Umiejętności pisania zawodowego — model answers
Język angielski w zastosowaniu zawodowym
How to use this document
These are model answers for the three in-class exercises (memo, proposal, report). They are not the only correct answers — a different group could write a different memo that scored just as highly. They show one version that satisfies every requirement and explain why particular choices were made.
For each exercise, this document:
- Reproduces the brief so the requirements are visible alongside the answer.
- Builds the document section by section with annotated asides at each step.
- Highlights vocabulary and grammar choices in bold in the text and explains them in margin-style notes.
- Closes with a short post-mortem identifying the easiest places to lose marks.
The annotations use the lesson notation from the course:
- L1 — globalisation, present perfect, first conditional
- L2 — strategy and management vocabulary, sentence adverbials, cause-and-effect
- L3 — ethics and leadership vocabulary, modal verbs
- L4 — finance / people / innovation vocabulary, passive voice, reported speech
Read the model answer first, then re-read it with the annotations. The point is not to memorise the wording: it is to see which features of Lessons 1–4 do which work in each genre.
Exercise 1 — Compliance memo
The brief (reminder)
You are the Chief Compliance Officer. Inform all department heads about new whistleblower procedures and code-of-ethics updates effective 1 May 2026. Under 250 words. Required language:
- At least 5 vocabulary items from L1–L4 (drawn from the ethics/compliance set).
- At least three different modal verbs (one obligation, one advice, one possibility).
- At least two sentence adverbials.
Step-by-step build
Step 1 — the header
Write the header before anything else. It is the easiest part of the memo, and getting it right means you do not have to come back to it under time pressure.
MEMORANDUM
To: All Department Heads From: Anna Nowak, Chief Compliance Officer Date: 18 April 2026 Subject: New whistleblower platform and revised code of ethics, effective 1 May 2026
- The subject line is specific and dated. “Policy update” would be a fail. The subject must let the reader decide in three seconds whether to act now or later.
- “All Department Heads” (not “All Staff”) signals correctly that action lives at the department-head level — they are the ones who must cascade the changes.
- The date is the date the memo is sent, not the date the policy takes effect. The policy date belongs in the subject.
Step 2 — the opening (1–2 sentences)
The opening must do two things in two sentences: name the context (what changed and why), then name the purpose (what this memo does).
Following the Executive Committee’s review of the Q1 risk assessment, the Board has approved a revised code of ethics and updated whistleblower procedures, effective 1 May 2026. This memo sets out the changes, your responsibilities as department heads, and the deadlines for cascade and acknowledgement.
- “has approved” — present perfect (L1). Recent action with continuing relevance is exactly what the present perfect is for. Past simple (“approved”) would feel disconnected from the present situation; the present perfect signals “this just happened, and it matters now.”
- “risk assessment”, “code of ethics”, “whistleblower” (L3). Three of your five vocabulary items in the first sentence. They are doing real semantic work — not jargon for its own sake.
- The opening explicitly previews the structure of the memo (“changes, responsibilities, deadlines”). Department heads now know what they are reading.
Step 3 — the body
Use headings and bullet points. Memos are scanned, not read.
What is changing
Consequently, from 1 May 2026:
- All staff must complete a 60-minute online compliance training module.
- All staff must disclose any potential conflict of interest through the new digital declaration form.
- A confidential whistleblower platform (internal portal plus external independent hotline) will replace the current ethics inbox.
- Due diligence requirements for new external partnerships have been strengthened.
- “Consequently” — sentence adverbial (L2, connective). Signals that what follows is the logical consequence of the opening. One word, but it does the work of a connecting clause.
- “must” used twice — modal verb of obligation (L3). Both are non-negotiable institutional requirements. Two musts in a row is fine here because the items are parallel and the obligation is identical.
- “will replace” — modal of certainty. Not advice, not possibility: the platform replacement is a decided fact.
- “have been strengthened” — passive voice (L4). The agent (the Compliance Office) is implicit and irrelevant; the change itself is what matters. Active voice (“We have strengthened…”) would shift attention to the actor and weaken the institutional tone.
Your responsibilities
- Cascade this memo to all team members by 22 April 2026.
- Ensure at least 90% training completion within your department by 15 May.
- Return the signed acknowledgement form to HR by 22 April.
- Importantly, you should flag any capacity or resource concerns to my office before 25 April, so that they can be addressed in advance.
