📘 Synthesis & Scoring Methodology

📊 Family A — Core Synthesis Layer
🎯 Significance Scoring · Synthesis Summary · Executive Brief · Stakeholder Perspectives · Stakeholder Impact

📋 Document Owner: CEO | 📄 Version: 1.4 | 📅 Last Updated: 2026-05-15 (UTC) 🔄 Review Cycle: Quarterly | ⏰ Next Review: 2026-07-21 🏢 Owner: Hack23 AB (Org.nr 5595347807) | 🏷️ Classification: Public

---

🎯 AI-FIRST Methodology Card

🚦 Read this card before writing a single paragraph. It names the artifact this methodology owns, the gate check it satisfies, the evidence-density target it must hit, and the Pass-1 / Pass-2 discipline required by .github/copilot-instructions.md §5 (AI-FIRST Quality Principle).

Field	Value
Purpose	Family A — step-by-step production of the 9 always-on core synthesis artifacts (executive brief, synthesis, significance, classification, SWOT, risk, threat, stakeholder).
Inputs	Per-document analyses (Family E); cross-reference-map.md; data-download-manifest.md; classification outputs
Outputs	executive-brief.md, synthesis-summary.md, significance-scoring.md, classification-results.md, swot-analysis.md, risk-assessment.md, threat-analysis.md, stakeholder-perspectives.md, folder README.md
Owning artifact(s)	All 9 Family A artifacts
Owning gate check	Checks 1, 3, 4 (evidence in SWOT + significance), 5 (Mermaid), 6 (Pass-2 done), 7 (Family C structure passed-through)
Citation density target	≥ 1 evidence anchor per ~120 words of analytical claim (synthesis-summary, stakeholder); SWOT ≥ 2 anchors per quadrant; significance-scoring ≥ 1 anchor per ranked item
Banned phrases	Enforced via political-style-guide.md §Machine-readable banned-phrase list
Threshold source	reference-quality-thresholds.json → thresholds[articleType][artifact] (fallback defaults.coreArtifactFloor)

✅ Pass-1 checklist (creation — minimal viable artifact)

• DIW 6-dimension weighting applied to every significance-scoring entry
• Executive-brief BLUF + 3 Decisions sections present (Check 7)
• Produce every required sub-section listed in the owning template
• Add ≥ 1 evidence anchor (dok_id, vote id, named MP, or primary-source URL) per analytical claim
• Apply the correct WEP confidence band for the run's horizon (72h / week / month / quarter / year / cycle)
• Include ≥ 1 themed Mermaid diagram with style … or themeVariables config (where structurally meaningful)
• Cross-link the relevant template under analysis/templates/ and the gate check it satisfies

🔁 Pass-2 checklist (read-back & improve — AI-FIRST mandatory)

• Cross-check SWOT quadrant counts ≥ 2 evidence-anchored entries each (Check 4)
• Confirm party arithmetic in coalition references is neutral and reflects Tidö support, not active coalition (see per-artifact-methodologies.md)
• Re-read the file end-to-end; flag every claim that lacks an evidence anchor and add one
• Replace every banned phrase listed in political-style-guide.md §Machine-readable banned-phrase list with an evidence-anchored alternative
• Tighten WEP language: never above likely without ≥ 3 cycle-aged sources for year/cycle horizons
• Strengthen Mermaid (color-coded style … directives, themeVariables, ≥ 5 nodes where the structure admits it)
• Add ≥ 1 second-order effect, cui-bono note, or counterfactual where the artifact admits one
• Verify citation density meets the per-file target below and the gate's evidence-density rules

🟢 Exemplar (good — pattern-match this)

(synthesis lede) "FiU committee report H901FiU1 (2026-04-22, 17 sponsors led by Edin (S)) reroutes SEK 6.1 bn from defence to climate. SD–M–KD bloc loses on punkt 5.3 (vote 173–176). Cui bono: opposition coalition wins fiscal narrative; cost: tightens FöU appropriation gap. WEP: highly likely Riksdag passes (3-source consensus, [B2])."

