Field Note 001
Everyone loves to play in the sandbox
But eventually someone has to move some dirt.
A sandbox is a fine place to learn. The data is clean, the stakes are imaginary, and the demo looks like progress. Most of the organizations we meet are not short on demos. They are short on work that survived contact with the operation.
The gap is rarely the model. It is the handoff. Who owns the output on a Tuesday. Which system it has to land in. What “better” means in a number the business already tracks. Without those, a pilot is a presentation with a login.
Moving dirt is less glamorous than the sandbox, and it is the job. Pick a decision that already happens. Put the tool in the path of that decision. Measure the before and the after in the unit the operator cares about — errors, minutes, missed loads, rework.
Enterprise strength is the unglamorous part: access, definitions, security, an owner, a way to turn it off. Market speed is refusing to spend a year building a platform before a single workflow gets better. You can have both. You cannot have either if the work stays in the sandbox.