Claude Sonnet 5 on “A set of monthly reports”: with Minto, worse on structure, quality unchanged
Structure 6 → 4 of 8 −2 Quality 7 → 7 of 10 0
The judge on the output with Minto
It handles missing versus buried logic well and uses levels sensibly, but promotes resourcing as a key branch and misses the two failure-point classes.
| Structure | Without Minto 6 | With Minto 4 | Change |
|---|---|---|---|
| Top | 1 | 1 | |
| Key-line composition | 1 | 1 | |
| Levels | 2 | 1 | −1 |
| Order and kind | 2 | 1 | −1 |
| Quality | Without Minto 7 | With Minto 7 | Change |
|---|---|---|---|
| Top | 1 | 1 | |
| Same-kind grouping | 2 | 2 | |
| Explainable order | 2 | 1 | −1 |
| MECE | 1 | 1 | |
| Visible, proportionate structure | 1 | 2 | +1 |
Outputs and the judge’s words are quoted as written, in English.
Without Minto
control prompt · blind labelout-10No hard failures
Its evidence is well nested and its sequence is coherent, but it elevates the diagnosis and capacity constraint as though they constituted the missing answer.
Audit: Period Graph Books memo
Overall diagnosis
The memo has no governing thought. It opens with a factual statement about a departmental handover, then lists production steps, then issues, then an "overall assessment" that actually contains the real point — but that point arrives last, buried, and is followed by a weak, non-committal question rather than a request tied to an action.
Defect 1: No pyramid — narrative order, not logical order
The structure follows the order events happened (transfer → production steps → issues → assessment), not an order derived from a single top point supported by grouped reasons. A reader has to reach the final paragraph to learn what the memo is actually arguing: that the process is fragmented and needs restructuring, and that the current team can't do it alone. That should be the opening sentence, not the conclusion of a chronological account.
Defect 2: The real point is stated as an aside, not the lead
"The production process badly needs streamlining... The current team cannot manage the required changes as well as ongoing production" is the actual message and the actual ask (implicitly: give us dedicated resources/authority to fix this). It's introduced under a throwaway heading ("Overall assessment") after two sections of description. Minto's structure requires this to be the top box, with everything above it existing only to support it.
Defect 3: Situation–Complication–Question–Answer is missing or malformed
- Situation: weakly present (Financial Analysis took over four books since Period 5).
- Complication: present but scattered — "issues appeared" is mentioned up top, then re-explained later.
- Question: never made explicit. The reader isn't told what question the memo answers.
- Answer: implied only in the final line as a question back to the reader ("Could we discuss how best to proceed?"), which reverses the responsibility — the memo should answer its own question with a recommendation, not offload it.
Defect 4: "Production" section is description, not argument
The four numbered production steps are procedural detail with no clear grouping logic and no stated relevance to the argument. Under Minto, subordinate groupings must answer "why" or "how" for the point above them. Here they just narrate a workflow. If retained, they need to be recast as evidence for a specific claim (e.g., "the process is fragmented" — supported by showing triple entry, multiple departments, manual steps) rather than a standalone procedural inventory.
Defect 5: "Issues" section conflates categories of problems
Data inconsistency, calculation errors, unexplained changes, and re-entry burden are all listed at the same level without being grouped by cause or type. Some are process defects (fragmentation, multiple entry points), others are consequences (bad numbers reaching management). Minto requires groupings to be logically distinct (MECE) and summarizable by a single sentence — currently they're just an enumerated list of symptoms with no organizing idea.
Defect 6: One example doing the work of a general claim
The detailed walk-through of "one book" (loaded into a corporate program, transferred, re-entered, returned) is used as if it illustrates the whole system, but it's presented as an isolated anecdote rather than explicitly framed as representative evidence for the top-level claim of fragmentation. As written, it reads as a digression rather than a supporting point.
Defect 7: Closing line undermines the argument's purpose
"Could we discuss how best to proceed?" signals the memo hasn't actually reached a conclusion — it's an information dump asking the reader to do the synthesis. If the top point is "the process needs restructuring and the current team can't do both fix and run it," the memo should end with a specific recommendation or request (e.g., staffing, timeline, decision needed) that flows logically from that point, not an open-ended invitation to talk.
What must change
- State the governing thought first: process is broken due to fragmentation/lack of ownership, current team can't fix it while also running it, therefore [specific ask] is needed.
- Rebuild S-C-Q-A explicitly: Situation (handover since Period 5), Complication (errors emerged), Question (what should be done), Answer (the governing thought above).
- Regroup supporting detail into 2–3 MECE categories (e.g., "multiple entry points create inconsistency," "no single owner catches errors," "team is stretched between fixing and running") each with a one-line summarizing point, not a flat list of symptoms or a chronological workflow description.
- Convert the "one book" example into explicit supporting evidence for one of those groupings, not a stand-alone narrative.
- Replace the closing question with a specific recommendation or decision request that the governing thought logically demands.
With Minto
prompt with the skill · blind labelout-15No hard failures
It handles missing versus buried logic well and uses levels sensibly, but promotes resourcing as a key branch and misses the two failure-point classes.
