Deciding at 60% Information

P64.tradeoff-narratives.02 · Audience: guest, it-ml, language-pro · Prerequisites: Saying No — and Being Thanked for It

Saying No — and Being Thanked for It The Scope Cut That Saved the Project

LLM grading (this page):Real gradingOff — no LLM calls

A newly hired platform lead is told her mandate is "scaling AI capabilities across Europe" — no team chart, no budget line, three countries with three different ideas of what that means. Her colleague waits four months for the picture to clarify; it never does, and the waiting is itself read as a decision. She instead spends one week writing three lists — what is already known, what is unknowable until tried, what can safely be decided later — and then commits to the smallest step that would teach her the most: a six-week pilot in one country. The same play runs at every scale — a teacher planning week one for a class they have never met, an interpreter rendering a German sentence whose verb has not arrived. Ambiguity is not a hole where a decision should be; it is the normal working condition of anyone senior enough to be handed problems with no owner's manual — and this module teaches how to operate in it, and how to tell the story afterwards.

🗺️ So far you have built: the professional no (module 01): the three-part refusal, ranking asks against one agreed criterion, and the follow-through that turns refusals into trust. That module assumed the options were at least visible — three named asks on the desk, one slot. This one removes even that comfort: the mandate is vague, the information is partial, and waiting for certainty quietly becomes the worst decision available.

In this module you will: how to decide when you can see only sixty percent of the picture. You will learn to map the fog — known, unknown, decidable-later — so ambiguity becomes a work plan instead of a mood; to sort decisions into reversible doors and one-way doors, which sets how much certainty each one deserves; to find the smallest committing step that buys the most information; and to communicate a provisional decision with authority instead of hedging. Then the part interviews probe: telling an ambiguity story so the structure you imposed is the star. The ladder closes it with a fog-map exercise on a mandate you are given, then your own deciding-at-60% story.

Step 1 / 6 — Map the fog: known, unknown, decidable-later

Ambiguity feels like a wall because it arrives as one undifferentiated cloud: "improve how the team works", "scale this across Europe", "make the course more engaging". The first move is never to answer it; it is to sort it into three lists. Known: what is already established fact — the budget, the deadline, the two complaints everyone repeats. Unknown: what nobody can tell you today and only action will reveal — whether the French team will adopt anything, whether the quiet students are bored or lost.Decidable-later: real decisions that do not need deciding yet, and the trigger that will make each one ripe. Twenty minutes of sorting converts a mood into a work plan — and the third list is the secret weapon, because most of what masquerades as urgent ambiguity is simply a question whose time has not come.

ⓘ Concept: The fog map: three lists that turn ambiguity into a work plan

fog → three lists: known (established fact) · unknown (answerable only by action) · decidable-later (real decision + the trigger that ripens it)

Why it matters — Unsorted ambiguity produces the two classic failure modes: paralysis (waiting for a clarity that is not coming) and thrash (deciding everything at once, mostly wrong). The map dissolves both. The known list is usually longer than it feels — writing it down shrinks the fog by half before you have decided anything. The unknown list converts anxiety into experiments, because each entry's real label is 'answerable only by trying something'. And the decidable-later list is explicit permission to not-decide — the difference between strategic patience and drift is precisely whether the deferral is written down with a trigger ('we choose the platform after the pilot reports') or just silently postponed. The map is also the first artefact worth showing a stakeholder: it proves the ambiguity is being governed, before a single result exists.

✅ Key takeaways

Sort ambiguity into three lists before answering it: known (established fact), unknown (answerable only by action), decidable-later (a real decision plus the trigger that ripens it). Twenty minutes of sorting turns a mood into a work plan.
Classify the doors: reversible decisions with small blast radius are taken fast at ~60%, one-way doors earn the slow path — and a one-way door can often be re-engineered into a reversible one (pilot before rollout, internal before public).
Move by the smallest committing step: a real deliverable with a date, cheap to be wrong, aimed at the top unknown — so the fog shrinks whichever way it lands. Steps that commit without teaching are gambles; steps that teach without committing are theatre.
Announce provisional decisions as decision + confidence + tripwire: state the call plainly, say where your confidence is and is not, and pre-agree what would reopen it. Fake certainty and hedge-drowning both cost you; a named tripwire preserves authority through revisions.
In the telling — including the interview's 'inherently ambiguous role' question — the structure you imposed is the star: mandate, fog map, doors, smallest step, tripwire, result. Told this way, even a bet that failed lands as machinery that worked.
Module 01 ranked asks you could see and this one governed the fog around them; both were trades made while the plan was being drawn. The trade made mid-flight — when the plan is already slipping and something must be cut before it cuts itself — is the final module's subject.

Next: P64.tradeoff-narratives.03 — The Scope Cut That Saved the Project

None of this requires a technical field. The fog map works on a vague headship brief, an unscoped translation project, and a Europe-wide AI mandate alike — wherever work is handed over without an owner's manual, which is the surest sign it was worth handing to a professional.

Ask the mentor about this module

Ask a question about this content. The mentor explains and grounds its answer in what you are studying; asking is recorded as a learning signal, not a grade.

Images, PDF or text. Kept on this device only.

Keeping your files on this device

Keep a copy of the files I attach, in this browser

Off by default. The mentor always gets your file; this only decides whether your own copy stays here. Copies live in this browser only - they do not follow you to another device, and clearing site data removes them.

Ctrl/Cmd + Enter to send

🎓 Practice ladder

2 graded rungs · ~20 min

Two mentor-graded craft rungs. Rung 1 gives you a genuinely vague mandate — "improve how the team works" — and asks for the full fog map plus the first reversible committing step with its tripwire. Rung 2 is your own deciding-at-60% story, told so the structure is the star; a low-stakes story is always an accepted substitute. The mentor grades the telling, never whether the bet paid off.

Rung 1 — Map the fog, take the first step (explorer)

Loading exercise…

Rung 2 — Your 60% decision, structure first (practitioner)

Loading exercise…

Try it yourself

A scratch console for this page's ideas — ungraded, nothing you run here is recorded.

Scratch console

A scratch console with the scientific stack (pandas, numpy, scikit-learn). Runs on the server — no network, resource-limited and measured.

Output appears here.

My notes on this module

Loading your notes...

Where next?

Continue

The Scope Cut That Saved the Project →

Trade-off Narratives · P64

Later in Trade-off Narratives

The Scope Cut That Saved the Project

Go up a level