Practice · Real cases
Real cases: twenty people, twenty situations, one framework
The book proves the framework on a fintech campaign, a software migration, a bank proposal and an art festival. These twenty cases take it everywhere else — a coaching class, a factory floor, a clinic, a construction site, a kitchen table — with the same eight steps and not one of them changed.
Planned for release phase 2. The page is complete as a draft; interactive parts run in the browser and do not yet save to an account.
In brief
Each case follows one person with a real problem through all eight steps: what they wrote as the end state, what their own flow missed and what AI review added, how the Decision Filter split the work, which tool categories they used, the criteria they wrote before seeing anything, what failed those criteria, and what integration caught — usually one small fix and one structural problem traced back to its step. Half the protagonists are women and half men; most are students or early in their careers, the book's audience, with people in their forties and fifties where the situation calls for them. The people are composites, not clients; the situations are the ones the author sees every week.
What the twenty have in common
Nobody in these cases became good at AI. Each of them wrote down what they wanted before touching a tool, found that their own list of tasks had gaps a tool could see and judgements a tool could not, and wrote criteria before they saw output. Almost every case has a structural problem at integration, and almost every structural problem traces to Step 1 — an audience named and then not served, a texture described and then not tested. That is not a flaw in the people. It is what the framework is for: making the right work easier to find.
Read one that resembles your situation, then open the workbench and run your own.