🔴 Anti-exemplar (failure mode — never ship this)

(failure mode) "The Riksdag debated the budget. Several parties spoke. The opposition raised concerns. The vote was close." — no dok_id, no MPs, no vote count, no WEP, banned-phrase "raised concerns".

🔗 Cross-links

• Template(s): analysis/templates/intelligence-assessment.md, analysis/templates/executive-brief.md
• Gate check: .github/prompts/05-analysis-gate.md
• AI-FIRST canon: .github/copilot-instructions.md §5 · ai-driven-analysis-guide.md
• Style canon: political-style-guide.md · osint-tradecraft-standards.md
• Catalog row: artifact-catalog.md

---

🔄 Tradecraft Anchors

Element	Value	Reference
F3EAD Stage	ANALYZE → DISSEMINATE	This methodology covers the synthesis and dissemination phases of the intelligence cycle
PIRs Served	PIR-1 (Coalition Stability), PIR-5 (Fiscal Trajectory), PIR-7 (Democratic Norms) — all PIRs inform synthesis	See political-style-guide.md §PIR/EEI Catalog
Admiralty Floor	Evidence must reach [B2] or higher for inclusion in executive brief	See political-style-guide.md §Admiralty Code
WEP Requirement	All forward-looking claims use WEP probability language	See political-style-guide.md §WEP + ODNI
ICD 203 Gate	Standards 5 (customer relevance), 6 (logical argumentation), 8 (accurate judgments)	See political-style-guide.md §ICD 203
SAT(s)	Key Assumptions Check (synthesis-summary), Brainstorming (stakeholder-perspectives)	See political-style-guide.md §SATs

---

🎯 Purpose

This methodology governs the Family A — Core Synthesis layer of every Riksdagsmonitor analysis workflow. The synthesis layer turns raw per-document evidence (Family E) and structural metadata (Family B) into decision-ready intelligence products that answer four questions:

1. What happened today? (synthesis-summary.md)
2. Which items matter most? (significance-scoring.md)
3. Who is affected and how? (stakeholder-perspectives.md + stakeholder-impact.md)
4. What should a decision-maker do in ≤90 seconds of reading? (executive-brief.md)

Every output in this family is produced for every daily workflow — no exceptions.

flowchart LR
    classDef evidence fill:#E3F2FD,stroke:#1565C0,color:#0D47A1
    classDef scoring fill:#FFF8E1,stroke:#FFC107,color:#3E2723
    classDef synthesis fill:#E8F5E9,stroke:#4CAF50,color:#1B5E20
    classDef brief fill:#F3E5F5,stroke:#7B1FA2,color:#311B92
    classDef stakeholder fill:#FFF3E0,stroke:#FF9800,color:#BF360C

    E[Family E<br/>per-document analyses]:::evidence
    B[Family B<br/>manifest + xref map]:::evidence
    S1[significance-scoring.md<br/>📊 rank + DIW]:::scoring
    S2[synthesis-summary.md<br/>📋 narrative synthesis]:::synthesis
    S3[stakeholder-perspectives.md<br/>🎭 party-by-party view]:::stakeholder
    S4[stakeholder-impact.md<br/>💥 who wins / who loses]:::stakeholder
    S5[executive-brief.md<br/>⚡ ≤90s decision brief]:::brief

    E --> S1
    B --> S1
    S1 --> S2
    S1 --> S3
    S3 --> S4
    S2 --> S5
    S4 --> S5

---

📐 Evidence & Confidence Standards

All outputs in this family inherit the platform-wide rules:

Rule	Applied To
Every claim cites a dok_id, named actor with party, vote count, rm session, or primary-source URL	All five outputs
5-level confidence scale — 🟦 VERY HIGH / 🟩 HIGH / 🟧 MEDIUM / 🟥 LOW / ⬛ VERY LOW	Required per scored row
Color-coded Mermaid using the canonical palette from the AI-Driven Analysis Guide	≥1 per output
Swedish policy terms glossed on first use (e.g. jobbskatteavdrag — earned-income tax credit)	Synthesis + executive brief
Party neutrality — equal analytical depth for S, M, SD, C, V, KD, MP, L when evidence supports coverage	All outputs

---

🔢 Part 1 — Significance Scoring (significance-scoring.md)

Purpose

Rank every document processed in the workflow so downstream products (synthesis, executive brief) can focus attention where it matters. Significance scoring is the single source of truth for document priority across Family A, C, and D products.

