Why AI pilots stall
The pattern is almost never the model. It is that the workflow was never redesigned, so the tool got added to the old shape of the work.
thinking
insights.
Research, frameworks and honest post-mortems on the work — what actually happens when organizations adopt AI, restructure operations, or try to make six systems behave like one.
First pieces publishing soon
where this stands
Nothing is published here yet.
We would rather show you an empty shelf than fill it with restated industry commentary. Below are the questions we keep running into in the work — the ones worth writing about properly. When something is ready it will appear here.
If one of these is a live question for you right now, ask us directly — the answer exists well before the essay does.
open questions
what we keep meeting.
The cost of undocumented process
What it actually costs when the operation lives in one person’s head — measured in onboarding time, decision latency and key-person risk.
AI governance for small teams
What a usable AI policy looks like for an organization with no legal department and no appetite for a forty-page document nobody reads.
Church technology after the platform
Most ministries have bought the tools already. The remaining problem is the seams between them, and who owns each one.
Creative as a capability, not a campaign
Why three good agencies can leave a team no better off than before, and what an operations layer changes about that.
Reading an operation you didn’t build
The diagnostic method: how to find the real constraint in an organization in the first two weeks, without relying on what people tell you it is.
next step
skip the essay.
If one of these is the question in front of you this quarter, the fastest path is a conversation, not waiting for us to publish it.