MBA Business Communication Concept Card Template

Business Communication Concept Card: [Your memo, deck or course]

A concept card on writing and presenting for decision makers, the communication thread that runs through every MBA core course, case write-up, internship and job. It covers the pyramid principle (answer first, then grouped support), the structure of an executive summary and a decision memo, building a storyline whose slide titles state the point, data visualisation for executives, and plain writing, with a fully worked before-and-after memo rewrite. It is for MBA students, applicants preparing written cases and presentations, and managers who write for senior readers. The examples use Northfield Outdoor Co., Corvane Robotics and Tallow Creek Medical Center, all fictional; all data are invented. Source basis: Minto's pyramid principle (Minto 2009, revised edition), Flesch (1948), Tufte (1983), Cleveland and McGill (1984) and US federal plain language guidance, written up in original words and examples. Every number was computed in Python and is shown with the Excel formula where one applies. Replace every [bracketed] value and every row marked Example.

Document control

When to use it

Use this card whenever a busy reader must understand, decide or act on what you write: a case write-up, a recommendation to a professor playing the client, an internship final presentation, a board paper or an email to a senior manager.

Do not use the answer-first structure blindly where the reader must be brought along slowly: a hostile audience, bad news to people affected, or a culture where a direct recommendation from a junior writer would be resented. Even then, know your answer first and decide consciously how and when to reveal it.

The idea in plain language

Senior readers have little time and many documents. They read the first lines, decide whether the rest matters, and look for what they must decide. Good business writing is therefore built from the top down:

1. Answer first. State the recommendation or main conclusion at the start, then support it. This is the core of Minto's pyramid principle: ideas form a pyramid, with one governing thought at the top, a few key supporting points below it, and evidence below each of those.
2. Group and order the support. Each level summarises the ideas below it. Ideas in a group are of the same kind (three reasons, three steps, three options), they do not overlap and together they cover the point (often called MECE: mutually exclusive, collectively exhaustive). The order is logical: by time, by structure or by importance.
3. Answer the reader's next question. Each statement raises a question ("why?", "how?", "what does it cost?") and the level below answers it. Minto describes this as a question-and-answer dialogue between writer and reader.
4. Introduce with a story the reader already accepts. The situation (what the reader knows), the complication (what changed or what is at stake) and the question it raises lead straight to your answer. Keep it short: two to four sentences.

Slides follow the same logic. Each slide title is a full sentence that states the point of the slide ("Dalton Street is the only loss-making store"), not a topic label ("Store performance"). Read in sequence, the titles alone tell the whole argument. Charts make one comparison obvious and do not distort the data. Sentences are short, active and concrete, with numbers where numbers matter.

Formulas and notation

Flesch Reading Ease (Flesch 1948):

$FRE = 206.835 - 1.015\,\frac{W}{S} - 84.6\,\frac{Y}{W}$

For example, a passage of 120 words in 4 sentences with 198 syllables scores $206.835 - 1.015 \times 30 - 84.6 \times 1.65 = 36.8$ (hard); the same 120 words in 8 sentences with 168 syllables scores $73.2$ (fairly easy). Excel: =206.835-1.015*(120/4)-84.6*(198/120). Shorter sentences and shorter words both raise the score.

Lie factor (Tufte 1983): the ratio of the effect shown in a graphic to the effect in the data. A faithful chart has $LF$ close to 1.

$LF = \frac{\Delta_{shown}}{\Delta_{data}}$

Data-ink ratio (Tufte 1983): the share of a graphic's ink that shows data rather than frames, grids and decoration. It is a principle more than a measurement; remove ink that does not inform.

$DIR = \frac{\text{ink used to show data}}{\text{total ink used in the graphic}}$

Example (fictional). Corvane Robotics' gross margin rose from 30.0 to 31.5 percent, a relative change of $\Delta_{data} = 1.5/30.0 = 0.050$. A bar chart with the axis starting at 29.0 percent draws bars of height 1.0 and 2.5, a visible change of $\Delta_{shown} = (2.5 - 1.0)/1.0 = 1.500$. $LF = 1.500/0.050 = 30.0$: the chart exaggerates the improvement thirty-fold. Bars must start at zero; if a small change matters, show it as a number or with a dot plot labelled with the change.