Input

• All Family E per-document analyses produced during the workflow
• Family B cross-reference-map.md (to detect bundle / coordinated activity multipliers)
• Voting data for each document (when applicable)

Output — required structure

1. Scoring matrix table — one row per document, columns:
   • Rank · dok_id · Title (short) · Doctype · Democratic-Impact Weight (DIW) · Tier · Confidence · Top rationale (≤15 words)
2. Tier distribution summary — count of P0 / P1 / P2 / P3 documents
3. DIW color-coded distribution — horizontal bar Mermaid using the canonical palette
4. Drivers of top-ranked items — bulleted rationale for every P0 + P1 entry with dok_id citations

DIW formula — canonical

Democratic-Impact Weight (DIW) is scored 0.0 – 10.0 along six dimensions, each 0.0 – 1.0 normalised, summed, then multiplied by 10 ÷ 6 for the final 0–10 scale:

Dimension	Weight	Example high-score indicator
🏛️ Institutional reach	1.0	Constitutional change, election-law amendment, agency mandate
⚖️ Rights & freedoms impact	1.0	GDPR article 9 data, speech, assembly, due process
💰 Fiscal magnitude	1.0	≥1% of statsbudget or ≥50 000 direct beneficiaries
🗳️ Electoral salience	1.0	Pre-election window + top-10 voter-priority issue
🔀 Coalition pressure	1.0	Forces deviation from Tidöavtalet or opposition red line
📡 Media resonance	1.0	Coordinated opposition response or government flagship announcement

Worked example — DIW computation

Suppose prop. 2025/26:113 raises the wealth-tax threshold from SEK 1.5M to SEK 5.0M. Score each dimension on the 0.0 – 1.0 anchor scale (0.0 = no signal, 0.5 = moderate, 0.8 = clearly material, 1.0 = exemplar):

Dimension	Score	Why
Institutional reach	0.5	Tax-code amendment — touches statutory framework but no constitutional change
Rights & freedoms	0.2	Distributional, not rights-based
Fiscal magnitude	0.9	~SEK 8 bn / yr ≈ 0.4 % of statsbudget × 5-yr horizon = clearly material
Electoral salience	0.8	Pre-election window + #4 voter-priority issue per SCB-Kantar Apr-2026
Coalition pressure	0.7	M/KD/L aligned; SD ambivalent on top quintile carve-out — friction but not a red-line breach
Media resonance	0.7	DN, SvD, Aftonbladet leaders Day 0; Ekot Day 0+1
Sum	3.8
DIW = (3.8 / 6) × 10	6.33	P1 — High (triggers scenario + devils-advocate + stakeholder-impact)

The rationale for each anchor score MUST cite dok_id + at least one named actor. Anchor scores must not be back-fitted to a target tier; if you find yourself inflating Coalition pressure to push a P2 into P1, write the explanation honestly and let the score sit. Calibration debt accumulates fast — it is preferable to have a borderline 5.9 documented honestly than a 6.1 that does not survive Pass-2 scrutiny.