- “Importantly” — sentence adverbial (L2, focus). Marks the one item that is qualitatively different from the others. The first three items are mechanical (cascade, train, sign); the fourth requires judgement.
- “should” — modal of advice (L3). The expectation is real, but the action requires the department head’s discretion. Must would imply that everyone has problems to flag, which would be insulting; should leaves the choice with the reader.
- Numbers and deadlines everywhere. A memo without dates and percentages is a memo without teeth.
Why this matters
These changes reflect our continuing commitment to integrity, transparency, and accountability — values that our stakeholders, clients, and regulators increasingly expect us to demonstrate. Furthermore, robust compliance infrastructure may itself become a competitive differentiator in client due diligence reviews.
- The “why this matters” section is optional but valuable in change-management memos. It pre-empts the reader’s “why now? why this?” question.
- “integrity”, “transparency”, “accountability”, “stakeholders”, “due diligence” (L3 + L2). That is five vocabulary items in two sentences. The requirement of “at least 5” is comfortably met; the count is now well over.
- “Furthermore” — sentence adverbial (L2, additive). The second sentence adverbial. Adds a second reason without sounding like a list.
- “may” — modal of possibility (L3). Completes the obligation/advice/possibility triad. Note that may lives in the “why this matters” paragraph, not in the “what you must do” paragraph — possibility is the right register for forward-looking framing, not for instructions.
Step 4 — the closing
For questions about the training platform, please contact IT (
ithelp@firma.pl). For questions about the code of ethics or whistleblower procedures, please contact the Compliance Office (compliance@firma.pl).
- No flowery sign-off. Memos do not use “Kind regards” or “Best wishes” — those are email conventions. The named contacts at the end are the equivalent.
- Two contacts, two different topics. Specific, actionable, friction-free. The reader should never have to guess where to direct a question.
Word count and language audit
The full memo body runs to ~225 words — under the 250-word ceiling. Let us audit against the requirements:
| Requirement | Met? | Where |
|---|---|---|
| ≥ 5 L1–L4 vocabulary items | ✅ (10+) | risk assessment, code of ethics, whistleblower, compliance, conflict of interest, due diligence, integrity, transparency, accountability, stakeholders |
| Modal of obligation | ✅ | must complete, must disclose |
| Modal of advice | ✅ | should flag |
| Modal of possibility | ✅ | may itself become |
| ≥ 2 sentence adverbials | ✅ (3) | Consequently, Importantly, Furthermore |
| Active voice predominant | ✅ | One deliberate passive: have been strengthened |
| Present perfect (L1) | ✅ (bonus) | has approved |
Common ways to lose marks on this exercise
- Vague subject line. “Compliance update” fails. Always include the policy and effective date.
- All four modals at the same strength. A memo that uses must for everything reads as authoritarian; one that uses should for everything reads as toothless. Mix deliberately.
- Sentence adverbials clumped at the start of every sentence. Importantly, … Furthermore, … Consequently, … in three consecutive sentences is overuse. Three across the whole memo is right.
- Importing email greetings. “Dear colleagues, … Kind regards, Anna” is wrong for a memo. The header replaces the greeting; the contact lines replace the sign-off.
- Calls to action without deadlines. “Please cascade this in due course” is a non-instruction.
Exercise 2 — NordicPrime market-entry proposal
The brief (reminder)
NordicPrime Retail (Polish home-furnishings retailer, €142m revenue, 6.8% domestic market share) wants to enter the Czech and Slovak markets. You are Crossroads Consulting Partners. The executive summary is provided. You write four sections only: problem statement, proposed solution, key benefits, deliverables. Required language:
- At least 8 terms from L1–L4 (across globalisation, strategy, finance, innovation).
- At least one first conditional projecting an outcome.
- At least three sentence adverbials.
- Active voice predominant.
Step-by-step build
Section 1 — problem statement
The trap in problem statements is to describe the opportunity rather than the problem. The client already knows about the opportunity; what they need to see is that you understand the constraint that makes the opportunity hard to capture.
