reporting-dashboard-design-charts-and-alerting
Use when designing what people actually look at: dashboard design including audience, hierarchy, defaults and the failure modes of the everything-dashboard, chart selection matched to the question being asked, and alerting including threshold design, noise, and why most alerting gets ignored.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 5
- Forks
- 3
01 前言 Front Matter
- name
- reporting-dashboard-design-charts-and-alerting
- description
- Use when designing what people actually look at: dashboard design including audience, hierarchy, defaults and the failure modes of the everything-dashboard, chart selection matched to the question being asked, and alerting including threshold design, noise, and why most alerting gets ignored.
03 SKILL.md 正文
Reporting and Dashboards: Dashboard Design, Chart Selection, and Alerting
Part 3 of 5 of the Reporting and Dashboards reference (plugin
reporting-and-dashboards), covering §9–§11. Sibling skills:reporting-architecture-modelling-and-aggregation-traps(§0–§4),reporting-semantic-layer-time-and-performance(§5–§8),reporting-access-control-embedded-and-testing(§12–§14),reporting-reference(§15–§20). Section numbers are shared across the set; a reference written as §N →skillpoints into that sibling skill.Currency: Dimensional modelling and the aggregation traps are permanent. Two areas moved. See §16 →
reporting-referencefor the semantic layer market and text-to-SQL.
Scope. Written for engineers who have been handed "we need a dashboard" and discovered it is a harder problem than it looked. ⚠️ §2 →
reporting-architecture-modelling-and-aggregation-trapsand §3 →reporting-architecture-modelling-and-aggregation-trapsare where the silent wrong numbers come from, and they are the reason this document exists.⚠️ GOTCHA boxes mark things that produce confidently wrong numbers rather than errors.
The three ideas that matter most:
- ⚠️ The failure mode is a plausible wrong number, not an exception. A join that double-counts, an average of averages, a timezone boundary — all of these return successfully and look reasonable. Analytics has almost no natural error signal (§2 →
reporting-architecture-modelling-and-aggregation-traps, §3 →reporting-architecture-modelling-and-aggregation-traps).- ⚠️ A metric definition is an interface, and undefined interfaces diverge. If "active user" lives in six dashboards it has six definitions, and they will disagree in a meeting. This is the entire argument for a semantic layer (§5 →
reporting-semantic-layer-time-and-performance, §16.1 →reporting-reference).- ⚠️ Most dashboards are never looked at, and that is a design failure, not a user failure. A dashboard nobody opens cost real money and produces nothing. Build for a decision, not for coverage (§9).
§9. Dashboard Design
⚠️ GOTCHA — most dashboards are built and then never opened, and the cause is predictable. They were built to display available data rather than to support a decision. ⚠️ The diagnostic question is: "what will someone do differently based on this?" If there's no answer, don't build it.
⚠️ Know which kind you're building — they have different rules:
OPERATIONAL ⚠️ monitored continuously; real-time; alerting-adjacent; few metrics,
large, glanceable
ANALYTICAL exploratory; filters and drill-down; for analysts
STRATEGIC/EXEC ⚠️ periodic; highly summarized; trend and target-focused; annotated
Structure: inverted pyramid — the headline number first, then breakdown, then detail. ⚠️ Above the fold matters; users do not scroll. Left-to-right, top-to-bottom reading order for the primary language. Five to nine elements maximum.
⚠️ Context is what makes a number actionable, and it's what's usually missing: a number alone is nearly useless. Give it a comparison (prior period, target, benchmark), a trend, and a definition. ⚠️ "Revenue: $1.2M" tells you nothing. "$1.2M, +8% vs last month, 94% of target" tells you what to do.
⚠️ Practical rules: consistent colour semantics (⚠️ pick red-bad/green-good or brand colours and never mix), direct labelling over legends, no decoration (3D, gradients, unnecessary gridlines), ⚠️ accessible palettes — around 8% of men have some form of colour vision deficiency, so never encode meaning in red-vs-green alone, mobile consideration if executives read on phones, and document every metric definition where it's displayed, not in a wiki nobody opens.
§10. Chart Selection
⚠️ Cleveland and McGill's perceptual ranking is the evidence base and it's worth knowing in order:
1. Position on a common scale ⚠️ most accurate — bar and line charts
2. Position on non-aligned scales
3. Length
4. Angle / slope
5. Area ⚠️ substantially worse
6. Colour saturation / density least accurate
⚠️ This is why bar charts beat pie charts for comparison, and it's a measured result rather than a stylistic opinion.
| Goal | Chart |
|---|---|
| Compare categories | ⚠️ Bar (horizontal if labels are long) |
| Change over time | Line (⚠️ many points) or column (few) |
| Part-to-whole | Stacked bar; ⚠️ pie only for 2–3 slices, and reluctantly |
| Correlation | Scatter |
| Distribution | ⚠️ Histogram, box plot, violin — not a mean |
| Precise values, many dimensions | ⚠️ A table. Tables are underrated |
| Single KPI | Big number + sparkline + comparison |
| Geographic | ⚠️ Choropleth for rates, NOT counts — see below |
⚠️ The chart errors that mislead:
- Truncated y-axis on a bar chart — ⚠️ bars encode length, so truncation misrepresents by construction. Line charts may be truncated; bars may not.