Winner / loser quantification rubric

Every P0/P1 ranking row MUST narrate who wins and who loses in concrete, quantifiable terms before any forward-looking implication is drawn. The rubric:

Axis	Required form	Example
Identity	Named actors (party, ministry, committee, sector, demographic), not abstract groups	"M/KD/L coalition + top-decile income earners" not "the right"
Magnitude	A unit (SEK, % of voters, # of MPs, # of beneficiaries) and a horizon	"+ SEK 8 bn / yr from 2027 budget for top-decile"
Direction	Win / loss / mixed, with sign	"Win for top decile; loss for tax-revenue projection"
Confidence	5-level label backed by evidence reliability, not advocacy	🟩 HIGH — RUT-utvärdering 2024 + FiU's own modelling
Counter-narrative	One named actor's framed dissent	Quote V's Karin Pernebo (2026-04-22) on distributional effect

A synthesis ranking that names only winners and not losers (or vice versa) fails the gate: politics is zero-sum or near-zero-sum on most material questions, and asymmetric narration is itself a tradecraft tell. When the distribution is genuinely diffuse (e.g. an information-policy bill), say so explicitly with the phrase "diffuse-impact — no quantifiable concentrated win/loss" and back it with evidence.

Tier mapping

• P0 (Critical, DIW ≥ 8.0) → triggers every Family C + D template
• P1 (High, 6.0 ≤ DIW < 8.0) → triggers scenario, devils-advocate, stakeholder-impact
• P2 (Medium, 4.0 ≤ DIW < 6.0) → synthesised only
• P3 (Low, DIW < 4.0) → listed in manifest, not narrated

Quality gate

• Every document has a DIW in [0.0, 10.0]
• DIW cross-checks against Family E per-doc significance — ≤1.0 variance or reconciled
• Tier counts reconcile with Family B manifest total document count
• Top three rationales each cite ≥1 dok_id and ≥1 named actor

---

📋 Part 2 — Synthesis Summary (synthesis-summary.md)

Purpose

Produce the canonical narrative of the day/week/cycle. Every other Family A product derives its headline from this file.

Input

• significance-scoring.md (drives ordering)
• All Family E analyses (drives substance)
• Family B cross-reference-map.md (drives "bundle" and "pattern" callouts)
• Previous 5 workdays' synthesis-summary files (drives trend narration)

Output — required structure

1. Executive summary — 3 paragraphs, inverted-pyramid, each opening sentence citing a dok_id or named vote
2. Strategic finding cards — 3–7 cards, one per major theme; each card contains:
   • Finding (≤40 words) · Evidence (≥2 dok_id or votes) · Confidence label · Dissenting view · Forward indicator
3. Pattern table — coordinated filings, cross-party agreement, government flagship clusters
4. Mermaid thematic graph — color-coded nodes by theme, edges by cross-reference strength
5. Links to: executive-brief.md, stakeholder-perspectives.md, scenario-analysis.md (always produced, Family C), Family E index
6. Methodology footer — models used, sources, cutoff time

Required Mermaid — thematic graph example

graph TB
    classDef fiscal fill:#FFF8E1,stroke:#FFC107,color:#3E2723
    classDef security fill:#FFEBEE,stroke:#D32F2F,color:#B71C1C
    classDef climate fill:#E8F5E9,stroke:#4CAF50,color:#1B5E20
    classDef rights fill:#F3E5F5,stroke:#7B1FA2,color:#311B92
    classDef institutional fill:#E3F2FD,stroke:#1565C0,color:#0D47A1

    T1[Theme: Fiscal policy<br/>3 docs · DIW 7.4]:::fiscal
    T2[Theme: National security<br/>2 docs · DIW 8.1]:::security
    T3[Theme: Climate & energy<br/>4 docs · DIW 6.8]:::climate
    T4[Theme: Rights & rule of law<br/>1 doc · DIW 9.2]:::rights
    T5[Theme: Institutional reform<br/>2 docs · DIW 5.9]:::institutional

    T1 -->|shared sponsor M+KD| T3
    T2 -->|trade-off| T1
    T4 -->|constitutional| T5
    T3 -->|jurisdiction| T5

Quality gate

• Top-ranked P0/P1 items from significance-scoring appear in the first 5 strategic finding cards
• Every card cites ≥2 primary sources
• Confidence labels distributed realistically (not every card "HIGH")
• Swedish terms glossed on first use
• Thematic Mermaid includes all themes mentioned in cards

---

🎭 Part 3 — Stakeholder Perspectives (stakeholder-perspectives.md)

Purpose

Document how every actor with material stake interprets the day's findings — from their documented position, not the analyst's projection.

