Tests of a Strategy

Clarity, coherence, advantage logic, non‑obviousness, allocation, falsifiability.

The strategist’s work produces choices, not just analyses. Those choices must survive contact with markets, organizations, and time. A practical way to raise the odds is to subject any proposed strategy to a small set of tests—simple, repeatable questions that assess whether the plan is understandable, internally aligned, economically grounded, distinctive, resourced, and disprovable. This chapter presents six such tests—clarity, coherence, advantage logic, non‑obviousness, allocation, and falsifiability—and shows how to apply them in formulation, review, and governance.

The aim is not to assign a scholastic grade. Tests discipline conversation, force explicit trade‑offs, and create a record that can be audited later. When used well, they reduce the gap between a compelling narrative and a strategy that actually guides action.

I. How to Use the Tests

When. Use the tests at three moments: (1) during initial formulation to tighten the kernel; (2) just before committing resources; and (3) in post‑decision reviews to compare expected mechanisms with observed outcomes.

Who. The proposing team runs the tests first; a cross‑functional review (including finance, risk, and operations) repeats them; the approving forum (executive committee, board) signs off on the results, not only on the plan.

Artifacts. A one‑page strategy kernel (diagnosis, guiding policy, coherent actions) accompanied by a decision memo that records the answers to the six tests, the evidence cited, and the triggers for re‑assessment. The memo becomes part of the decision register (see Part V).

II. Test 1 — Clarity

What it asks. Can a competent outsider restate, in plain language, what we will do and not do, for whom, and why it should work?

Why it matters. Clarity is a prerequisite for coordination. Ambiguous strategies metastasize into conflicting projects, duplicated spend, and unaligned incentives.

How to check.

• Write the strategy in 100 words without jargon. It must include where to play (segments, geographies, jobs‑to‑be‑done), how to win (cost, differentiation, ecosystem position), and what we will not do (excluded customers, features, channels, or speeds).
• Produce a one‑page map of the few coherent actions with named owners and dates.
• Ask three managers outside the team to paraphrase the strategy. If their answers diverge, the text is unclear.
• Ensure every goal has an audience and a verb (“win self‑serve SMB in Tier‑2 cities by halving time‑to‑first‑value”) rather than an abstract aspiration.

Evidence of strength.

• Crisp where‑to‑play/how‑to‑win sentence; explicit trade‑offs.
• Shared vocabulary (definitions of segments, products, and metrics) published in a data dictionary.
• Communication artifacts reused across forums without re‑interpretation.

Red flags.

• “And” chains: “We will be premium and low‑cost and fast follower and disruptive.”
• Verb‑free nouns: “Digital transformation, customer centricity, and innovation at scale.”
• Strategy lives only in slides; different versions circulate for different audiences.

III. Test 2 — Coherence

What it asks. Do the chosen actions fit together, reinforce one another, and mesh with constraints (capabilities, platforms, regulation, culture)?

Why it matters. Isolated excellence rarely survives the system. Coherence is the property that makes the whole stronger than the parts—and prevents hidden contradictions from eroding value.

How to check.

• Draw a choice chain: a simple diagram linking 5–10 major choices (offer design, pricing, channel, service model, operations, data, organization) and annotate how each supports the others.
• Identify dependencies on shared platforms or scarce capabilities. Confirm capacity and sequencing with owners.
• Test for policy collisions: does a low‑touch sales model conflict with a customization promise? Do procurement savings undermine supplier quality needed for reliability claims?
• Use a pre‑mortem: assume failure in 24 months; list coherence breaks (“We raised price while cutting service, breaking trust with the segment we target”).

Evidence of strength.

• A small number of reinforcing routines (e.g., modular product architecture → faster release → premium reliability promise → pricing power).
• Sequencing makes sense: capability A is built before action B that depends on it.
• Non‑negotiables (safety, privacy, brand) are embedded, not bolted on.

Red flags.