Legend: ▲ top/answer · ● first-level group · ○ support · ⚠ mixed kinds/false grouping · ↑ buried conclusion · ⇄ wrong order · ⊗ overlap · ⊕ gap · ✂ cut/appendix
Subject: Period Graph Books
Since Period 5, Financial Analysis has been responsible for producing four graph-presentation books previously handled by Corporate Planning. This memo describes issues that appeared after the transfer. [Situation given; Complication only implied — no line states what specifically changed for the worse, or how badly, before the Question is asked]
Production [⚠ topic label, not a claim; this section answers "how does the process work," not the reader's question]
- Gather data from divisional forms, internal documents, and telephone calls.
- Calculate graph points manually or on the corporate computer…
- Transcribe the points to input sheets…
- Check the data and calculations.
[○ time-ordered, internally consistent — but sits at the top level with no payoff for the reader]
Issues [⚠ mixed kinds: general causes of fragmentation + one book's specific symptoms, presented as if same kind]
The transfer inserted another person into an already fragmented process. Divisional staff and three corporate departments participate, and the same information may be entered into computer systems three times. [○ general cause]
For one book, divisional data are loaded into a dedicated corporate program… [⊗ this is evidence for the general cause above, not a separate category — currently reads as a sibling, not a support]. Problems include inconsistent data between periods and regions, incorrect variable-cost calculations, unexplained changes to points that were previously correct, and prior-period data that must be re-entered.
Overall assessment
Most problems arise from a cumbersome fragmented process. No one controls the data from beginning to end, responsibility is unclear, and errors can fall between stages. The production process badly needs streamlining… [↑ this is the real answer to "what's wrong," buried three sections down instead of leading]. The current team cannot manage the required changes as well as ongoing production. [○ this is actually the seed of a real second branch — a staffing/resourcing constraint — currently demoted to a passing remark]
Could we discuss how best to proceed? [▲ this is where the top should be, and it isn't an answer — it's a request for a conversation. ⊕ management explicitly asked for corrective proposals; none are given anywhere in the memo]
Findings
| Location | Violation | Fix |
|---|---|---|
| Closing line "Could we discuss…" | Top does not answer the reader's question — ends in a question instead of a recommendation | Open with the recommended corrective action as the first line |
| Whole memo | ⊕ Gap — management asked for "corrective proposals"; none exist in the document | Add a group of proposed actions (the kind the reader's question demands) |
| "Overall assessment" | ↑ Buried conclusion — the real diagnosis ("cumbersome fragmented process, no end-to-end owner") sits at the end, three sections after the reader meets the symptoms | Promote it to the top, before or replacing the current opening |
| "Issues" section | ⊗ False grouping — a general cause (fragmentation, triple entry) and one book's specific symptoms are presented as parallel, same-level items | Make the general causes the first-level group; move the one-book detail underneath as its supporting evidence |
| "Production" section | Section answers "how does the process work" (mechanics), not "why does it fail" or "what should we do" — no payoff for the reader | Cut to an appendix, or keep only the steps that directly support a named failure point |
| "current team cannot manage the required changes" | A real second branch (resourcing/capacity constraint) is present but demoted to a closing aside instead of standing as its own group | Promote to a first-level group alongside the process-fragmentation cause |
| Intro paragraph | Situation is stated but the Complication (what specifically broke, and how badly) is never made explicit before the memo reaches its "question" | State the complication in one line (e.g., specific recurring error types since the transfer) so the reader reaches the question already asking it |
Score
| Axis | Score | Note |
|---|---|---|
| Top | 0 | No answer anywhere; memo ends in a question, not a recommendation |
| Same-kind groups | 1 | Production is internally consistent (steps); Issues mixes cause and evidence |
| Order | 1 | Time order present in Production only; Issues and Overall assessment are unordered |
| MECE | 0 | Gap: no corrective proposals despite being requested; overlap: book-specific detail duplicates the general cause |
| Display | 1 | Headings exist, but no key line and the real conclusion is not where the eye lands |
Total: 3/10 — not structured yet
Three fixes that buy the most
- Replace the closing question with a stated recommendation as the top line — this alone would move the memo from "not structured" toward baseline.
- Split "Issues" into causes (first level) with the one-book detail demoted to supporting evidence underneath.
- Add the missing corrective-actions group — the memo currently has no answer to the question it was commissioned to answer.
How the test was built
Claude Sonnet 5 received the same task twice. The control prompt is one framing line and the task. The prompt with Minto is the same task preceded by the full text of the skill. The judge read both outputs among 16 under blind labels and did not know which one had the skill.
Prompt without Minto 2 424 B
You are working on a business document using the Minto Pyramid Principle.
<!-- Faithful paraphrase of Minto (2010), Appendix B, Exhibits B-5 and B-6.
Original wording for this fixture. -->
# Fixture 03: Period Graph Books
**Mode:** `audit`
**Language:** `en`
## Context
Corporate Financial Analysis recently took over production of four monthly
graph books used in presentations to senior management. Management knows that
errors and inconsistencies have appeared and asked for corrective proposals.
