Impart · The Router

Practitioner Cheat Sheet

Use live, in the room. Built from the Router Interview, 2026-08-17. Verified against three validation cases.

The rule this sheet exists to enforce Never route on what someone says they need. A person who could describe an entire funding value chain, could name exactly where she wanted to stand, and had taken zero steps toward it, reported that she needed to understand the landscape better. Taken at face value that routes to Terrain and costs her another quarter of reading. A sincere self-report is persuasive because it is sincere. Be unpersuaded.

The four questions

Run in sequence. Stop at the first one that routes.

1Terrain check

“Who are the biggest operators in the field you're working in, and why haven't they solved this?”

Cannot name them	Terrain scan. Nothing else is diagnosable yet.
Names them, cannot say why they underperform	Terrain scan. Naming is not reading.
Names them and can say why	Continue to Q2

“There are none” is never an opportunity. It is insufficient research. Ask about the domain where they are stuck, not the one they are fluent in. A fluent answer to the wrong Q1 is the easiest way to misroute someone.

2Serve or own

“Why haven't you just joined one of those organizations?”

Realizes there is no real reason	Serve Contribute to an existing effort. A successful outcome, not a lost sale.
Has a real reason	That reason is the founding thesis. Continue to Q3.

Skip for anyone not proposing to found something: consultants, operators, career movers.

3The honesty test

“Name the specific decision you can't make with what you already know.”

Cannot name one, or names something decidable today with a caveat	Not an information gap. Go to Q4.
Names a concrete, genuinely unmakeable decision, first time	Real gap. Route by type below.
Names the same gap they named last time	Gate A gap that never closes is not a gap.

Never ask “would more information help?” Almost nobody says no. Information-seeking is the most comfortable form of avoidance in this field: it feels like progress, produces artifacts, and defers the action indefinitely.

What is missing	Route
What am I even pursuing	Thing
Where value moves, what roles exist, where I would stand	Terrain
Who I am being, what must I believe, what will resist me	Footing
The order of operations	Build

4The action test

“What's the next action, and is anything actually blocking it?”

Available, unblocked, not taken	Gate Belief.
Blocked by something obtainable inside the window	Gate They could get it and are not.
Blocked by something not obtainable inside the window	Partner Real capability gap.
Being taken, no result yet	Stay Volume or time. Push through.

The availability rule. Not doing something blocked is weak evidence. Not doing something free, open, and requiring nobody’s permission is strong evidence. Every school lets you serve.
The window rule. Belief drives capability acquisition, not capability availability. Time, licensure, capital gated by someone else, a four-year curve against a six-month window: those are real. Everything else is usually belief.

The three traps

1. The sincere information request

Tell Can describe the value chain in detail, can name their position, has not moved toward it, and reports needing to understand the landscape better.

Reality Belief gap. Terrain would teach them what they already know.

Guard Q3. Make them name the unmakeable decision.

2. “I don’t have the resources”

Tell The stall is described as a resource shortage.

Reality Usually a misattribution. Not having resources is a fact about the world. Not taking action toward procurement is a belief about self or process.

Guard Ask what the last procurement action was, and when.

3. Reads the chain, not creating value, cannot name a position

Tell Looks identical to a belief gap.

Reality A position problem. They are standing in the wrong place in a chain they read correctly. The person who loves tattoos and is failing as an artist should be doing marketing.

Guard The question is not “do you understand the chain.” It is “can you name where you would stand in it, and are you moving there?”

Indicator sets

⚠ Drafted, not extracted. These were inferred from the interview rather than taken from Julian. They are the last unwritten piece of the router and the difference between a router that reasons and one that detects. Correct them.

Thing

Impact	Can name specifics they have ruled out, and the ruled-out list is growing. Can name a role in a value chain, not just a field.
Pivot	Still describing the interest in adjectives after three weeks. The “thing” changes between conversations. Cannot name one thing ruled out.
Cadence	Weekly. Short phase, fast drift.

Terrain

Impact	Can name incumbents and why they underperform. Can sketch the value chain. Can name three or more roles besides the obvious one.
Pivot	Research volume rising while position naming stays flat. Describes the chain but not where they would stand. This is the Trap 1 signature.
Cadence	Weekly while active, then on material change only.

Footing

Impact	Answers “who am I being” in present tense, without aspiration. Can name what will resist them. Has a rehearsed response, written not intended.
Pivot	Answers “who am I being” with a future-tense aspiration. Success language entirely external.
Cadence	On transition, never on a calendar. Ceiling 90 days. Revisited too often, Footing becomes rumination.

Build

Impact	Deliverables shipping. Referrals appearing without spend, meaning Step 7 is working. Review producing actual changes.
Pivot	Refining price and process repeatedly with no change in result. Success measured entirely by external proxies and nothing internal. Same review outcome three cycles running.
Cadence	Quarterly if highly engaged, twice yearly in the middle, annually as the floor.

The annual review is an anti-abandonment device. Silent drift is the normal end state for a build, and it is silent because nothing is scheduled to notice. A standing review makes abandonment a decision made at a date.

Build’s escape hatch

Build’s review can only refine downward: strategy, deliverables, process, price. Nothing inside the loop can conclude the idea is wrong, which is the sunk-cost trap. Add one question to every review.

“Has anything in the terrain, or in who I am being, changed enough that the idea itself no longer holds?”

No	Refine inside Build
Terrain moved	Terrain positioning depth
I have changed	Footing light revisit
Neither, still not working, and I can name where I would stand	Gate

Without an upward exit, a cycle is a trap with a calendar.

When ambiguous: time-box it

For a bounded, learnable gap achievable inside the window, assign it with a short return date and let the box be the test.

Closes it	It was information. Continue.
Does not close it, with a deadline live	It was never information. Gate

Costs two weeks instead of a quarter, and resolves the case empirically instead of by guess.

How each framework behaves

Gate

Belief System Audit

Baseline once for everyone, including in success. Then on stall.

Phase

Thing

Once, when any Ikigai circle is unresolved.

Utility

Terrain

Repeatedly, at three depths: scan, value chain, positioning.

Stance

Footing

Once as a phase, then persists as posture.

Cycle

Build

Continuously, on cadence. Exits only by abandonment.

Diagnose Thing with Ikigai. Treat it with elimination. People hold clearer views about what they hate. Using the four circles as the exercise is the common failure and produces someone staring at a diagram they cannot fill in.

Two elements are enough for motion. Four are needed for durability. Community and task alone can produce results; nobody has to wait until all four are solid to act.

Verified against

	Routed via	Result
Marcus	Q4: school visit, unblocked, not taken	Gate
Dana	Q3 names something decidable today → Q4 pitch costs an email, not taken	Gate
Ray	Q3: “fiscal sponsor or own 501(c)(3)”, concrete and genuinely unmakeable	Terrain time-boxed

Dana is the proof. Her self-report says Terrain. Q3 and Q4 both override it and land on the gate. That is this sheet doing the job it exists for.

Impart · The Router · v1.0, generated 2026-08-17. Source: impart-skills/impart-router/references/cheatsheet.md. Full extraction in Impart-Router-Interview.md, architecture in Impart-Framework-Router.md, claim limits in Impart-Evidence-Base.md. None of the three ships in the download; this sheet is a snapshot of them.