Start With a Useful Capability

When someone asks for a system, we want to understand the difficulty that made them ask. The request may be for an entire proposal platform; the immediate problem may be finding the requirements that matter in a long document.

Helping a reviewer find those requirements and check them against the source is a capability with a clear purpose. It gives us something to try with the person who knows the work. We can see what it misses, whether it saves effort, and whether it changes what they need next.

A craftswoman attaches a large metal handle to an enormous wooden drawer while a colleague tries opening it. — One useful handle Start with the change that lets someone use what is already there.

Make the first useful thing specific

A capability makes someone better able to do something. It might help them compare records, prepare for a conversation, examine a decision, or spot a mistake. It can live in tools they already use. It does not need to become a separate application to count.

Small scope helps us understand the result, but it does not make every experiment harmless. A draft for its author’s inspection and a decision sent directly to a customer have different consequences. The way we try the capability should fit those consequences without turning a simple experiment into a miniature procurement programme.

Let use change the plan

A useful first version may remain a small tool. It may reveal that a larger system is needed. It may show that the real problem is an unclear responsibility or poor source material, and that adding software would leave it untouched.

We are willing to follow any of those answers. Starting small is a way to learn what deserves investment. It is not a promise that enough small tools will eventually solve every problem.

Original illustration generated with AI for this belief.