cp-recommend

/cp-recommend — Consulting Process: Recommend

"Start with the answer. The Pyramid Principle inverts the academic convention: state the recommendation first, then provide the supporting arguments, then the evidence. Executives make decisions — they need the answer, not the journey."

Executes the Recommend phase of the McKinsey-style Consulting Process. Produces the executive recommendation deck structured using the Pyramid Principle and the SCQA framework.

THYROX Stage: Stage 5 STRATEGY.

Gate: Recommendation storyline reviewed by engagement lead before finalizing deck for client.

---

Consulting Process Cycle — focus on Recommend

flowchart LR
    I[Initiation\nProblem Def] --> D[Diagnosis\nMECE + Issue Tree]
    D --> S[Structure\nHypothesis + Workplan]
    S --> R[Recommend\nPyramid + SCQA]:::active
    R --> P[Plan\nImplementation]
    P --> Im[Implement\nExecution Support]
    Im --> E[Evaluate\nImpact Measurement]
    classDef active fill:#4a9eff,color:#fff

Pre-condition

• cp:structure complete: Consulting Workplan approved, all P1 and P2 analyses executed.
• Findings from at least 80% of planned analyses are available.
• At least one hypothesis has been confirmed as the primary driver.

---

When to use this step

• When analyses are complete and findings need to be synthesized into a recommendation
• When you need to structure the recommendation for an executive audience (C-suite, board, steering committee)
• When multiple findings need to be integrated into a single coherent storyline

When NOT to use this step

• Without completed analyses — synthesizing incomplete findings produces a weak recommendation
• For operational-level communication (working sessions with client counterparts) — use simpler formats
• If the client has already decided — check whether the engagement is advisory or decisional before building a full recommendation deck

---

Activities

1. SCQA — Situation, Complication, Question, Answer

SCQA is the storyline framework that structures the opening of any recommendation. It moves the audience from shared context to the consultant's recommendation.

The four elements:

Element	Definition	Length	Tone
Situation	The current state — facts the audience already agrees with	2-3 sentences	Neutral, factual
Complication	What changed, what's wrong, or what's at risk — the tension that creates urgency	2-3 sentences	Urgent, specific
Question	The natural question the complication raises in the audience's mind	1 sentence	Implicit or explicit
Answer	The recommendation — the direct response to the question	1-2 sentences	Decisive, direct

Example:

Situation: RetailCo has maintained 8% operating margin for six years, consistently
           outperforming the industry average of 5.5%.

Complication: Over the past three years, margin has declined to 4%, driven primarily
              by unmanaged discounting in the SMB segment — discounts averaging 22%
              vs a 10% list price policy — costing the company $18M annually.

Question: [Implicit: How do we restore margin to historical levels?]

Answer: RetailCo should implement a structured discount governance program in SMB,
        which our analysis shows could recover $12-15M in annual margin within 12 months.

See full SCQA and Pyramid Principle guide: pyramid-principle-scqa.md

2. The Pyramid Principle — vertical and horizontal logic

The Pyramid Principle (Barbara Minto, McKinsey) structures the recommendation as a pyramid: the answer at the apex, supported by arguments in the middle, supported by evidence at the base.

Structure of the Pyramid:

                    [ANSWER]
                   /    |    \
          [Arg 1]   [Arg 2]   [Arg 3]
          /  |  \   /  |  \   /  |  \
        E1  E2  E3 E4  E5  E6 E7  E8  E9

Vertical logic — each level proves the level above:

• Answer: "RetailCo should implement discount governance"
• Arg 1: "Unmanaged discounting is the primary margin driver" → proves this is the right problem
• Evidence for Arg 1: discount waterfall analysis showing 22% avg discount vs 10% policy

Horizontal logic — arguments at the same level are MECE:

• Arg 1: The problem is real and quantified (diagnosis)
• Arg 2: Discount governance is the highest-impact solution (alternatives assessed)
• Arg 3: Implementation is feasible within 12 months (feasibility confirmed)

MECE test for arguments: The three arguments together should answer the question completely — no overlap, no gaps.

3. Vertical logic test

For every argument, ask: "Does this directly support the answer above it?"

Answer	Argument	Vertical logic test	Pass/Fail
"Implement discount governance"	"Discounting costs $18M/year"	Does this prove the recommendation? Partially — it shows the problem, not the solution.	Partial — needs pairing with feasibility argument
"Implement discount governance"	"Governance programs in comparable companies recovered 70-80% of margin leakage within 12 months"	Does this prove the recommendation is the right action? Yes.	Pass

4. Horizontal logic test (arguments are MECE)

Arguments at the same level must be:

• ME: No argument duplicates another (no two arguments make the same point)
• CE: Together, the arguments answer the question completely

Example — bad horizontal logic:

Arg 1: The problem is large ($18M)
Arg 2: The problem is growing
Arg 3: We are losing customers because of the problem

These three arguments all support "the problem is serious" — not MECE relative to each other AND they don't address the solution at all.