NordicPrime’s market share in Poland has grown from 4.3% to 6.8% in three years, but the Polish home-furnishings market is now maturing: category growth has slowed from 6% (2022) to 2% (2025), and revenue growth at NordicPrime has decelerated correspondingly. Consequently, sustaining the trajectory that NordicPrime’s investors expect will require diversification beyond the domestic market. The Czech and Slovak markets, both inside the same trade bloc as Poland, offer 4–6% category growth and a shared supply chain infrastructure — but two established competitors (one Swedish, one German) already operate at scale across both markets, and recent due diligence has flagged outsourcing and labour-cost risks. Without a structured entry, NordicPrime risks paying the cost of late entry while a third competitor (rumoured to be a Hungarian incumbent) consolidates the mid-price segment.
- Open with the client’s own numbers. Market share moving from 4.3% to 6.8% is NordicPrime’s growth; the slowdown from 6% to 2% is the category’s. That gap between firm-level and category-level growth is the problem in two numbers.
- “market share”, “revenue”, “diversification”, “trade bloc”, “supply chain”, “due diligence”, “outsourcing” (L1, L2, L4). Seven vocabulary items in one paragraph — you have already met most of the 8-term requirement.
- “Consequently” — sentence adverbial (L2). Connects the problem (slowing domestic growth) to the implication (need for international diversification). Without the adverbial, the two sentences would read as separate observations.
- “has grown”, “has slowed”, “has decelerated” — present perfect (L1). Trends with continuing relevance. Past simple would freeze the events in the past and lose their connection to the current decision.
- The final sentence names what is at stake. The third competitor is the cost of inaction. Naming that competitor (even speculatively) makes the urgency tangible.
- Active voice throughout. “NordicPrime risks paying” not “the cost is risked by NordicPrime”. Proposals are not the place for passives.
Section 2 — proposed solution
The reader should be able to visualise what happens in each phase of the engagement. Three phases of three weeks each are easier to absorb than one ten-week monolith.
Our ten-week study comprises three phases:
- Diagnostic phase (Weeks 1–3): comparative assessment of the two markets on regulatory regime, tariff and non-tariff barriers, consumer-spending data, and competitive density. Twenty in-market interviews with retailers, distributors, and category buyers.
- Benchmarking phase (Weeks 4–6): evaluation of three entry modes — joint venture with an established local distributor, franchising, or greenfield subsidiary — with due diligence profiles of three candidate partners per chosen country.
- Roadmap phase (Weeks 7–10): prioritised 36-month implementation plan, first-mover advantage assessment relative to the unannounced Hungarian entrant, scalability stress-test of the chosen operating model, and IP-protection review.
- Phases as bullets, not paragraphs. The reader can scan and count. Three roughly equal phases also signal that the engagement is managed, not open-ended.
- Each phase ends with a quantifier or a count. “Twenty interviews”, “three candidate partners per country”, “36-month plan”. Vague phases lose contracts.
- “tariff”, “joint venture”, “due diligence”, “first-mover advantage”, “scalability” (L1, L2, L4). Five additional vocabulary items, deployed where they fit the analysis. By the end of the proposal you will be well over the 8-term floor.
- No modal verbs of advice or possibility here. The proposal commits to delivering each phase — will (implicit) is the only acceptable modal.
Section 3 — key benefits (with the first conditional)
Based on comparable engagements in the CEE retail sector, NordicPrime can expect:
- Strategically, an entry sequence that captures the underserved Czech mid-price segment first, with Slovakia following in Year 2.
- If NordicPrime enters the Czech market by Q3 2026, it will secure first-mover advantage over the Hungarian competitor and lock in a 12-to-18-month head start in distributor relationships.
- A 2–4% target-market share within 24 months of entry, generating €18–28 million in incremental revenue.
- A 25–35% improvement in working-capital turnover on the international book through a more efficient supply chain.
- An estimated five-year ROI of 2.6–3.2×, compared with 1.8× for an equivalent domestic capacity expansion.
- “Strategically” — sentence adverbial (L2, strategic signpost). A single word that signals “this is the headline”. It also lets the bullet that follows be a noun phrase rather than a full sentence.
- The first conditional in bold (L1). “If NordicPrime enters the Czech market by Q3 2026, it will secure first-mover advantage…” This is the structural centrepiece of the benefits section — a projected outcome the client can act on. Note the specific date and specific competitor: vague conditionals (“if you choose us, you will succeed”) are worthless.
- “first-mover advantage”, “revenue”, “ROI” (L2, L4). Three more vocabulary items.