Input

• significance-scoring.md (drives which items to cover)
• search_anforanden + search_dokument output per party for the covered rm
• Family E per-doc analyses (carry party statements)
• SCB / Opinionsläget polling (where identity of affected group matters)

Output — required structure

1. Actor × Issue matrix — rows: actors (8 Riksdag parties + government + key myndigheter + EU + external); columns: top 5 issues from significance-scoring
   • Cell contents: stance glyph (✅ support · 🟧 conditional · 🟥 oppose · ⬜ silent) + 1-line rationale with source
2. Deep profiles — one per tier-1 actor (government + each of the 8 parties) with:
   • Documented position (dok_id, anförande quote, press release URL)
   • Strategic rationale (why they hold this position — supported by voting history)
   • Internal tension indicators (where party cohesion is stressed)
   • Forward move (likely next action in 7–30 days, labelled with confidence)
3. Cross-actor alignment Mermaid — color-coded edges (green = alignment, red = conflict, yellow = partial)
4. Links to stakeholder-impact.md (material consequences) and synthesis-summary.md

Required Mermaid — actor alignment

graph LR
    classDef gov fill:#E3F2FD,stroke:#1565C0,color:#0D47A1
    classDef oppleft fill:#FFEBEE,stroke:#D32F2F,color:#B71C1C
    classDef oppright fill:#FFF8E1,stroke:#FFC107,color:#3E2723
    classDef ext fill:#F3E5F5,stroke:#7B1FA2,color:#311B92

    M[Moderaterna]:::gov
    KD[Kristdemokraterna]:::gov
    L[Liberalerna]:::gov
    SD[Sverigedemokraterna]:::gov
    S[Socialdemokraterna]:::oppleft
    V[Vänsterpartiet]:::oppleft
    MP[Miljöpartiet]:::oppleft
    C[Centerpartiet]:::oppright
    EU[EU Commission]:::ext

    M ---|aligned| KD
    M ---|aligned| L
    M ---|Tidö| SD
    S ---|aligned| V
    S -.partial.- MP
    C -.issue-based.- S
    EU ---|directive pressure| M

Quality gate

• All 8 Riksdag parties covered (even if "silent" with rationale)
• Every stance has a source citation
• Internal tension section present for any party whose vote split >5% on this cycle's items
• Forward-move predictions include confidence labels and time windows

---

💥 Part 4 — Stakeholder Impact (stakeholder-impact.md)

Purpose

Translate political positions into material consequences for citizens, sectors, and institutions. Where stakeholder-perspectives answers "who thinks what", stakeholder-impact answers "who wins, who loses, by how much, when".

Input

• stakeholder-perspectives.md (actor list)
• significance-scoring.md (which documents have measurable impact)
• SCB microdata (Swedish-specific ground truth), IMF WEO/FM/IFS for quantitative scale (primary economic source), World Bank for non-economic residue only (WGI governance / environment / social / education participation / defence historicals)
• Family E per-document implementation feasibility sections

Output — required structure

1. Impact matrix — rows: affected group; columns: magnitude · direction · timeline · reversibility · affected count · confidence
2. Winners & losers panel — top 5 winners, top 5 losers, with:
   • Quantified benefit/loss (SEK, beneficiary count, service-level change)
   • Evidence (SCB, Riksrevisionen, budget-proposition math)
   • Confidence label
3. Sector heat map — Mermaid with color-coded impact intensity
4. Equity lens — impact disaggregated by income quintile, region (storstad/glesbygd), age group, gender where data allows
5. Mitigations / compensations — government-offered offsets with dok_id
6. Links to implementation-feasibility.md (Family D, always produced)

