to work harder, you get tired people and the same delay, because their effort was never the constraint. The constraint was the shape of the pipeline. Redraw the pipeline. Take out the handoff. Give one person the whole arc. The delay was structural, so the fix has to be structural.

Capacity

A capacity problem is the honest one. There is simply more work than the team can do in the time, and no amount of process cleverness changes that arithmetic. The sign of a capacity problem is that everyone is already working on the right things, in the right order, with no obvious waste, and it still does not fit.

Capacity problems have only three real fixes. Add capacity, cut scope, or move the date. That is the entire menu. The failure mode is refusing to pick one and instead hoping. Hope is not a fix. If it is genuinely capacity, then pretending it is a motivation problem and asking for more hustle is a way to burn your best people while the date slips anyway. Say the true thing. We have more than we can carry. Here is what we are cutting, or here is what we are adding, or here is the new date.

Incentives

The third one is the quietest and the most dangerous, because it hides. An incentive problem means the people doing the work are, rationally, optimizing for something other than shipping this on time. Maybe the reward is for looking busy, not for finishing. Maybe finishing early just gets you handed the next fire. Maybe the person who raises the risk early gets blamed, so no one raises it until it is too late to fix cheaply.

You cannot fix an incentive problem with a better schedule or more people, because the schedule and the people are not the issue. The issue is that the system is paying for the wrong behavior. The fix is to change what gets rewarded. Make it safe and normal to surface a slip early. Reward the person who flagged the crack while there was still time, instead of the person who quietly hoped it would hold. Until the incentives point at shipping, everything else you try is friction against the actual gradient.

Why naming it is the work

The reason to force the diagnosis is that the three failures look identical from a distance. They all show up as a red date on a status update. But the fix for a process problem makes a capacity problem worse, and the fix for a capacity problem does nothing for an incentive problem, and treating an incentive problem as a process problem just adds ceremony to a game people are already playing.

So when something is late, I do not ask how do we go faster. I ask which of the three is this. Get that right and the fix is usually obvious. Get it wrong and you will spend a lot of energy solving a problem you do not have, while the one you do have keeps shipping things late.