Assumptions and limits

• Answer first assumes you have an answer. In exploratory work, state the question, what you know and what you will do next; do not fake a recommendation.
• Readability scores measure surface features. They ignore logic, accuracy and whether the reader knows the terms. A text full of short, vague sentences can score well and say nothing. Syllable counts also differ between tools by a few points.
• Culture and hierarchy matter. Some organisations and countries expect context before the conclusion, or agreement from stakeholders before a written recommendation. Adapt the order of the introduction, not the clarity of the thinking.
• Decks are not documents. A slide deck read without a presenter needs fuller sentences than one that is presented. Decide which you are writing.
• Grouping in threes is a habit, not a rule. Use as many groups as the logic needs, usually two to five.
• Charts can mislead without lying. Choice of baseline, period and comparison group shapes the message; disclose them.

How managers use it

• Decision memos. Recommendation in the first sentence, then the reasons, the numbers, the risks and the specific decision needed with a date. Many senior readers decide from the first paragraph and read the rest only if they disagree.
• Executive summaries. A stand-alone page that a reader can forward without the report. It answers what, why, how much, what risk and what next. Write it last, put it first.
• Ghost decks and storylines. Before any analysis is finished, teams write the slide titles they expect to prove, then do the analysis and rewrite titles to match what the data show. The storyline keeps work focused; the rewrite keeps it honest.
• Board and investor papers. Boards need the question they are asked to decide, the options considered and the risks, not the full working (startup/board-deck).
• Email to senior people. Subject line states the ask ("Approval needed by 15 Jan: Dalton Street closure"); first line repeats it; details below.
• Writing as thinking. Writing the pyramid exposes gaps: a key point with no evidence, a group that overlaps, a recommendation that does not follow. Fix the thinking, then the prose.

Plain language is also public policy in some places: in the United States, the Plain Writing Act of 2010 requires federal agencies to write public documents clearly, and the government's plain language guidance (now published on digital.gov) sets out the practices: know the audience, organise for the reader, use short sentences, active voice and common words, and test with real readers. The guidance is written for government but applies to any business document.

Data visualisation for executives

Cleveland and McGill (1984) tested how accurately people read quantities from different graphical encodings. Position along a common scale was read most accurately, followed by position on non-aligned scales, length, then angle and slope, then area, with volume and colour saturation least accurate. The practical rule: encode the comparison that matters as position or length on a common baseline.

Executive chart checklist: a title that states the point; units and period in the subtitle or axis; bars from zero; direct labels instead of a legend where possible; one highlight colour for the item that matters and grey for the rest; the source and date below the chart.

The chart below applies the checklist to the Northfield store review (fictional data, contribution after store costs, USD m, last financial year).

The total contribution of the eight stores is USD 2.20m; Dalton Street alone subtracts USD 0.42m from it.

Worked example A (fully solved): rewriting a decision memo

Situation (fictional). A retail analyst at Northfield Outdoor Co. must ask the chief operating officer to approve closing the Dalton Street store. The underlying numbers (USD m): revenue 3.60; gross margin 42 percent, so gross profit 1.512; store costs 1.93 (rent 0.86, staff 0.84, other 0.23); contribution $1.512 - 1.93 = -0.418$. After closure, the team expects 55 percent of the store's revenue to move online (1.98) at a 38 percent gross margin (0.7524), less fulfilment at 9 percent of that revenue (0.1782) and a pickup-point fee of 0.12, giving a contribution of 0.4542. One-off exit costs are 0.90 (lease exit 0.65, severance 0.25).

Step 1: compute the numbers the reader needs.

Step 2: the original draft (before).

As you know, the retail team has been looking at store performance for some time now, and over the last quarter we carried out a detailed review of all eight stores, including footfall, conversion, staffing and lease terms, with input from finance, property and the e-commerce team. The review raised a number of interesting points about how customer behaviour is changing, particularly the growth in online orders from postcodes near some of our stores. In the case of Dalton Street, revenue was USD 3.60m last year and, after rent, staff and other costs, the store made a loss of about USD 0.42m, although it should be noted that there are several factors to consider, including the lease break option that becomes available next March and the possibility that some customers would move to our website. Taking all of this into account, it may be worth considering whether the store should be closed, subject to further discussion.