Required Mermaid — sector heat map

graph TB
    classDef critneg fill:#D32F2F,stroke:#B71C1C,color:#FFFFFF
    classDef highneg fill:#FF9800,stroke:#E65100,color:#FFFFFF
    classDef neutral fill:#9E9E9E,stroke:#424242,color:#FFFFFF
    classDef highpos fill:#4CAF50,stroke:#1B5E20,color:#FFFFFF
    classDef critpos fill:#1565C0,stroke:#0D47A1,color:#FFFFFF

    Health[Healthcare<br/>+2.3B SEK · 12mo]:::highpos
    Defence[Defence industry<br/>+8.1B SEK · 36mo]:::critpos
    Agri[Agriculture<br/>−450M SEK · 6mo]:::highneg
    Culture[Cultural grants<br/>−890M SEK · 12mo]:::critneg
    Edu[Primary education<br/>neutral · pending]:::neutral

Quality gate

• ≥1 quantitative figure per winner/loser row
• ≥1 equity-lens cut where SCB data exists
• Reversibility assessed (one-shot / ongoing / multi-cycle)
• Mitigations cited with dok_id or confirmed absence

---

⚡ Part 5 — Executive Brief (executive-brief.md)

Purpose

Deliver a ≤90-second read that gives a decision-maker the day's intelligence takeaway. If a reader only reads one file, this is it.

Input

• synthesis-summary.md (for the canonical narrative)
• significance-scoring.md (for the ranked items)
• stakeholder-impact.md (for the consequence line)
• Family C scenario-analysis.md (always produced) for futures paragraph

Output — required structure (strict 400–600 word budget)

1. Top line — single-sentence headline of the day (≤25 words)
2. Three bullets — the only three things that matter — each ≤30 words, each citing a dok_id or named vote
3. So what? paragraph — consequences for the next 7–30 days
4. Watch-list — 3–5 forward indicators (from Family D forward-indicators.md when present) with dates
5. One chart — a single Mermaid diagram (color-coded) — usually a quadrant or a short timeline
6. Links to: synthesis-summary.md, significance-scoring.md, stakeholder-impact.md, scenario-analysis.md

Required Mermaid — decision quadrant example

quadrantChart
    title Decision-weighting quadrant — today's items
    x-axis "Low fiscal magnitude" --> "High fiscal magnitude"
    y-axis "Low political salience" --> "High political salience"
    quadrant-1 "⚡ Act now"
    quadrant-2 "🔍 Watch closely"
    quadrant-3 "📎 File"
    quadrant-4 "📊 Analyse"
    "prop 2025/26:108": [0.82, 0.78]
    "mot 2025/26:3412": [0.34, 0.71]
    "bet FiU2": [0.91, 0.52]
    "SOU 2026:18": [0.58, 0.31]

Quality gate

• Word count within 400–600
• Every bullet has a primary-source citation
• No claim without a corresponding entry in synthesis-summary.md
• Mermaid diagram readable at a glance (≤10 nodes)
• Watch-list items have ISO dates

---

🛠️ Production Workflow — step-by-step

flowchart TD
    classDef input fill:#E3F2FD,stroke:#1565C0,color:#0D47A1
    classDef step fill:#E8F5E9,stroke:#4CAF50,color:#1B5E20
    classDef gate fill:#FFF8E1,stroke:#FFC107,color:#3E2723
    classDef output fill:#F3E5F5,stroke:#7B1FA2,color:#311B92

    I1[Family E per-doc analyses]:::input
    I2[Family B manifest + xref]:::input

    S1[Step 1 — Compute DIW per doc<br/>→ significance-scoring.md]:::step
    G1{Gate 1 — DIW reconciles<br/>with per-doc Family E?}:::gate
    S2[Step 2 — Narrate synthesis<br/>→ synthesis-summary.md]:::step
    S3[Step 3 — Build actor matrix<br/>→ stakeholder-perspectives.md]:::step
    S4[Step 4 — Quantify consequences<br/>→ stakeholder-impact.md]:::step
    G2{Gate 2 — Every claim cites<br/>primary source?}:::gate
    S5[Step 5 — Distill to 400-600 words<br/>→ executive-brief.md]:::step
    G3{Gate 3 — Brief readable<br/>in ≤90s?}:::gate

