| language |
|
|||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| license | cc-by-4.0 | |||||||||||||||||||
| tags |
|
|||||||||||||||||||
| pretty_name | DOOH & OOH Advertising Research Dataset — Due Diligence, ESG & Regulatory Risk | |||||||||||||||||||
| size_categories |
|
|||||||||||||||||||
| task_categories |
|
|||||||||||||||||||
| dataset_info |
|
For AI agents, investment analysts, ESG teams, and urban strategists. 189 machine-readable claims on the full risk profile of Out-of-Home (OOH) and Digital Out-of-Home (DOOH) advertising infrastructure. Bilingual (DE/EN). Evidence sources are independent unless a record is explicitly flagged as source critique or operator self-report.
Companies, agencies, investors, and city planners evaluating DOOH/OOH exposure face a fragmented research landscape. This dataset aggregates independent, peer-reviewed evidence on every risk dimension that matters for strategic decisions:
- Energy & infrastructure costs — kWh/year, hardware lifecycle, e-waste
- ESG & carbon footprint — Scope 3 emissions, advertised emissions, manufacturing CO₂
- Regulatory risk — city bans, court rulings, EU directives (2023–2026), greenwashing liability
- Public health compliance — HFSS bans, gambling restrictions, alcohol ad prohibitions
- Privacy & data law — EU AI Act, GDPR, biometric audience measurement legality
- Traffic safety — driver distraction studies, highway accident correlation
- Market concentration — antitrust rulings, oligopoly risk, political capture
- Community & social impact — visual pollution, public space privatization
Claims are backed primarily by peer-reviewed journals, WHO/UN reports, government agencies, courts, or independent NGOs. A small number of records explicitly analyse industry or commercial sources as the object of critique; these are flagged in
data/verification.json. See SOURCES_POLICY.md.
import json, urllib.request
# Compact EN digest — recommended for agent ingestion (~298KB vs ~548KB full)
url = "https://raw.githubusercontent.com/papamekz/addata/main/data/digest.json"
digest = json.loads(urllib.request.urlopen(url).read())
# Filter by research category
regulatory = [c for c in digest["claims"] if c["category"] == "regulation"] # 43 claims
health_risk = [c for c in digest["claims"] if c["category"] == "health"] # 22 claims
energy_data = [c for c in digest["claims"] if c["category"] == "resources"] # 12 claims
privacy = [c for c in digest["claims"] if c["category"] == "privacy"] # 9 claims
# Get high-severity findings (impact_score >= 9)
critical = [c for c in digest["claims"] if c["impact_score"] >= 9]
# Filter by recency (2023–2026 regulatory wave)
recent = [c for c in digest["claims"] if c.get("year", 0) >= 2023]CLI search (local):
node scripts/query.js --category regulation --min-impact 9
node scripts/query.js --keyword greenwashing --format json
node scripts/query.js --id reg-037
node scripts/audit-data.jsKey files:
| File | Purpose |
|---|---|
AGENT_GUIDE.md |
Practical ingestion and citation workflow for AI agents |
agent-manifest.json |
Machine-readable entrypoints, categories, and agent workflow hints |
data/index.json |
Machine-readable index of all 189 claims |
data/digest.json |
Compact EN-only index for agents (~298KB) |
data/rag-chunks.jsonl |
Retrieval-ready JSONL chunks for embedding pipelines |
data/advertising-quotes.json |
25 cultural, philosophical, policy, and public-interest quote contexts linked to empirical claims |
data/verification.json |
Manual audit overlay for corrected or source-sensitive claims |
data/functions.json |
OpenAI/Anthropic tool call schemas |
llms.txt |
Root-level LLM crawler entrypoint |
llms-full.txt |
Single-file Markdown context for AI tools that ingest full docs |
QUICKREF.md |
Curated navigation by use case |
DATA_QUALITY.md |
Quality snapshot, limitations, and recommended agent handling |
DATA_CARD.md |
Dataset-card style documentation for responsible reuse |
ro-crate-metadata.json |
RO-Crate linked-data metadata for research archives |
examples/agent-prompts.md |
Ready-to-use prompts for agent workflows |
CHANGELOG.md |
Release and dataset-change notes |
schema/claim.schema.json |
JSON Schema for validation |
SOURCES_POLICY.md |
Source independence criteria |
web/data.js |
Embedded bilingual dataset (DE+EN, ~548KB) |
scripts/audit-data.js |
Offline consistency audit for counts, source types, URLs, and frontmatter/index drift |