Step 3: diagnose. The draft is chronological (what we did, then what we found, then a tentative idea). The recommendation is in the last sentence and hedged ("it may be worth considering"). The decisive numbers (payback, downside) are missing. There is no ask and no deadline, although the lease break creates one. Sentences average 39 words.

Step 4: the rewrite (after).

We recommend closing the Dalton Street store at the lease break next March and opening a pickup point nearby. The store lost USD 0.42m last year on revenue of USD 3.60m. If 55% of its sales move online, profit improves by USD 0.87m a year and the USD 0.90m exit cost pays back in about one year. Even if no customer moves online, closure still saves USD 0.30m a year. We need your approval by 15 January to give notice on the lease.

Step 5: measure the change. Syllables were counted with a simple vowel-group rule, so the Flesch scores are a heuristic, not an exact measurement: a second rule-based counter gave about 41 before and 77 after. Read the direction and size of the change, not the decimals.

Step 6: check the rewrite against the numbers and the risks. Every figure in the rewrite matches Step 1 (0.87 is 0.8722 rounded; 0.30 is 0.298 rounded). The rewrite states the downside honestly: with zero retention the closure still saves money each year, but the three-year NPV is slightly negative (USD -0.146m), so the case depends on keeping some customers. The full memo should say this in the risk paragraph, together with what the team will do to keep customers (pickup point, targeted offers). Hiding the downside to make the recommendation look stronger would be a mistake in both ethics and persuasion: the chief operating officer's finance team will find it.

Step 7: the rest of the memo. Below the opening paragraph: (1) why the store loses money (rent rose at the last renewal, footfall fell), (2) why customers can be kept (online order growth from nearby postcodes, pickup point), (3) the economics with the table and the zero-retention case, (4) risks and mitigations (staff redeployment, local reputation, retention lower than planned), (5) next steps and owners. Each paragraph starts with its point.

Worked example B (half solved): a storyline of action titles

Situation (fictional). Corvane Robotics, a growth-stage warehouse-robotics company, will ask its board to approve a second production line. The analysis shows: order backlog has grown for six quarters and now equals nine months of output; the current line runs at 94 percent utilisation and lead times have doubled; two large customers have said they will not reorder at current lead times; the second line costs USD 14m and pays back in about four years in the base case; the main risk is a slowdown in warehouse automation spending, which a phased build reduces.

Questions to complete:

1. Write the four missing titles (slides 2, 4, 5, 6) and complete slide 7.
2. Which slides form the "why act" group and which the "why this option" group in the pyramid?
3. For slide 3, which chart would you use, and what would its title and highlight be?
4. Write the situation, complication and question for the opening (two to three sentences).

Exercise C (unsolved): an executive summary for the emergency department

Situation (fictional). Tallow Creek Medical Center's emergency department staffs 3 physicians between 10:00 and 14:00. A queueing analysis (mba/mba-queueing-card, Worked example A) shows an average wait for a physician of 35.11 minutes, with 52.0 percent of patients waiting more than 15 minutes. With a fourth physician the average wait falls to 5.33 minutes and the share waiting more than 15 minutes to 13.0 percent. The finance team estimates that a fourth physician in this block costs USD 0.29m a year (given). The clinical director's targets are an average below 15 minutes and no more than 20 percent waiting longer than 15 minutes.

1. Write the governing thought in one sentence.
2. Write an executive summary of at most 200 words for the hospital's chief executive, answer first, with the ask and a date.
3. Draw a chart (on paper) that shows the choice, and write its action title.
4. List two questions the chief executive is likely to ask, and say where in your summary each is answered.
5. Measure your draft's words per sentence and compare it with the "after" memo in Worked example A.