• Each function has a plausible plan alone, but the plans compete for resources or violate one another’s assumptions.
• Platform deprecations required by the plan lack owners or dates.
• Reliance on “heroics” (exception processing, rush orders, ad‑hoc discounts) to reconcile contradictions.

IV. Test 3 — Advantage Logic

What it asks. What are the mechanisms by which this strategy creates and sustains superior economics compared with rivals? Why us, and why now?

Why it matters. Competitive advantage must be more than a hope. It should be a causal theory that connects resources and position to unit economics and returns on capital.

How to check.

• State the mechanisms explicitly: cost drivers (scale, experience, utilization, supply access), differentiation drivers (features, reliability, brand, service), and structural drivers (network effects, switching costs, regulation, control of bottlenecks, standards).
• Translate mechanisms into unit economics: price, variable cost, contribution margin, acquisition cost and payback, retention/cohort curves, capacity and yield.
• Show why these economics are defensible: what rivals would have to acquire, learn, or concede to match them.
• Anchor timing: what makes now the right moment (technology cost curves, regulation windows, customer readiness)?
• Identify the weakest link in the chain—often the assumption that turns a plausible story into a fragile one (e.g., willingness‑to‑pay for a new attribute, or complementor participation).

Evidence of strength.

• A concise diagram or paragraph tracing: asset/capability → activity choice → customer outcome → unit economics → returns/ROIC → reinvestment loop.
• Sensitivity analysis showing which assumptions move value the most and how exposure is hedged.
• Milestones that test the mechanism (e.g., “If attach rate of complement C does not exceed 25% in cohort month 3, network effect is too weak to sustain pricing tier Z.”)

Red flags.

• “We will be best‑in‑class” without a mechanism; reliance on “brand” as a residual category.
• Economics that work only at unattainable scale, with no credible path to reach it.
• Confusing correlation with cause (e.g., copying practices of current winners without context).

V. Test 4 — Non‑Obviousness

What it asks. What about this strategy departs materially from the default path or common playbook, creating a chance to outperform? What insight or asymmetry are we exploiting?

Why it matters. If the plan is obvious and cheap to imitate, the rents it creates will be competed away. Non‑obviousness does not require theatrics; it requires edge.

How to check.

• Write the counterfactual: what most competent rivals would do in our position. Highlight the differences.
• Identify the asymmetry we possess (data, access, trust, cost position, regulatory permission, complementary assets) that others lack or undervalue.
• Price the cost of being wrong: non‑obvious bets should be sized so that errors are survivable; options and stage‑gates (Ch. 21, 23) should be present.
• Surface the mechanism of insight: what customer, operational, or ecosystem learning produced the divergence? Can we keep that learning proprietary or compounding?

Evidence of strength.

• A clear “why others won’t or can’t follow” argument.
• A testable, contrarian hypothesis with a near‑term milestone (e.g., “Self‑serve onboarding can deliver enterprise‑grade adoption if time‑to‑first‑value < 10 minutes; we can achieve this via architectural change X that competitors cannot easily replicate.”)
• Use of under‑priced assets (e.g., overlooked segments, neglected channels, stranded capacity) with a plan to scale if the thesis proves right.

Red flags.

• Strategy indistinguishable from industry clichés (“customer intimacy,” “innovation,” “platform play”) without concrete choices.
• “Me too” responses to rivals’ moves, justified as table stakes.
• Over‑sized contrarian bets without options or hedges.

VI. Test 5 — Allocation

What it asks. Does the strategy move resources—capital, people, time—away from lower priorities and toward the chosen path? Are governance and incentives aligned with the choices?

Why it matters. Strategy that does not change budgets and calendars is commentary. Resource movement is the visible symptom of commitment.

How to check.

• Compare before/after budgets, headcount, and platform capacity by vector (Core/Grow/Seed) and by business unit. Look for concentration on the few things that matter.
• Confirm stage‑gated funding for uncertain bets, with explicit evidence thresholds and kill rules.
• Inspect decision rights (RAPID) to see whether owners can actually change pricing, product, channel, or policy in the service of the strategy.
• Verify incentives: variable pay and recognition schemes should reward outcomes consistent with the plan (e.g., retention and reliability, not just bookings).
• Measure portfolio velocity: the time from insight to reallocation; the reallocation ratio quarter‑over‑quarter.