    O[Family A complete<br/>5/5 outputs]:::output

    I1 --> S1
    I2 --> S1
    S1 --> G1
    G1 -->|pass| S2
    G1 -->|fail| S1
    S2 --> S3
    S3 --> S4
    S4 --> G2
    G2 -->|pass| S5
    G2 -->|fail| S2
    S5 --> G3
    G3 -->|pass| O
    G3 -->|fail| S5

Typical time budget (automated workflow, single day)

Step	Share of Family A runtime
Step 1 — Significance scoring	20 %
Step 2 — Synthesis summary	35 %
Step 3 — Stakeholder perspectives	15 %
Step 4 — Stakeholder impact	15 %
Step 5 — Executive brief	15 %

---

✅ Family-A Completion Checklist

• significance-scoring.md — matrix · distribution chart · tiers · rationale per P0/P1
• synthesis-summary.md — 3-paragraph exec summary · 3–7 finding cards · pattern table · thematic Mermaid
• `stakeholder-perspectives.md$ — \text{actor} \times \text{issue} \text{matrix} · \text{deep} \text{profile} \text{per} \text{tier}-1 \text{actor} · \text{alignment} \text{Mermaid}
• $stakeholder-impact.md` — impact matrix · winners/losers · sector heat-map Mermaid · equity lens
• executive-brief.md — 400–600 words · 3 bullets · so-what · watch-list · 1 Mermaid
• Every file passes evidence gate (primary-source citations)
• Every file passes confidence-label gate (5-level scale)
• Every file uses canonical color palette
• All 5 files cross-linked to each other and back to Family B/E

---

🔗 Template bindings

Template	Methodology section
analysis/templates/significance-scoring.md	Part 1 above
analysis/templates/synthesis-summary.md	Part 2 above
analysis/templates/stakeholder-impact.md	Part 4 above (also covers stakeholder-perspectives.md)
analysis/templates/executive-brief.md	Part 5 above

---

📐 Cross-references to other methodology layers

• Upstream (inputs): per-document-methodology.md (Family E) · structural-metadata-methodology.md (Family B)
• Downstream (always produced on every run): strategic-extensions-methodology.md (Family C) · electoral-domain-methodology.md (Family D)
• Style & voice: political-style-guide.md
• Master protocol: ai-driven-analysis-guide.md

---

🔐 ISMS Alignment

Control	How this methodology satisfies it
ISO 27001 A.5.23 (Information security for use of cloud services)	All sources are public cloud APIs with documented endpoints in manifest
ISO 27001 A.5.34 (Privacy and PII protection)	Stakeholder profiles handle political-opinion data under GDPR Art. 9(2)(e)/(g); data minimisation enforced
NIST CSF ID.RA-4	Significance scoring = continuous risk assessment of institutional activities
NIST CSF DE.AE-3	Cross-party alignment Mermaid supports anomaly detection
CIS 3.1 Data management process	Provenance enforced via Family B manifest linkage

---

📜 Changelog

• v1.4 (2026-05-15) — Aligned version badge with header text (both now 1.4); bumped date to 2026-05-15.
• v1.3 (2026-04-25) — Line-by-line DIW worked example (wealth-tax proposition scored across all 6 dimensions) and winner/loser quantification rubric (5-column format + diffuse-impact escape hatch).

---

📄 Document Control

Owner: CEO (Intelligence Program) · Reviewer: CISO + Editorial Lead · Review Cycle: Quarterly Next Review: 2026-07-21 · Related: ai-driven-analysis-guide.md, political-style-guide.md

---

Generated following Riksdagsmonitor Synthesis & Scoring Methodology v1.0 — Family A Core Synthesis Layer.