Connections to other cards

• mba/mba-case-write-up and mba/mba-case-prep-sheet: the case method rewards an answer-first recommendation backed by analysis.
• mba/mba-group-project-report and mba/mba-internship-final-presentation: storylines and action titles at length.
• mba/mba-case-interview-prep: consulting interviews test structured, top-down communication of an answer.
• mba/mba-queueing-card, mba/mba-cvp-breakeven-card and mba/mba-npv-irr-card: sources of the numbers that a decision memo summarises.
• mba/mba-ethics-guide and mba/mba-stakeholder-esg-card: honest disclosure of downside and of who is affected.
• business/memo, business/executive-summary, business/one-pager and startup/board-deck: ready-to-fill formats.
• writing/style-guide and data/dashboard-specification: house style and chart standards.

Common mistakes and misconceptions

• Chronological structure. "First we did, then we found" makes the reader wait for the point. Lead with the answer.
• Hedged recommendations. "It may be worth considering" signals uncertainty the analysis may not justify. State the recommendation and put real uncertainty in a risk section with numbers.
• Topic titles on slides. "Financials" says nothing. A title is a sentence with a verb that states the conclusion.
• Titles the chart does not prove. Each title must be supported by the evidence on that slide; otherwise the reader stops trusting the deck.
• Groups that overlap or miss something. Reasons like "cost", "rent" and "profitability" overlap; rewrite them into distinct groups.
• No ask, no deadline. A decision memo ends with what you need from the reader and by when.
• Hiding the downside. Senior readers look for the weak point; showing it, with a mitigation, builds credibility.
• Distorting charts. Truncated bar axes, 3D effects, dual axes and inconsistent scales mislead even when every number is correct. Check the lie factor.
• Too much on one slide. One message per slide; move detail to an appendix.
• Jargon and acronyms. Write for the least specialised reader who must decide; define any term they may not know.
• Numbers without context. "USD 0.42m loss" means more next to revenue, the other stores and the cost of fixing it.

Self-test questions

1. State the pyramid principle in two sentences.
2. What are the situation, complication and question in an introduction, and why should the situation contain nothing new to the reader?
3. Rewrite the slide title "Market overview" as an action title for a fictional slide showing that the market for warehouse robots grew 18 percent last year while Corvane's sales grew 31 percent.
4. What does MECE mean, and why does overlap between groups weaken an argument?
5. Compute the Flesch Reading Ease of a passage with 90 words, 6 sentences and 135 syllables. Give the Excel formula.
6. A line chart shows a cost rising from 100 to 104, drawn with an axis from 99 to 105 so the line rises across almost the whole height. Is that a problem in the same way as a truncated bar chart? Explain.
7. According to Cleveland and McGill (1984), which encoding is read most accurately, and what does that imply for a pie chart with seven slices?
8. List the five questions an executive summary should answer.
9. When would you not lead with the recommendation, and what would you do instead?
10. In Worked example A, why does the memo mention the zero-retention case, and what does its negative three-year NPV imply for the recommendation?

My notes and next steps

Sources and further reading

• Minto, B. (2009). The Pyramid Principle: Logic in Writing and Thinking (rev. ed.). Financial Times Prentice Hall. Earlier editions from Minto International, London (a revised second edition is catalogued for 1978). (The pyramid structure, the question-and-answer dialogue and the situation, complication and question introduction.)
• Flesch, R. (1948). A new readability yardstick. Journal of Applied Psychology, 32(3), 221-233.
• Tufte, E. R. (1983). The Visual Display of Quantitative Information. Graphics Press. (Lie factor and data-ink ratio.)
• Cleveland, W. S., and McGill, R. (1984). Graphical perception: theory, experimentation, and application to the development of graphical methods. Journal of the American Statistical Association, 79(387), 531-554.
• US General Services Administration, Digital.gov. Plain language guide series (adapted from PlainLanguage.gov), https://digital.gov/guides/plain-language (accessed 2026-10-28). Public domain, US government work.
• OpenStax (2019). Principles of Management, chapter 16 on managerial communication, https://openstax.org/details/books/principles-management (accessed 2026-10-28). CC BY-NC-SA 4.0. Open resource.
• Further reading (by name only): Knaflic, Storytelling with Data; Few, Show Me the Numbers; Williams, Style: Lessons in Clarity and Grace; Zinsser, On Writing Well.