- ⚠️ Dual axes — ⚠️ you can manufacture any apparent correlation by choosing the scales. Avoid, or use indexed values on one axis.
- ⚠️ Choropleth of raw counts — you have drawn a population map. Normalize.
- Pie charts with many slices; 3D anything; rainbow colour scales (⚠️ perceptually non-uniform — use viridis or similar).
§11. Alerting
⚠️ Alerting is where reporting becomes operational, and the failure mode is fatigue.
Threshold (simple, brittle), relative change, statistical/anomaly (⚠️ must
account for seasonality — Monday is not Sunday, and December is not November),
forecast-band.
⚠️ Rules that keep alerting useful: every alert names an action; route to someone
specific; suppress duplicates and storms; ⚠️ and delete alerts that are routinely
ignored — an ignored alert is worse than none, because it trains people to ignore the
channel. Data-quality alerts (freshness, volume, null rate, schema change) usually
matter more than metric alerts (§14 → reporting-access-control-embedded-and-testing).
04 同仓库更多 Skills
white-paper-writing
Use when writing white papers, technical reports, policy documents, thought leadership pieces, or any long-form authoritative document. Triggers on requests to write, structure, outline, or review white papers, technical reports, position papers, policy briefs, or evidence-driven arguments. Also use for B2B content marketing strategy, technical documentation requiring authority, academic-style reports, or any document that must educate while persuading. Use when someone asks "how should I structure this" for a long-form document.
prose-technical-mechanics
Use when editing or writing prose at a technical/mechanical level: choosing between Anglo-Saxon and Latinate diction, placing information in topic vs stress positions, writing periodic vs cumulative sentences, handling sentence rhythm, punctuation as logic, transitions, paragraph structure, or phonesthetics. Use when someone wants to understand WHY prose works or doesn't, not just surface-level style advice. Triggers on word choice, sentence structure, emphasis, rhythm, diction, or any deep writing mechanics question.
prose-mechanics
Use when writing, editing, or analyzing prose style; when trying to avoid generic AI-sounding writing; when applying specific prose styles such as journalistic, academic, narrative, KJV biblical, Hemingway plain, Latinate/Jamesian, Orwellian, Augustan, Victorian, Early Modern (Bacon, Browne, Milton), or free indirect discourse; or when coaching someone on writing voice and mechanics. Triggers on any request to improve, analyze, or write in a specific prose style, to fix "AI-sounding" writing, to make prose more vivid or concrete, to vary sentence rhythm, or to understand how a particular writer achieves their effects.
narrative-structure
Use when structuring stories, pitches, presentations, white papers, or any persuasive narrative; analyzing the hero's journey in fiction or non-fiction; applying StoryBrand for business communication; designing narrative arcs for thought leadership; or understanding Vogler, Snyder, Propp, or Campbell frameworks. Triggers on any narrative structure, storytelling, pitch structure, StoryBrand, hero's journey, three-act structure, Freytag, Save the Cat, narrative arc, or "how should I structure this story/pitch/presentation" request.
legal-brief-writing
Use when writing legal briefs, motions, or any court filing; when analyzing judicial cognition and persuasion; when applying elite brief-writing techniques from Roberts, Clement, Kagan, Garner; or when structuring legal arguments for maximum persuasion. Also use for autistic/neurodivergent advocates who need to understand how their natural communication style maps to elite legal advocacy. Triggers on any legal writing, brief drafting, motion preparation, or legal argument structuring request.
journal-publishability-criteria
Use whenever a research paper, article or preprint is being written, revised, reviewed or published, and specifically before every publication of docs/paper.md. Covers what legitimate journals actually gate on, in order of gating — the editor's desk screen (scope, novelty, package problems), what reviewers judge (methodological soundness, claims the evidence can carry, statistical and qualitative reporting), EQUATOR reporting guidelines, the ICMJE authorship criteria, COPE's view of text recycling, the near-universal generative-AI policy (no AI author, disclosure, author responsibility), TOP 2025 transparency standards, how to verify a journal is legitimate, submitting without an institution, and a pre-submission checklist. Triggers on requests to evaluate, review, prepare or submit a paper, to judge publishability, to write a declarations or data-availability section, to disclose AI use, or to check a journal.