- Ranges, not point estimates. 2–4%, €18–28 million, 2.6–3.2×. Ranges signal honest uncertainty without undermining conviction — the difference between credible hedging and weak-kneed hedging.
- The final bullet uses comparison. Quoting the ROI of an alternative use of the same capital is one of the most persuasive moves available. It pre-empts the “what else could we do with this money?” question.
Section 4 — deliverables
Deliverables are tangible: a report, a workshop, a list. Outcomes (above) are projections; deliverables are objects.
On completion, NordicPrime will receive:
- A written diagnostic report identifying the recommended country sequence and entry mode.
- A 36-month implementation roadmap with milestones, resource requirements, and investment phasing.
- A shortlist of three qualified local partners per country (distributors and franchisees), with preliminary due-diligence dossiers.
- A half-day strategy workshop with NordicPrime’s Executive Committee to present findings, stress-test assumptions, and agree the Q3 entry decision.
- Importantly, all deliverables are written in English and Polish; the diligence dossiers are also available in Czech and Slovak.
- Each deliverable is a thing, not a process. “A report”, “a roadmap”, “a shortlist”, “a workshop”. The client can imagine receiving each one.
- “Importantly” — sentence adverbial (L2, focus). Used here to surface a logistical detail that would otherwise be buried but matters for stakeholder management. Three sentence adverbials across the four sections (Consequently, Strategically, Importantly) — requirement met.
- The bilingual / multilingual line is a credibility move. It demonstrates that you understand the client operates across language boundaries.
Language audit
| Requirement | Met? | Where |
|---|---|---|
| ≥ 8 L1–L4 vocabulary items | ✅ (15+) | market share, revenue, diversification, trade bloc, supply chain, due diligence, outsourcing, tariff, joint venture, first-mover advantage, scalability, ROI, working capital, IP, competitive advantage |
| ≥ 1 first conditional | ✅ | “If NordicPrime enters the Czech market by Q3 2026, it will secure first-mover advantage…” |
| ≥ 3 sentence adverbials | ✅ (3) | Consequently, Strategically, Importantly |
| Active voice predominant | ✅ | All commitments active; no agentless passives |
Common ways to lose marks on this exercise
- Hedging in the executive summary or benefits. “We hope to deliver…”, “It is possible that…”, “We will try to…”. Hedging belongs in reports, not proposals. Use ranges to convey honest uncertainty.
- Generic consultancy language. “A leading provider of innovative solutions”, “world-class expertise”, “bespoke methodology”. None of these names a thing the client receives.
- Unquantified benefits. “Significant improvement in efficiency” fails. “A 25–35% improvement in working-capital turnover” succeeds.
- Phases without deliverables. A phase that does not end with a tangible output is a phase that does not exist.
- Overusing the same modal. Will, will, will, will… In benefits sections, mix will (commitment) with conditional projections (if X, then Y will…) to avoid monotony.
Exercise 3 — GlobalTech retention and innovation report
The brief (reminder)
GlobalTech Innovations Sp. z o.o., 470-employee software firm, has a combined retention and innovation crisis. The data appear as three charts (people, innovation, finance) in the exercises. Your group writes:
- The executive summary (3–5 sentences, self-contained).
- At least two findings paragraphs, each describing one of the three charts, with at least one passive-voice sentence and one reported-speech sentence.
- The three recommendations, each with a different modal verb and a one-line justification.
Required language:
- At least 10 vocabulary items across people (L4), innovation (L4), finance (L4), and strategy (L2).
- Mixed modal verbs in the recommendations.
- Passive voice in findings where the agent is implicit.
- Reported speech when summarising stakeholder inputs.
- Sentence adverbials to mark logical structure.
- Calibrated hedging — present where warranted, absent where not.
Step-by-step build
Step A — the executive summary
The executive summary must be self-contained: a board member who reads only this paragraph must know what is wrong, why it matters, and what is being recommended.
Engineer churn at GlobalTech has risen from 13% to 28% over two years, while the industry median has held at 13%. Onboarding satisfaction has dropped from 62% to 39%, burnout is rated “severe” in four of seven product teams, and time-to-market on the last three product releases has increased by 85% against the 2022 baseline. Consequently, revenue growth has decelerated from 14% to 3% and profit margin has contracted from 12% to 7%. The data suggest that the retention problem is internal and structural — not market-driven — and that without intervention in the current quarter the recovery horizon is likely to extend beyond eighteen months. The board must approve a revised retention strategy within four weeks; the firm should restructure onboarding around a 90-day competency framework; and a four-day-week pilot in one product team could be used to test burnout mitigation at low cost.
