Communication & Deliverables

Module 05

Communication & Deliverables

The Pyramid Principle, the SCQA opening, structuring a recommendation, decks that land, and data storytelling. Brilliant analysis nobody acts on is worth nothing.

Here's the truth that surprises new consultants: the quality of your thinking matters less than your ability to make a busy, skeptical decision-maker believe it and act. Communication isn't the gift wrap on the analysis — it is half the job. This module is how you stop being right in private and start being effective.

The Pyramid Principle — answer first

Barbara Minto's idea, and the most important communication habit in consulting: lead with the answer, then support it. Beginners build up to their conclusion like a detective novel — context, analysis, then finally the recommendation. Executives hate this; they want the punchline immediately, then the reasoning, then the detail. Structure it as a pyramid: one main recommendation at the top, supported by three-ish key arguments, each supported by evidence. The reader can stop at any level and still have a coherent answer.

SCQA — how to open without rambling

A reliable way to start any document or pitch: Situation (the stable context everyone agrees on) → Complication (what changed or went wrong) → Question (the question that raises) → Answer (your recommendation). "You've grown 40% a year for three years [situation]. This year growth stalled to 5% [complication]. Why, and what do we do? [question] Reposition for enterprise buyers — here's why [answer]." It gets you from a blank page to a sharp opening every time, and it earns the right to be heard before you make your case.

Structure every recommendation the same way

  1. The recommendation — what to do, in one clear sentence.
  2. The reasons — the two to four arguments that make it right (grouped MECE).
  3. The evidence — the data and analysis under each reason.
  4. The 'so what' and next step — what changes if they act, and the first concrete move.

If you can't state the recommendation in one sentence, you're not ready to present — go back to the analysis.

Slides — one idea per slide, and the title carries it

The single biggest upgrade to anyone's decks: write the slide's main message as its title, in a full sentence. Not "Q3 Revenue" (a label) but "Q3 revenue fell 18%, driven entirely by churn in enterprise accounts" (a message). If someone reads only your slide titles in order, they should get your whole argument. Each slide then carries one idea, with the chart as evidence for it — never a cluttered dashboard the audience must decode.

Telling a story with data

A chart's job is to make one point instantly. Before making any chart, write the sentence it must prove, then strip away everything that doesn't serve it — extra series, decoration, 3-D effects, rainbow colours. Use colour to direct attention to the one bar that matters, not to brighten the slide. The best data storytelling feels almost too simple; the work was deciding what to leave out.

The rule of three, and the meeting itself

People remember three things, not seven. Force your case into three key points and they'll recall it; give them ten and they'll recall none. In the room, state your recommendation in the first minute. Welcome interruptions — a client poking holes is engaged, and you'd rather surface the objection now than have it kill the project silently later. Know the one number that anchors your case, and know the strongest argument against you better than your critics do. Confidence with humility: "Here's what I'd do and why — and here's what would change my mind."

Key takeaway

Open with SCQA, answer first, then support. Make every slide title a full-sentence message, every chart prove one written sentence, and force your case into three points. Brilliant analysis delivered as a mystery novel loses to decent analysis delivered as a pyramid.

For Orelis & the app

Output design is a feature, not an afterthought. Most AI tools dump a wall of text — the consulting move is to structure it: lead with the single recommendation, then reasons, then evidence, each expandable. If your app generates a clean, answer-first one-pager the user can hand to their boss, you've turned "the AI gave me advice" into "the AI gave me something I can act on and defend." That's the difference between a toy and a tool.

Test yourself

Q1Rewrite this slide title into a message: 'Marketing Spend by Channel, 2024.' (Invent a plausible insight.)
Show a worked answer
Anything that states the point rather than the topic, e.g.: 'Two-thirds of 2024 marketing spend went to paid social, yet it drove only a fifth of new customers.' Now the title makes an argument and the chart is just evidence for it. A reader skimming only titles still learns something — the whole test.
Q2You have ten minutes with a CEO after three weeks of analysis. In what order do you present, and why?
Show a worked answer
Recommendation first (the one-sentence answer), then two-to-four reasons, then evidence only as deep as time and questions demand, then the concrete next step. The CEO's attention is the scarce resource; leading with the conclusion lets them engage and probe immediately, and if you're cut off at minute three they've still heard the answer. Saving the conclusion for the end risks never delivering it.
Q3Open a recommendation using SCQA for a retailer whose foot traffic is fine but online sales are collapsing.
Show a worked answer
Situation: your stores still draw steady foot traffic and revenue. Complication: online sales have fallen 35% over two quarters even as the market shifts online. Question: why are we losing online, and how do we reverse it? Answer: the checkout flow added friction in the last redesign — fixing it is the fastest, cheapest lever, and here's the evidence. Notice it earns attention with shared context before asserting the answer.