We built a relay race to serve emergencies that move like wildfire.
The project cycle is the machine underneath everything we do. Assess the need, design the response, raise the money, deliver, then evaluate. Each stage hands off to the next, and the whole loop turns roughly once a year through the planning documents that govern it. On paper it is orderly and accountable. In a protracted or fast-moving crisis it is also the slowest thing in the room. By the time the assessment that opened the cycle becomes a funded program on the ground, the situation that assessment described has often already changed. We are not serving the crisis in front of us. We are serving a photograph of it, taken months ago.
This is not a failure of effort, and it is not anyone’s fault in particular. The cycle was designed for a world where a situation would hold still long enough to be assessed, planned for, and funded in sequence. That world existed for some crises and still does for a few. For most of the places we now work, it has quietly stopped being true. The design is shared across the sector, which means the redesign is ours to do together.
Why the loop runs slow
The lag is not in any single stage. It is in the handoffs between them. We run the cycle as a relay, where each runner waits for the baton before starting to move. The assessment must finish before the design can begin. The design must be complete before the appeal goes out. The money must arrive before delivery starts in earnest. The evaluation waits politely until the very end, after which its lessons are filed for a cycle that has already moved on. Each stage is reasonable on its own. Stacked in strict sequence, they add up to a response that is structurally behind the need it was built to meet.
Linearity carries a second cost we rarely name. A cycle that runs once a year teaches the whole system to think once a year. A change in the situation that arrives in month four has no stage ready to receive it, so it waits for the next assessment, the next plan, the next appeal. The signal is real. The structure simply has no slot for it until the loop comes back around. We end up rich in information and poor in the timing to use it.
We are not serving the crisis in front of us. We are serving a photograph of it, taken months ago.
What a faster cycle looks like
An anticipatory, continuously adapting cycle does not mean working without rigor or abandoning the discipline of assessment and evaluation. It means changing the shape of the loop from a relay into something closer to a current, where the stages run alongside each other and information moves through all of them at once. Three shifts make it concrete.
First, act before the assessment is complete, where the risk is known. For many of the shocks we respond to, the danger is forecastable well before it lands. A flood season, a harvest failure, a displacement pattern that repeats. We can agree in advance what we will do when a threshold is crossed, pre-position the resources, and release them on the trigger rather than on the appeal. This turns the front of the cycle from a starting gun into a state of readiness. The assessment then refines a response already in motion instead of being the gate that holds it back.
Second, fund the cycle, not the project. A large share of our slowness is inherited from how money is committed. When financing arrives tied to a single design for a single year, the plan cannot move without renegotiating the money, and renegotiating the money takes a cycle of its own. Where we and the parties who fund this work hold influence over those terms, we can build in pre-agreed triggers, standing contingency lines, and the authority to shift resources between activities as conditions change. The people who fund the work want what we want, which is money that reaches the need while the need is still current. Adaptive financing is not looseness. It is the operational form of planning for a world that moves.
Third, make evaluation continuous and put it at the front. An evaluation that arrives after the cycle closes is a lesson delivered to no one in time. If we instead read the signal from delivery weekly and let it revise the plan in real time, evaluation stops being the final stage and becomes the nervous system of the whole loop. Monitoring that can change a target this month is worth more than a perfect report next year. What we learn in delivery should be allowed to rewrite what we do in delivery.
None of this requires new technology or a longer planning document. It requires permission to run the stages together instead of in line, and to update the response on a rhythm set by the crisis rather than the calendar.
The test of a good cycle is not whether it completed every stage in the correct order. It is whether the help arrived while it still matched the need. That is a harder thing to measure and a more honest one. Building a cycle that can keep pace with the crises we actually serve is work we can begin now, in the next plan we write, and it is work worth doing together.