Evidence of strength.

• A small number of funded, staffed initiatives with clear owners and dates; a published stop list of what will cease.
• A line of sight from strategy to platform roadmaps (what is being built or deprecated to make the plan possible).
• A reserved flex pool to reallocate as evidence arrives.

Red flags.

• “Last‑year‑plus” budgets; broad but shallow initiatives; peanut‑butter spread across units.
• Governance that requires broad consensus to act; veto rights scattered widely.
• Incentive plans unchanged from prior years despite new priorities.

VII. Test 6 — Falsifiability

What it asks. What observable evidence would demonstrate that the strategy—or a key part of it—is wrong or requires revision? What signposts and triggers convert observation into action?

Why it matters. Falsifiability disciplines belief. It prevents escalation of commitment and converts uncertainty into staged learning.

How to check.

• For each mechanism in the advantage logic, specify leading indicators that would move first if the mechanism is true (e.g., time‑to‑first‑value, attach of complements, early churn).
• Define signposts (external events) that separate scenarios (e.g., regulation passes with clause X; cost curve crosses threshold Y).
• Convert them into triggers with bands and time windows (e.g., “If referral share < 15% for two consecutive cohorts by month 3, pause expansion and revisit onboarding.”).
• Record review dates in the decision register; appoint owners to watch each trigger.
• Commit to post‑decision reviews: short notes explaining what was learned and how the plan changed (or why it didn’t).

Evidence of strength.

• A short appendix of if‑then rules linked to owners and forums.
• Options designed into the plan (deferral rights, exit clauses, modularization), with explicit premia and exercise windows.
• Willingness to stop or pivot in public when thresholds are missed.

Red flags.

• Success defined only by long‑lag outcomes (annual revenue) with no drivers.
• Triggers so loose that they never fire, or so tight that they whipsaw action.
• Ex‑post rationalization replacing ex‑ante criteria.

VIII. Running a Strategy Quality Review

A Strategy Quality Review (SQR) is a time‑boxed session—typically 90 minutes—for any material proposal. The agenda mirrors the six tests:

1. Clarity (10 minutes). Presenter delivers the 100‑word statement and the one‑page map. Three reviewers paraphrase the plan; discrepancies are captured.
2. Coherence (15 minutes). Walk the choice chain and dependencies; platform owners confirm feasibility and sequencing.
3. Advantage logic (25 minutes). Present the mechanism‑to‑economics chain and sensitivity; reviewers attack the weakest link.
4. Non‑obviousness (10 minutes). Contrast with default path; state the asymmetry and the price of being wrong.
5. Allocation (15 minutes). Before/after budget and capacity; stage‑gates and kill rules; decision rights and incentives.
6. Falsifiability (15 minutes). Signposts, triggers, and review dates; owners confirm.
7. Decision (10 minutes). Approve, approve with conditions (documented in the decision log), or return for revision. Record the outcome and schedule the first post‑decision review.

The SQR is not a fishbowl for performance theater; it is a disciplined rehearsal that sharpens the plan and leaves an auditable trail.

IX. Applying the Tests at Different Levels

Corporate portfolio. Clarity and non‑obviousness focus on where to compete (businesses to emphasize or exit); coherence evaluates the corporate center’s parenting advantage; advantage logic stresses synergies vs. illusions; allocation inspects capital across units; falsifiability sets divestiture and M&A triggers.

Business unit. Clarity centers on segment and channel choices; coherence inspects product–channel–service fit; advantage logic drills into unit economics; non‑obviousness tests for a wedge; allocation checks platform roadmap and sales capacity; falsifiability uses market and cohort signposts.

Function or platform. Clarity defines service levels and scope; coherence checks API and process integration; advantage logic justifies build vs. buy vs. partner; non‑obviousness identifies the capability edge; allocation moves headcount and capex; falsifiability ties to SLOs and incident thresholds.