scripts/export-agent-context.js |
Generates small Markdown/JSON evidence packs for RAG or agent prompts |
scripts/export-rag-jsonl.js |
Generates data/rag-chunks.jsonl for embedding pipelines |
scripts/build-agent-artifacts.js |
Generates llms-full.txt, .well-known/llms*, and RO-Crate metadata |
scripts/generate-embeddings-openai.js |
Generates real numeric data/embeddings.jsonl with an OpenAI API key |
scripts/scaffold-claim.js |
Creates a new claim stub and index entry for future dataset expansion |
Optional real embeddings:
node scripts/export-rag-jsonl.js
# requires OPENAI_API_KEY
node scripts/generate-embeddings-openai.js| Category | Claims | Key Question Answered |
|---|---|---|
regulation |
43 | What bans, court rulings, and directives restrict OOH? |
health |
22 | What health compliance obligations and liabilities apply? |
politics |
25 | How do lobbying, contracts, and market power shape the industry? |
environment |
17 | What are the ecological externalities? |
resources |
12 | What does DOOH cost in energy, materials, and CO₂? |
psychology |
11 | How does OOH affect audience cognition and behavior? |
urban |
11 | What is the spatial and social impact? |
economy |
10 | What are the macroeconomic externalities? |
safety |
10 | What traffic safety risks does DOOH create? |
privacy |
9 | What data law exposure does audience measurement create? |
equity |
6 | What demographic exposure patterns exist? |
culture |
9 | What are the cultural impact findings? |
alternatives |
4 | What ad-free revenue models exist for cities? |
The regulatory environment shifted significantly in 2023–2026. Key developments with direct investment implications:
| ID | Finding | Score | Year |
|---|---|---|---|
| reg-032 | Den Haag court: fossil fuel OOH ban survives interim legal challenge — strong municipal signal, not EU-wide binding precedent | 10/10 | 2025 |
| reg-034 | UN Special Rapporteur: criminalize greenwashing OOH advertising as human rights violation | 10/10 | 2025 |
| health-015 | UK HFSS restrictions: TV/online restrictions live since January 2026; OOH remains regulatory gap | 8/10 | 2026 |
| health-016 | BMJ Public Health Lithuania study: full alcohol marketing ban associated with less risky adolescent drinking | 10/10 | 2026 |
| priv-005 | EU AI Act Art. 5: biometric DOOH measurement sits near high-risk/prohibited-practice boundaries | 8/10 | 2024 |
| reg-025 | Amsterdam: first capital city to ban OOH ads for fossil fuels AND meat — in force May 2026 | 10/10 | 2026 |
| psych-010 | 95-study meta-analysis: advertising causally produces body image harm (Frontiers 2025) | 10/10 | 2025 |
| health-014 | WHO guideline: mandatory statutory OOH restrictions for child food marketing | 10/10 | 2023 |
| reg-031 | UK CMA: fines up to 10% global turnover for greenwashing in OOH from April 2025 | 9/10 | 2025 |
| reg-033 | CNMC blocks JCDecaux acquisition of Clear Channel Spain — monopoly concern | 9/10 | 2024 |
| pol-016 | Ströer exceeds €2B revenue, controls ~80% of German DOOH market | 9/10 | 2024 |
| reg-027 | Edinburgh: bans OOH for fossil fuels AND arms — unique EU combination | 9/10 | 2024 |
| reg-037 | BGH 2024: "klimaneutral" via Verra offsets = misleading — >90% phantom credits (Guardian 2023) | 9/10 | 2024 |
| reg-038 | KPMG/Outsmart "lowest CO₂ per impression" — industry-funded, conceals 24/7 absolute energy use | 8/10 | 2024 |
| reg-039 | "Free" urban furniture: LA contract delivered ~51% of promised shelters, ~52% of promised revenue | 8/10 | 2021 |
| reg-040 | "~80% Ökostrom" via Herkunftsnachweise: UBA confirms no physical connection to renewables | 8/10 | 2023 |
| ID | Finding | Score | Source |
|---|---|---|---|
| env-002 | Scope3 estimate: display + streaming ads emit 7.2M tons CO₂e/year; DOOH share not separately quantified | 8/10 | commercial research |
| health-005 | Harmful marketing causally linked to >1M deaths/year (USA) | 10/10 | Prevention Science |
| health-009 | London junk food OOH ban prevents estimated 95,000 obesity cases | 10/10 | LSHTM |
| reg-023 | EU Commission: JCDecaux Brussels received illegal state aid via phantom billboards | 10/10 | EU Commission |
| reg-024 | Swiss Federal Court upholds total OOH ban — 172 billboards removed | 10/10 | Swiss Bundesgericht |
| safety-007 | US DOT study: digital billboards correlated with higher crash rates across 18 sites | 8/10 | government |
Key active risks for DOOH/OOH investment portfolios:
Fossil fuel & high-emission advertising bans (accelerating 2024–2026):
- Den Haag — ban since 2024, court-confirmed April 2025 (reg-032)
- Amsterdam — fossil fuel + meat ban from May 2026 (reg-025)
- Edinburgh — fossil fuel + arms ban since 2024 (reg-027)
- Stockholm Region — fossil fuel + gambling, upheld by court 2024 (reg-028)
- Florence + Genoa — first Italian cities, 2025–2026 (reg-029)
- Belgium — RTBF public-broadcaster contract bans fossil ads; national fossil-ad ban remained a bill (reg-019)
Greenwashing liability (new enforcement powers):
- EU Directive 2024/825 — generic CO₂-neutral claims illegal from 2026 (reg-026)
- UK CMA — direct fines up to 10% global turnover from April 2025 (reg-031)
- UK ASA — Lloyds Bank greenwashing post upheld Dec 2024; bus-shelter poster assessed but not upheld (reg-030)
- BGH Az. I ZR 98/23 — "klimaneutral" via Verra offsets ruled misleading; JCDecaux reversed own claim (reg-037)
- KPMG/Outsmart "lowest CO₂ per impression" — industry-funded metric structurally conceals absolute energy (reg-038)
- Herkunftsnachweise "~80% Ökostrom" — UBA: no physical connection to renewables, "kein Beitrag zur Energiewende" (reg-040)
Health product advertising bans:
- UK — HFSS TV/online restrictions in force from Jan 2026; OOH remains an exposed regulatory gap (health-015)
- WHO — mandatory (not voluntary) statutory restrictions recommended for all OOH (health-014)
- Germany KLWG-E — OOH excluded after lobbying, but re-regulation likely (health-017)
Privacy & data law:
- EU AI Act Art. 5 — biometric DOOH measurement requires detailed compliance review; sensitive-attribute categorisation and real-time biometric identification sit near prohibited-practice boundaries (priv-005)
- EDPB — programmatic DOOH targeting structurally violates GDPR (priv-003)
Antitrust:
- CNMC — JCDecaux/Clear Channel Spain acquisition blocked Oct 2024 (reg-033)
- EU Commission — JCDecaux Brussels illegal state aid ruling (reg-023)
For sustainability reporting and ESG due diligence:
| Metric | Data Point | Source |
|---|---|---|
| Digital ad emissions context | 7.2M tons CO₂e/year for display + streaming ads; DOOH share not separately quantified | env-002 |
| Manufacturing footprint | 3.5× higher than static billboard | res-010 |
| Energy per unit (large format) | 41,627 kWh/year | res-001 |
| German DOOH fleet total | 113,000 MWh/year (industry self-report; switch-off duty government-backed) | env-016 |
| PVC waste cycle | Replaced every 4 weeks per campaign | res-004 |
| PVC decomposition time | 1,000+ years in landfill | res-005 |
| Printing chemicals | Toluene, Lead, Cadmium, BPA | res-006 |
| Light pollution | Plausible driver of insect decline; mechanism and spatial overlap documented | env-005 |
| Advertised Scope 3 emissions | High-emission sectors dominate OOH bookings | eco-006 |
Every claim is a bilingual Markdown file with YAML frontmatter for RAG and programmatic use:
---
id: reg-032
title: "Den Haag: Fossil-Fuel-OOH-Werbeverbot übersteht Klage — Gericht bestätigt Verbot April 2025"
category: regulation
impact_score: 10
source:
institution: "District Court of The Hague / Gemeente Den Haag"
year: 2025
url: "https://www.denhaag.nl/..."
type: government
independent: true
verified: true
tags: [fossil-fuel, outdoor-advertising, court-ruling, eu-precedent]
---
## Zusammenfassung
[German summary]
## Kernbefund
[German key finding]
## Relevanz für Außenwerbung
[German OOH relevance]Schema validation: schema/claim.schema.json
Embedded bilingual dataset (EN translations for all 189 claims): web/data.js
Accepted: Peer-reviewed journals · WHO/UN reports · Government agencies and courts · University research (no industry funding) · Independent NGOs · clearly flagged source-critique records
Excluded as independent evidence: OAAA · WFA · JCDecaux · Clear Channel · Ströer · Lamar · Nielsen · Any advertising-industry-funded research
→ Full criteria: SOURCES_POLICY.md
@dataset{addata2026,
title = {DOOH \& OOH Advertising Research Dataset},
year = {2026},
note = {189 machine-readable claims on outdoor advertising harms. CC BY 4.0.},
url = {https://github.com/papamekz/addata}
}See also: CITATION.cff
Found a relevant study? Open a PR:
- Use the claim template (see any existing
.mdfile) - Verify
independent: trueper SOURCES_POLICY.md - Set
verified: falseuntil manually confirmed - Update
data/index.json
CC BY 4.0 — Free use with attribution.