Know more, do less

Project approvals & business decisions rely on three basic questions:

• When will it be done?

• How much will it cost?

• Is it worth doing?

When a project is unlike anything the team has done before, early estimates become overly optimistic. And, if a timeline is built on dependencies, we end up too far behind to make mid-course corrections.

The remedy is to measure the "risk" upfront through lean experimentation (and by "risk," I mean the unknowns that defy accurate estimation).

Identify those unknowns, slice them into thin experiments, dedicate resources, time-box the effort, and run them just long enough to make the work "knowable." Then we can scale estimates accordingly.