Reference

Templates: Strategic Choice

Part of the Strategic Choice skill · loaded on demand from SKILL.md

Copy-ready structures for each stage. Placeholders in angle brackets are allowed here and nowhere in a client deliverable.

Contents

  1. Framed-choice statement
  2. Option story card
  3. Conditions sheet (WWHTBT)
  4. Barrier and test plan
  5. Test results log
  6. Cascade choice record
  7. Activity-system map and fit tests
  8. Trade-off register
  9. Signpost register
  10. Rejected-options table
  11. Partner review checklist

1. Framed-choice statement

Decision: <one question with at least two mutually exclusive answers>
Decision-maker: <name, role>            Decision date: <date>
Options on the table: <A: status quo stated as a strategy>, <B>, <C>, <D>
What choosing one forecloses: <one sentence per option>
Constraints not open for relaxation: <balance sheet, ownership, regulation, horizon>
Trigger for the decision: <what happened that made the question urgent>

2. Option story card

One per option. Three to five options. The status quo always gets a card.

Option <letter>: <name>
Positioning base: <variety | needs | access | status quo | none (operational effectiveness only)>
Advocate: <name>                        Sceptic: <name>

Happy story (three sentences, written by the advocate, not yet evaluated):
Where it plays: <customers, geographies, categories, channels>
How it wins: <why the chosen customer picks us over the best alternative>
Why it works: <the activity or position the alternative cannot match>

We would stop doing: <specific current activities and the revenue or attention they consume today>
Trajectory if all conditions hold: <revenue and margin in year 3, one line>

3. Conditions sheet (WWHTBT)

One per option. Six to twelve conditions. Each is a declarative statement that could be false. The sceptic's objection must appear, in their words.

IDHeadingCondition (affirmative, falsifiable)Confidence by executive (H/M/L each)Lowest-confidence holder
<X>1Industry: segment size and growth
<X>2Industry: structure and attractiveness
<X>3Customer value: channel
<X>4Customer value: end customer
<X>5Relative position: capabilities
<X>6Relative position: costs
<X>7Competitor reaction
<X>8Sceptic's condition (verbatim)

Prompts for writing conditions under each heading:

  • Industry: How big must the segment be? Growing at what rate? What must be true of buyer power, supplier power, entry, substitutes, rivalry for returns to be earned?
  • Customer value: What must the customer value, relative to the alternative, and by how much? What must they be willing to give up or pay? What must the channel gain?
  • Relative position: What cost or capability level must we reach, by when? What must we be able to build or hire? What must we be able to stop?
  • Competitor reaction: What must rivals not do, or be unable to do? What must a new entrant be unable to replicate in the time available?

4. Barrier and test plan

One row per barrier condition (one to three per option).

OptionCondition IDWhy it is the barrier (lowest confidence, kills option if false)Test designed by (sceptic)MethodSample or scopePass mark (set before running)Cost and durationOwnerDue

Rules: the sceptic agrees in writing that a pass would change their view; the advocate agrees a fail would retire the option; tests are sequenced lowest confidence first across all options; a test that cannot be run in time becomes an assumption in section 9, not a pass.

5. Test results log

OptionCondition IDPass markResult (data, not interpretation)Verdict (pass, fail, conditional, untested)Threshold changed after result? (must be no)Sceptic's stated view after result

Stop rule: once an option records a fail on a barrier condition, mark all its remaining tests "not run, option eliminated" and record the cost saved.

6. Cascade choice record

Client: <name>          Decision: <framed choice>          Date approved: <date>
Chosen option: <letter and name>      Positioning base: <variety | needs | access>
Cascade questionAnswer (specific enough to exclude something)Explicitly excludedEvidence (condition IDs) or assumption
Winning aspiration: what does winning mean, against whom, by when, measured how?
Where to play: customers, geographies, categories, channels, value-chain stages
How to win: low cost or differentiation; why the chosen customer picks us over the best alternative; what sustains it
Capabilities: three to six, each marked exists / to be built, with the investment
Management systems: measures, processes, reviews, incentives that sustain the capabilities and monitor the conditions

Bottom-up coherence check (write one sentence per link):

  • Which management system sustains which capability?
  • Which capability delivers the how-to-win?
  • Why does the how-to-win work in the chosen where-to-play and not elsewhere?
  • How does the where-to-play, won, reach the aspiration?

Financial trajectory (from value-driver-modeling): year-by-year revenue and EBITDA for the chosen option and for the status quo; one-off costs; working-capital effect.

7. Activity-system map and fit tests

Three to five themes. Activities are things done in a particular way, not functions.

Theme (higher-order)Activities that deliver itReinforced by (name the other activity and the mechanism)Trade-off it requires

Fit tests (answer each with specifics; a blank is a fail):

  • First order, consistency: name any activity that works against how-to-win. Remove it or correct the cascade.
  • Second order, reinforcement: name at least three pairs where one activity makes another cheaper or better.
  • Third order, optimisation: name at least three activities the position lets the firm stop doing, and confirm it has stopped them.
  • Interlock: name the activity a rival would find hardest to copy and the two or more activities that make it hard.
  • Deletion test: remove any one theme. If the remaining activities still make sense alone, the system has no interlock.

8. Trade-off register

We will not doCurrent revenue, cost, or attention it consumesWhy doing it would break the system (image, activity conflict, coordination)Who will ask us to do it anywayStanding answer

Rule: if no line costs something the CEO would find painful, the register is incomplete.

9. Signpost register

Every conditional or untested condition gets a row, plus a standing growth-trap row.

Condition IDSignal that it is failingMeasureThresholdReview frequencyOwnerResponse if triggered
Growth trapProposal to broaden where-to-playAny board paper proposing new segments, geographies, channels, tiers or verticalsAnyEach board cycleCEOFull WWHTBT before approval

10. Rejected-options table

OptionPositioning baseOutcomeDeciding condition, pass mark, and evidenceCost of the testWould revisit if

Rule: "the team preferred another option" is not an entry. Every rejection cites a condition and data.

11. Partner review checklist

Run before the record goes to executive-communication.

  • The framed choice has at least two mutually exclusive answers, one of which is the status quo stated as a strategy with a trajectory.
  • Three to five options, each with an advocate, a three-sentence story, and a "stop doing" line that costs something real.
  • No two options could be pursued simultaneously without straddling.
  • Every condition is affirmative and falsifiable; the sceptic's objection appears verbatim for each option.
  • Conditions cover all four headings (industry, customer value, relative position, competitor reaction) for every option.
  • Each barrier condition has a test designed by the sceptic with a pass mark recorded before the result.
  • Tests were run lowest confidence first; testing stopped for eliminated options.
  • No threshold changed after results arrived.
  • Each cascade answer excludes something specific; the two largest competitors could not say the same where-to-play and how-to-win sentences.
  • The bottom-up coherence chain is written out and holds.
  • Activity map has three to five themes, activities not functions, at least three reinforcing pairs, at least three stopped activities, and a named hardest-to-copy activity.
  • Trade-off register contains at least one item the CEO will find painful.
  • Every conditional or untested condition has a signpost with measure, threshold, owner and response.
  • Every rejected option cites a failed or unconfirmed condition with data.
  • Segment revenues or ARR sum to totals; every margin row equals gross margin minus the cost lines shown; blended retention and margin figures are revenue-weighted averages of the segments; year-3 totals equal the sum of segments.
  • Hand-offs named: war-gaming (competitor conditions and scenarios), execution-planning (programme), kpi-architecture (scorecard), executive-communication (storyline).