X. A Simple Scoring Rubric (No Tables Required)

For each test, assign 0–5 based on the following guidance:

• 5 – Exemplary. Clear, specific, and evidence‑backed; explicit trade‑offs; triggers and owners named.
• 4 – Strong. Minor gaps; mechanisms plausible; allocation moves visible.
• 3 – Adequate. Understandable but generic; some contradictions unresolved; limited movement of resources.
• 2 – Weak. Vague wording; unproven mechanism; little or no reallocation; no triggers.
• 1 – Poor. Buzzwords; contradictory actions; me‑too; budgets unchanged; unfalsifiable.
• 0 – Not addressed. Missing content.

The score is less important than the commentary that explains why. Keep the rubric in the decision register; revisit the original score in post‑decision reviews.

XI. Common Failure Patterns (and Antidotes)

• Narrative drift. The story evolves across audiences. Antidote: freeze a canonical 100‑word statement and glossary; require paraphrase checks.
• Laundry‑list plans. Many initiatives, none critical. Antidote: force a stop list and concentration test (top three initiatives account for most discretionary spend).
• Wishful synergies. Corporate plans assume synergies without mechanism. Antidote: name the synergies as projects with owners, costs, and milestones—or remove them from the case.
• Price without mechanism. Margin expansion premised on price hikes absent differentiation or alternatives. Antidote: tie pricing power to attributes customers value and to switching costs; test via experiments.
• Under‑resourced pivots. Strategic shifts declared without platform or talent changes. Antidote: move headcount and capex first; adjust incentives before messaging.
• Trigger amnesia. Thresholds set, then forgotten. Antidote: route breaches automatically to forums; require written decisions to defer.
• Over‑fit to a tool. Forcing a two‑by‑two or a generic playbook onto a context it does not fit. Antidote: start from mechanisms; select tools only if they interrogate those mechanisms.

XII. A 30‑Day Strategy Quality Sprint

When a strategy is fuzzy or contested, run a short quality sprint:

Week 1 — Clarify and map.

• Draft the 100‑word statement and the one‑page action map.
• Identify the 8–10 most consequential choices and draw the choice chain.

Week 2 — Prove the mechanism.

• Build a lightweight unit‑economics model; run sensitivity to find the fragile assumptions.
• Write the non‑obvious wedge and the asymmetry we exploit.

Week 3 — Commit and guard.

• Propose before/after allocations; define stage‑gates and kill rules; update decision rights and incentives needed.
• Draft signposts and triggers; schedule post‑decision reviews.

Week 4 — Decide.

• Hold the SQR; record conditions, owners, and dates; publish the final kernel and memo.

This sprint does not replace deep analysis; it ensures that analysis produces decisions and commitments rather than slides.

XIII. What “Good” Looks Like (Illustrative Fragments)

• Clarity. “We will win mid‑market healthcare providers in Region A by offering a self‑serve analytics product that cuts claims processing time by half. We will not pursue national payers in this cycle.”
• Coherence. “We shift to self‑serve onboarding → requires modular templates and guardrails → reduces implementation cost → enables price tier S; success depends on platform capability P and a 24/7 support rota in two languages.”
• Advantage logic. “Proprietary labeled data from X clinics improves model precision by 20% → reduces false positives → lifts provider productivity → willingness‑to‑pay supports 30% price premium; cost to replicate is high due to data access constraints.”
• Non‑obviousness. “Competitors target payers; we target providers where switching costs are lower and consensus cycles shorter. Our partner network provides reach and trust incumbents lack.”
• Allocation. “Shift 15 FTE and $8M capex from custom integrations to template and data‑pipeline rebuild; freeze two regional expansions; stage‑gate product Y pending retention thresholds.”
• Falsifiability. “If activation < 40% by Day 7 for two consecutive cohorts, halt paid acquisition and revise onboarding; if regulation Z passes with clause Q, suspend cross‑border data flow and move compute to Region B.”