Example — good horizontal logic:

Arg 1: Unmanaged discounting is the primary margin driver (diagnosis confirmed)
Arg 2: Discount governance is the most impactful lever among alternatives assessed (solution selected)
Arg 3: Implementation is feasible with existing systems within 12 months (feasibility)

Each argument answers a different dimension of "why this recommendation" — MECE.

5. Storyline structure — from SCQA to deck

A standard executive recommendation deck follows this flow:

Slide / section	Content	Pyramid level
Executive summary	SCQA + answer + 3 key arguments (1 slide)	Apex
Context	Situation details — agreed facts	Situation
The problem	Complication quantified — what's wrong and how much	Complication
What we analyzed	Issue Tree overview + analytical approach	Process
Finding 1 — [Arg 1 title]	Evidence supporting argument 1 (2-3 slides)	Evidence
Finding 2 — [Arg 2 title]	Evidence supporting argument 2 (2-3 slides)	Evidence
Finding 3 — [Arg 3 title]	Evidence supporting argument 3 (2-3 slides)	Evidence
Recommendation	Answer restated + 3 arguments summarized	Apex restated
Implementation overview	High-level plan (detailed in cp:plan)	Supporting
Financials	Expected impact ($, %) and timeline	Supporting
Risks and mitigations	Key risks, not exhaustive	Supporting
Next steps	Decision needed from sponsor	Call to action

See full slide deck structure template: recommendation-deck-structure.md

6. The executive summary slide — most important slide

The executive summary slide is the single most important slide in the deck. It is often the only slide a senior executive reads before the presentation.

Structure of the executive summary slide:

Title: "RetailCo should implement discount governance to recover $12-15M margin"

Situation: [1-2 sentences — agreed context]
Complication: [1-2 sentences — the problem with numbers]

Recommendation: [1 sentence — the answer]

Supporting arguments:
• [Arg 1 — 1 sentence with key number]
• [Arg 2 — 1 sentence with key number]
• [Arg 3 — 1 sentence with key number]

Expected impact: $12-15M annual margin recovery within 12 months
Investment required: $0.8M in system and process changes

Rule: If the client reads only the executive summary slide, they should understand the full recommendation, the problem it solves, and why it is the right answer.

7. Storyline review — before finalizing deck

Check	Pass / Fail	Notes
SCQA flows naturally (S→C→Q→A)
Answer is stated in the first slide (not at the end)
3-5 arguments support the answer (not more, not fewer)
Arguments are MECE (no overlap, no gap)
Each argument has at least 2 pieces of evidence
Vertical logic: each evidence directly supports its argument
Executive summary slide can stand alone
No slide presents data without a clear "so what"
Recommendation is specific and actionable

---

Expected Artifact

{wp}/cp-recommend.md — use template: recommendation-deck-structure.md

---

Red Flags — signs of Recommend done poorly

• "Boiling the ocean" deck — more than 30 slides, covering every analysis run; the deck should show the best 20% of the work, not all of it
• Answer buried at the end — academic structure (data → analysis → conclusion) for an executive audience is a common failure; executives want the answer first
• Arguments repeat each other — if Arg 1 and Arg 2 are both saying "the problem is large," only one argument is needed
• Every slide is a data dump — slides should have headlines that state the finding ("Discounting costs $18M/year") not descriptive titles ("Discount Analysis")
• No quantified impact — a recommendation without a $-amount or %-change is qualitative advice, not a consulting recommendation
• Risk section is missing — acknowledging risks demonstrates credibility; omitting them signals overconfidence

Anti-rationalization

Rationalization	Why it's a trap	Correct response
"The client needs to see all our analysis to trust us"	Volume of analysis ≠ quality of recommendation; too much data buries the answer	Show only the analyses that directly support the 3 key arguments
"The conclusion at the end is more persuasive"	For C-suite audiences, the opposite is true — they want the answer immediately	State the recommendation on slide 1; build the case after
"We can't be too direct — the recommendation is sensitive"	Indirect recommendations are interpreted as lack of conviction	State the recommendation directly; acknowledge sensitivity in the risk section

---

Estado en now.md

Al INICIAR este step:

methodology_step: cp:recommend
flow: cp

Al COMPLETAR (Storyline approved by engagement lead):

methodology_step: cp:recommend  # completado → listo para cp:plan
flow: cp

Siguiente paso

When recommendation deck is approved by engagement lead → cp:plan

---

Limitations

• The Pyramid Principle is optimized for executive communication; for technical or scientific audiences, a modified inductive structure may be more appropriate
• SCQA assumes the audience agrees with the Situation — if the Situation is disputed, start with data to establish it before introducing the Complication
• Vertical logic can be tested formally; horizontal MECE is more subjective for qualitative problems — get a second opinion from a peer outside the workstream

---

Reference Files

Assets

• recommendation-deck-structure.md — Full slide deck structure template with section-by-section guidance, executive summary template, and storyline review checklist

References

• pyramid-principle-scqa.md — Complete Pyramid Principle guide: vertical logic, horizontal MECE, SCQA framework, slide headline writing, and common structural failure modes