## Before
```text
Subject: Period Graph Books
Since Period 5, Financial Analysis has been responsible for producing four
graph-presentation books previously handled by Corporate Planning. This memo
describes issues that appeared after the transfer.
Production
1. Gather data from divisional forms, internal documents, and telephone calls.
2. Calculate graph points manually or on the corporate computer, including
rolling values for revenue, cost, and percentages.
3. Transcribe the points to input sheets. Analysts update the current period,
then Color Graphics re-enters the points into its own database and generates
the graphs.
4. Check the data and calculations.
Issues
The transfer inserted another person into an already fragmented process.
Divisional staff and three corporate departments participate, and the same
information may be entered into computer systems three times.
For one book, divisional data are loaded into a dedicated corporate program,
calculated, transferred to Color Graphics input sheets, entered again, and
returned for updates. Problems include inconsistent data between periods and
regions, incorrect variable-cost calculations, unexplained changes to points
that were previously correct, and prior-period data that must be re-entered.
Overall assessment
Most problems arise from a cumbersome fragmented process. No one controls the
data from beginning to end, responsibility is unclear, and errors can fall
between stages. The production process badly needs streamlining, both for each
book and for common calculations. The current team cannot manage the required
changes as well as ongoing production.
Could we discuss how best to proceed?
```
## Task
Audit this document against the Minto Pyramid Principle. Diagnose the defects
and explain what must change. Do not rewrite the document.
Return only the result, with no explanation of how you produced it.
Prompt with Minto five parts, in order · 29 997 B
Below is a skill written as an instruction. Read it in full and apply it to the task at the end.===== SKILL.md =====SKILL.md at f4813ce 17 235 B===== references/rules.md =====references/rules.md at f4813ce 6 464 B===== references/templates.md =====references/templates.md at f4813ce 3 742 B===== TASK =====the whole case inputReturn only the result, with no explanation of how you produced it.
Case input before.md
<!-- Faithful paraphrase of Minto (2010), Appendix B, Exhibits B-5 and B-6.
Original wording for this fixture. -->
Fixture 03: Period Graph Books
Mode: audit
Language: en
Context
Corporate Financial Analysis recently took over production of four monthly
graph books used in presentations to senior management. Management knows that
errors and inconsistencies have appeared and asked for corrective proposals.
Before
Subject: Period Graph Books
Since Period 5, Financial Analysis has been responsible for producing four
graph-presentation books previously handled by Corporate Planning. This memo
describes issues that appeared after the transfer.
Production
1. Gather data from divisional forms, internal documents, and telephone calls.
2. Calculate graph points manually or on the corporate computer, including
rolling values for revenue, cost, and percentages.
3. Transcribe the points to input sheets. Analysts update the current period,
then Color Graphics re-enters the points into its own database and generates
the graphs.
4. Check the data and calculations.
Issues
The transfer inserted another person into an already fragmented process.
Divisional staff and three corporate departments participate, and the same
information may be entered into computer systems three times.
For one book, divisional data are loaded into a dedicated corporate program,
calculated, transferred to Color Graphics input sheets, entered again, and
returned for updates. Problems include inconsistent data between periods and
regions, incorrect variable-cost calculations, unexplained changes to points
that were previously correct, and prior-period data that must be re-entered.
Overall assessment
Most problems arise from a cumbersome fragmented process. No one controls the
data from beginning to end, responsibility is unclear, and errors can fall
between stages. The production process badly needs streamlining, both for each
book and for common calculations. The current team cannot manage the required
changes as well as ongoing production.
Could we discuss how best to proceed?
Task
Audit this document against the Minto Pyramid Principle. Diagnose the defects
and explain what must change. Do not rewrite the document.
The judge’s gold gold.md
Gold 03: Period Graph Books
Source structure: Minto (2010), Appendix B, Exhibits B-5 and B-6.
Book-derived defects
| Defect | Severity |
|---|---|
| The document never states the changes required; it therefore does not answer management's question | Critical |
| It describes the current process instead of presenting a top and key line | Critical |
| "The purpose is to describe issues" is a topic, not an answer | Critical |
| The process description is incomplete and does not isolate each system's failure points | Medium |
| Problems in graph-point generation and later presenter changes are not clearly separated | Medium |
The buried conclusion is that the process needs streamlining, but this is not
turned into an actionable answer.
Expected replacement structure
Reader question: What changes should we make to prevent unreliable graphs?
Top: Remove the opportunities for error across data transfer, graph-point
generation, and final presentation changes.
Key-line kind: changes/actions.
- Create a direct data link to the corporate computer.
- Create a reliable computerized routine for graph-point generation.
- Require presenter changes to be revalidated before use.
Audit outputs should name the missing answer, distinguish it from a conclusion
that is merely buried, annotate only visible evidence, provide no rewritten
memo, and avoid inventing defects in unseen material.
Judge gpt-5.6-sol, effort high · Claude Sonnet 5, effort low · skill from commit f4813ce · run v1.7.0-final