Answer key

Worked example B. (1) Example titles: slide 2 "Our order backlog has grown for six quarters and now equals nine months of output"; slide 4 "Two of our largest customers will not reorder at current lead times"; slide 5 "The second line costs USD 14m and pays back in about four years in the base case"; slide 6 "A phased build limits our exposure if automation spending slows"; slide 7 "... and phase two at the [month YYYY] review, once backlog and orders confirm demand". (2) "Why act": slides 2, 3 and 4 (demand exceeds capacity and customers are at risk). "Why this option": slides 5 and 6 (it pays back and the risk is managed). Slide 1 is the governing thought; slide 7 is the ask. (3) A line chart of quarterly utilisation and lead time, or a bar of utilisation with a reference line at a sustainable level, titled as on the slide; highlight the latest quarter. Link to mba/mba-queueing-card: at 94 percent utilisation waiting time is on the steep part of the curve, which is why lead times doubled. (4) Situation: Corvane has grown on one production line. Complication: demand now exceeds what the line can make, lead times have doubled and large customers are at risk. Question: how should we add capacity without over-committing if the market slows?

Exercise C. Answers vary. A strong governing thought: "Add a fourth physician from 10:00 to 14:00 at USD 0.29m a year: it cuts the average wait for a physician from 35 to 5 minutes and meets both clinical targets, which three physicians miss by a wide margin." A strong summary states the recommendation, the two targets and the results with 3 and 4 physicians, the cost, the main risk (beds downstream may become the bottleneck; other hours not yet analysed) and the ask with a date. A good chart: a bar or dot chart of the share waiting over 15 minutes with 3, 4 and 5 physicians (52.0, 13.0 and 2.9 percent) and a reference line at the 20 percent target, titled "Four physicians meet the waiting target; three do not". Likely questions: "Can we do it with a cheaper role?" and "What about other hours?", answered in the options and next-steps lines.

Self-test.

1. Put the main conclusion at the top, then support it with a small number of grouped key points, each supported by evidence below it. Each level summarises the one below and answers the question the level above raises.
2. Situation: what the reader already knows and accepts; complication: what changed or what is at stake; question: what the reader now wants answered, which your governing thought answers. The situation contains nothing new so that the reader agrees with it and is ready to accept the complication.
3. Example: "The warehouse robot market grew 18 percent last year, and we grew faster, at 31 percent".
4. Mutually exclusive, collectively exhaustive: no overlap between groups and nothing missing. Overlap double-counts evidence and confuses the reader about which reason carries the weight.
5. $206.835 - 1.015 \times 15 - 84.6 \times 1.5 = 64.7$. Excel: =206.835-1.015*(90/6)-84.6*(135/90).
6. Less so: a line chart encodes values by position, not by bar length, so a non-zero axis is acceptable when the change itself is the message. But the scale should not be chosen to exaggerate; label the axis clearly and give the change as a number (4 percent).
7. Position along a common scale. Angles and areas of seven slices are read inaccurately; a sorted bar chart shows the same data more accurately.
8. What is recommended (or concluded), why, how much it costs and returns, what the main risks are, and what is needed from the reader by when.
9. Examples: delivering bad news to people affected by it, a hostile audience, or a culture that expects context first. Know the answer, lead with shared facts and the problem, bring the reader to the conclusion, and still state the conclusion clearly and early enough.
10. Because the reader's first question is "what if the customers do not follow?". The negative NPV over three years shows the closure is not risk-free: the recommendation depends on keeping some customers, so the memo should state the retention needed and the actions to achieve it.

Revision history

Standard and licence note

Based on the published ideas of Minto (the pyramid principle), Flesch (1948), Tufte (1983), Cleveland and McGill (1984) and US federal plain language guidance, which are cited above. The text, the memos, the storyline and all data are original; Northfield Outdoor Co., Corvane Robotics and Tallow Creek Medical Center are fictional. This card is not affiliated with or endorsed by any business school, consulting firm, publisher or author. Check your course's or employer's house style, which may differ. Template licence: MIT.