- Six instances of present perfect (L1): has risen, has held, has dropped, has decelerated, has contracted, has increased. The trends are recent, ongoing, and connect to a present decision — exactly the present perfect’s territory.
- “is rated” — passive voice (L4). The agent (the engagement-survey instrument) is irrelevant.
- “The data suggest that…”, “is likely to” — hedging. Calibrated, not reflexive: hedging in two specific places (the causal claim and the time-horizon claim), not everywhere.
- “Consequently” — sentence adverbial (L2). Connects the people/innovation findings to the financial consequences.
- Three modals at three strengths in the closing sentence: must (board action — non-negotiable), should (operational restructure — the recommended path), could (experimental pilot — one option among several). This is the calibrated mixing the recommendations section will expand on.
- Vocabulary count so far: churn, retention, onboarding, burnout, time-to-market, revenue, profit margin, competency framework, productivity — nine items in one paragraph.
Step B — findings paragraph for the people chart
Engineer churn has diverged sharply from the industry benchmark. Between 2022 and 2024, GlobalTech’s engineer churn rose from 13% to 28%, while the industry average was observed to remain flat at 13% across the same period. The gap of 15 percentage points in 2024 is more than twice the largest single-year jump recorded for any sector peer in the Polish Software Industry Association’s benchmarking dataset. The Head of People acknowledged in her February briefing that onboarding resources had been reduced by approximately 40% during the 2024 cost-cutting round, and that upskilling budgets had not been revised since 2022 in nominal terms. The data suggest that post-onboarding retention is the part of the system in which intervention is most likely to be effective.
- Mini-heading as a claim, not a label. “Engineer churn has diverged sharply from the industry benchmark” is the finding in one sentence; “Engineer churn” alone would not be.
- Passive-voice findings (L4): “was observed to remain flat”, “had been reduced”, “had not been revised”. The agent (the survey, the management) is implicit and unimportant; the change is what matters.
- Reported speech (L4): “The Head of People acknowledged… that onboarding resources had been reduced…” Note the tense backshift: was reduced → had been reduced. The reporting verb acknowledged is doing semantic work — it signals that this is a concession, not a neutral statement.
- “The data suggest that…” — hedging. Calibrated: the causal claim is hedged; the descriptive claim about the 15-point gap is not.
- The chart is described in the first two sentences, then interpreted. Description first, interpretation second — never blur the two.
Step C — findings paragraph for the innovation chart (alternative)
A second findings paragraph, on the innovation indicators chart, demonstrates the same pattern with different vocabulary.
Innovation capacity is weakening across two simultaneous dimensions. Time-to-market on GlobalTech’s three most recent product releases has increased by 85% against the 2022 baseline, while R&D spend has held at 14% of revenue against an industry median of 18%. Two competitors have launched disruptive products in GlobalTech’s core segment since 2024, both within twelve months of their announcements — well below GlobalTech’s current release cycle. Senior product leads have stated that the slowdown reflects engineer departures and the absence of senior reviewers in the affected teams, rather than any change in product strategy. Taken together, the people findings and the innovation findings appear to be causally linked: the retention problem is becoming an innovation problem.
- The mini-heading announces the structural claim — two simultaneous dimensions, not one.
- Reported speech (L4): “Senior product leads have stated that…” Reporting verb stated (neutral) chosen over acknowledged (concession): there is no concession here; the leads are simply describing the situation.
- Hedging: “appear to be causally linked” — the causal claim is honest about being interpretive, not proven.
- Vocabulary used: R&D, time-to-market, disruptive, retention, innovation. Five more, complementing the people-paragraph vocabulary.
- Cross-section synthesis at the end. “The retention problem is becoming an innovation problem” is the thread that pulls findings together for the conclusions section. Each findings paragraph should hand the next one a baton.
Step D — the recommendations (mixed modals)
The board must approve a revised retention strategy within four weeks. The current churn trajectory directly threatens the 2026 product roadmap and therefore Q4 revenue guidance. Delay compounds the cost of every subsequent quarter.
