Formal Reports at C1: Analysis, Caveats, and Decisions
C1 · Writing Lab · about 45 minutes
1 · Learn
The B2 report taught structure; C1 is about EPISTEMICS — how the report handles what it does not know. A senior report grades its own evidence (established / probable / contested), states the base rates behind its percentages, records the minority view in a footnote that a decision-maker cannot unsee, and separates three things junior reports merge: what we found, what we think it means, what we advise doing about it.
The signature register moves are calibrated: “The evidence supports…”, “On the data available…”, “We cannot exclude…”, “The working group was not unanimous; the minority held that…”. This is not timidity. It is the dialect of people whose recommendations carry budgets — they have learned what one overconfident report costs.
2 · See an Example
Executive Excerpt: Automation Audit, Meridian Logistics
FINDING 3 — The projected savings from warehouse automation rest on a volume assumption the sales plan itself does not endorse. The vendor's model ($2.1M annually from Year 2) assumes throughput growth of 14% a year. Finance's own forecast, in the same board pack, assumes 6%. Recomputed at the company's own numbers, the saving falls to approximately $0.9M and the payback period extends from three years to seven. We do not conclude that the automation is unjustified; we conclude that its justification, as currently documented, is borrowed from a future the company has not forecast.
CAVEAT — Utilization data cover eleven months and exclude the December peak, when the manual system's overtime — the very cost automation targets — is at its highest. The omission cuts in the project's favor for labor costs and against it for error rates; neither effect has been quantified.
DISSENT — The operations director's minority view is recorded at her request: that the audit prices labor but not fragility, and that the December 2023 manual recovery — when staff worked around a system outage for two days — has no automated equivalent at any modeled cost. The majority found the point real but unquantifiable; it is presented here so that the decision is taken with it in the room.
3 · Analyze the Example
- Internal inconsistency as the finding. The killer evidence is the company's OWN forecast contradicting the vendor's assumption — 14% vs 6% in the same board pack. Senior findings often need no new data, only juxtaposition.
- Conclusion split from observation. “We do not conclude X; we conclude Y” — the report polices the border between what the data shows and what anyone should do about it.
- Caveat with a direction. The missing December data is not a shrug: it cuts FOR the project on one variable and AGAINST it on another, and “neither has been quantified” marks the exact size of the unknown.
- Dissent given a chair, not a grave. The minority view is stated at its strongest, attributed, and its status (real but unquantifiable) ruled on openly. Boards trust reports that show them the losing argument.
Useful Vocabulary
One caveat: the data exclude December.
Throughput rose 14%.
The payback period extends to seven years.
Sales does not endorse the assumption.
The group was not unanimous.
Useful Expressions
- Recomputed at the company's own numbers, …
- We do not conclude X; we conclude Y.
- Neither effect has been quantified.
- … is recorded at her request: …
- … so that the decision is taken with it in the room.
Common Mistakes
✗ False certainty: “Automation will save $2.1M a year.” (inherited from the vendor's slide)
✓ “On the vendor's volume assumption, savings reach $2.1M; on Finance's forecast, approximately $0.9M.”
A senior report never lets a vendor's assumption wear the company's voice. Every inherited number keeps its owner's name attached.
✗ The buried caveat: a footnote in 8-point type on page 34 that reverses the headline finding.
✓ Caveats that change the decision travel WITH the finding — same section, same type size, with their direction stated.
Decision-makers skim by design. A report that hides its qualifications in the basement has, functionally, lied in the lobby.
✗ Manufactured consensus: “The working group recommends…” (when the group split 4–3)
✓ “The majority recommends…; the minority view, held by the operations director, is recorded below.”
Split decisions reported as unanimity detonate later — in the corridor, where you cannot answer. A recorded dissent is armor for the recommendation itself.
4 · Guided Practice
Recompute the headline: “The new system will cut processing time by 40%.” (pilot: 12 users, best week, vendor-supervised)
See a model answer
Example: “In a vendor-supervised pilot with twelve experienced users, processing time fell by up to 40% in the best week; the median improvement was 17%, and no unsupervised data yet exist.”
Write the caveat for: a satisfaction survey sent only to customers who completed a purchase.
See a model answer
Example: “CAVEAT — The survey reached only completed purchases. The dissatisfied-by-abandonment, probably the more informative group, are structurally absent from every figure that follows.”
Record the dissent in one sentence: the finance lead thinks the timeline, not the budget, is the real risk.
See a model answer
Example: “The finance lead dissents from the majority's focus on cost, holding that the timeline risk — a launch inside the audit season — is the one risk the budget cannot buy its way out of.”
5 · Write — Your Task
✍️ Write the core of an analytical report on a decision in your world (a purchase, a policy, a system change): one finding that juxtaposes two sources of evidence, one caveat with its direction and quantification status stated, one recorded dissent at full strength, and a recommendation that separates found / interpreted / advised into three explicitly labeled moves.
Goal: Finding + Caveat + Dissent + Recommendation · Length: 320–400 words
- Attach every inherited number to its owner (vendor, finance, the pilot).
- For each caveat ask: which way does it cut, and by how much — or mark it unquantified.
- Write the dissent so well that its holder would sign your version.
6 · Check — Self-Check
Work through your checklist honestly, then run the Self-Check. It counts words and sentences and spots mechanical points — it does not grade your English. You are the editor.
Read a model answer for the task
FINDING — The reading program's celebrated gains are real and mislocated. Test scores in participating schools rose 11 points over two years. Over the same period, scores in the matched non-participating schools — which received new libraries under a separate grant — rose 9. The program's marginal contribution, on the data available, is approximately two points and possibly zero; what the two interventions share, books and adult attention, is the likelier engine. CAVEAT — Matching was by district statistics, not randomization; unmeasured differences (a charismatic principal is not in our model) could account for the residual. DISSENT — The program's director records that the audit 'measures the shadow of reading, not reading': the two points may undervalue a habit whose returns arrive after the audit's window closes. The majority accepts the mechanism and notes it applies equally to the library grant. RECOMMENDATION — Continue the program at current funding; do not expand it until a randomized phase-in, which the calendar permits next year, separates its effect from the libraries'.
7 · Improve & Rewrite
Weak version — rewrite it in your notebook first, then compare:
Our investigation shows the new software is excellent and will save a lot of money. Everyone on the team agrees it should be bought immediately. There are no problems. The vendor says companies save 50% and we expect the same. This report recommends purchase.
See an improved version
FINDING — The software's case rests on a single vendor figure: a 50% reduction in processing time, observed at two reference sites, both larger than us and both named by the vendor. Our own two-week pilot, with nine users, produced a median improvement of 18% — respectable, and roughly a third of the headline. INTERPRETATION — The gap is most plausibly explained by selection: vendors exhibit their best rooms. CAVEAT — Our pilot ran in the quietest fortnight of the year; peak-season performance is unmeasured in either direction. POSITION OF THE TEAM — The team was not unanimous (5–2). The minority would purchase now, arguing the 18% alone clears the cost within two years. RECOMMENDATION — The majority recommends a paid peak-season trial with pre-agreed success metrics, the metrics — not the brochure — to decide the purchase in January.
Quick Quiz
Keep going
📘 Grammar deep-dive: The Subjunctive & Formal Patterns🌱 Vocabulary Builder: the B1 Money Business set🎓 TOEFL Writing: the exam version of this skill← Previous lesson · C1 — all lessons · Next: Persuasive Writing and Rhetoric: Ethos, Pathos, Logos — and Timing →
Ad space