GlobalTech should restructure onboarding around a 90-day competency framework, with named mentors and explicit milestones. Onboarding satisfaction is the single largest driver of first-year departures in the exit-interview data, and a structured framework can be implemented within one quarter using existing HR resources.
A four-day working week could be piloted in one product team for six months to test whether reduced workload mitigates burnout without compromising delivery. The intervention carries implementation risk and may not generalise, but a small-scale pilot would generate evidence at low cost; if outcomes are productivity-neutral or better, the approach ought to be considered for wider rollout.
- Three different modals at three different strengths (the core requirement of L3 in this exercise):
- must — non-negotiable institutional action; framed as required for business survival.
- should — the recommended approach, with feasibility named explicitly.
- could + ought to — experimental, low-cost, contingent. Could opens the door; ought to is a stronger conditional follow-up if the pilot succeeds.
- Each recommendation has a heading-style action sentence, then a one- or two-sentence justification. The action is owned (the board, GlobalTech, a pilot team); no recommendation hangs in passive limbo.
- Implicit first conditional in Recommendation 3 (L1): “if outcomes are productivity-neutral or better, the approach ought to be considered…” This is the kind of forward-looking projection the conclusions in a report are entitled to use.
- Passive in Recommendation 3: “could be piloted”, “the approach ought to be considered”. The agent (the management team that runs the pilot) is implicit; the action is what matters.
Language audit
| Requirement | Met? | Where |
|---|---|---|
| ≥ 10 L1–L4 vocabulary items | ✅ (15+) | churn, retention, onboarding, burnout, upskilling, performance review, talent pipeline, R&D, time-to-market, disruptive, scalability, revenue, profit margin, working capital, ROI |
| ≥ 1 passive-voice sentence per findings paragraph | ✅ | was observed to remain flat, had been reduced, had not been revised; could be piloted |
| ≥ 1 reported-speech sentence per findings paragraph | ✅ | Head of People acknowledged that…, Senior product leads have stated that… |
| Mixed modals in recommendations | ✅ (3) | must, should, could / ought to |
| Sentence adverbials | ✅ | Consequently, Taken together (and others if extending the report) |
| Calibrated hedging | ✅ | The data suggest that…, is likely to…, appear to be causally linked |
Common ways to lose marks on this exercise
- Reading the chart instead of describing it. “The chart shows engineer churn” is not a finding. The finding is what the data say and what it means.
- Hedging every sentence. A report that hedges everywhere reads as if the analyst is unwilling to take a position. Hedge causal and predictive claims; do not hedge factual descriptions.
- All recommendations at the same strength. Three musts reads as authoritarian; three shoulds reads as advisory mush. Mix deliberately, and let the strength of the modal match the strength of the evidence and the urgency of the action.
- Reported speech without backshift. “The CFO acknowledged that revenue is declining” is wrong; “The CFO acknowledged that revenue had been declining” (or was declining) is right.
- Findings and conclusions blurred. Findings present data; conclusions interpret it. If your findings paragraph already contains “this means that…”, you have moved into conclusions territory — that text belongs in the next section.
- Recommendations without ownership. “Onboarding should be restructured” (passive) loses the mark. “GlobalTech should restructure onboarding” (active) keeps it.
Cross-cutting summary
Across the three exercises, the same Lessons 1–4 features are doing different work in each genre:
| Feature | Memo | Proposal | Report |
|---|---|---|---|
| Modal verbs (L3) | Calibrate the strength of instructions to the recipient | Almost always will (commitment); avoid hedged modals | Calibrate the strength of recommendations to evidence |
| Sentence adverbials (L2) | Mark the structure of the memo for scanning | Signal logical movement and emphasis | Connect findings to conclusions |
| Passive voice (L4) | Used sparingly, for institutional policy | Avoided | Used in findings; avoided in recommendations |
| Hedging | Avoided | Avoided | Essential, but calibrated |
| Reported speech (L4) | Rare | Rare | Used to summarise stakeholder inputs |
| Present perfect (L1) | Recent action with continuing relevance | Track record and trends | Trends in findings; conclusions |
| First conditional (L1) | Rare | Projected outcomes for the client | Projected costs of inaction; conditional recommendations |
The same vocabulary item (e.g. due diligence) can appear in all three documents but does different work in each: a fact in the memo, a deliverable in the proposal, a finding in the report. Genre, not vocabulary, decides what each word is doing.