GEO Audit Report Veolia
Full analysis of Veolia's visibility across AI-generated responses. 18,200 responses analyzed from 9,720 prompts across 4 markets and 2 AI engines, evaluated through 36 personas.
ChatGPT
ChatGPT / Gemini
Among Defined Entities
This report analyses 19,141 AI responses (9,720 ChatGPT + 9,421 Gemini) generated from 9,720 scientifically validated prompts across 36 personas, 4 markets, 3 product lines, and 3 funnel stages. March 2026.
Brand Impact Score (BIS)
Composite index measuring overall brand performance in AI-generated responses. Range: 0 – 100. Combines sentiment, positional relevance, mention density, and competitive standing into a single actionable score.
Interpretation: 80+ dominant • 60–80 well positioned • 40–60 present but not dominant • <40 competition wins.
\(\text{normalizedSentiment} = \dfrac{\text{sentimentScore} + 1}{2}\)
\(\text{BIS} = \bigl(\text{normalizedSentiment} \times 0.30 \;+\; \text{positionScore} \times 0.25 \;+\; \text{mentionScore} \times 0.25 \;+\; \text{competitiveScore} \times 0.20\bigr) \times 100\)
Share of Voice (SOV)
Percentage of all AI-generated responses in which the brand is mentioned at least once. Indicates raw visibility regardless of sentiment or position.
Share of Branded Voice (SBOV)
Market share within the subset of responses that contain at least one branded mention. Measures competitive share of the "branded conversation" in AI outputs.
Mention Score
Share of entity mentions relative to all entity mentions within a single response. Range: 0 – 1. A higher value means the brand dominates the mention space in that response.
Position Score
Measures positional importance of the brand within a response using logarithmic decay. Range: 0 – 1. Entities mentioned earlier carry more weight, reflecting the primacy effect in AI-generated text.
\(w(\text{position}) = \dfrac{1}{\log(\text{position} + 2)}\)
\(\text{positionScore} = \dfrac{\text{entityWeightSum}}{\text{totalWeightSum}}\)
Reference weights: position 0 = 1.443 • position 1 = 0.910 • position 10 = 0.417
Sentiment Score
Normalized sentiment polarity of brand mentions within AI responses. Range: −1 to +1. Derived from a raw 1–5 evaluation scale, centered at zero for neutral sentiment.
Competitive Score
Relative ranking of the brand among all entities mentioned in a response. Range: 0 – 1. A score of 1.0 means the brand ranked first; lower values indicate more entities were ranked above it.
Key Findings
Veolia ranks #1 in Share of Voice across both LLMs, commanding a 2.1–2.6× margin over the nearest competitor. With 7,845 total mentions across 19,141 responses, the brand is the default reference entity in water, waste, and energy services prompts. Competitive scores of 0.777 (ChatGPT) and 0.843 (Gemini) confirm that Veolia appears even when competitors are explicitly queried.
ENGIE edges Veolia in ChatGPT Brand Impact Score (49.2 vs 48.6) despite having only 233 responses to Veolia's 2,067. ENGIE's advantage is driven by a stronger energy-transition narrative and tighter positioning in clean-energy prompts. While ENGIE's volume is far smaller, the BIS delta signals a narrative gap that AI models have absorbed from third-party content.
Gemini delivers 38% more response volume for Veolia (2,737 vs 2,067 responses), but ChatGPT delivers measurably better positioning. ChatGPT's position score of 0.269 outperforms Gemini's 0.214, and its mention score of 0.226 exceeds Gemini's 0.154. This indicates that ChatGPT places Veolia earlier and more prominently in its responses, even though Gemini mentions the brand more frequently in absolute terms.
Hazardous Waste is the weakest product line with a ChatGPT BIS of 47.2, trailing Water Technologies (50.1) and Bioenergy (50.0) by approximately 3 points. The gap is most acute at the Awareness stage (46.2 vs 49.6 for Water Technologies), suggesting that AI models frame Veolia's hazardous waste capabilities through a compliance lens rather than as an innovation-driven service offering.
Across product lines, BIS consistently drops from the Consideration stage to the Decision stage. In Water Technologies, the gap is narrow (50.2 to 50.5), but in Hazardous Waste it compresses from 48.7 (Consideration) to 46.8 (Decision) — a 1.9-point decline. This pattern indicates that when users ask AI for final purchase or vendor-selection guidance, Veolia's recommendation strength weakens relative to the exploratory phase.
The Middle East delivers Veolia's highest BIS at 51.4 with 613 responses and a sentiment of +0.408, sitting 4.4 points above the US market (47.0). Australia (47.8) and Spain (47.5) occupy the middle ground. The Middle East advantage reflects Veolia's deep infrastructure presence in the Gulf states and strong media coverage of desalination and water-reuse projects in the region.
Three Strategic Imperatives
Veolia leads all 15 defined competitors in Share of Voice and holds the #1 position in the competitive set, but its Brand Impact Score sits at approximately 48 — firmly in the "present but not dominant" zone. The brand is consistently mentioned, but not consistently recommended. The 32-point gap between Veolia's current BIS and the dominance threshold (80+) is closeable, but not through incremental improvements. Three targeted content investments — decision-stage depth, hazardous waste narrative reframing, and owned media infrastructure — can begin to close that gap within a 90-day execution window.
Measure Veolia's visibility, positioning, and sentiment in generative AI responses (ChatGPT and Gemini) across its three core business areas and four target markets.
The Prompt Atlas
Discovery Recommendation Comparison Evaluation Criteria Feature Prioritization Risk Assessment Planning Implementation Guidance Budget Clarification Problem Solving
AI Engines Evaluated
Markets Analyzed
Product Lines
Personas
8 archetypes × 4 countries + 4 neutral (country-agnostic) personas.
Competitive Set
Beyond these 15 defined competitors, the AI engines spontaneously mention 9,700+ other entities across all responses. The competitive field in generative AI is far larger and more fragmented than any predefined competitor list.
Funnel Stages
Share of Voice (SOV) measures the percentage of all AI-generated responses in which a brand appears at least once. It is the most direct indicator of raw visibility: a brand with high SOV is being surfaced by the model regardless of sentiment or ranking position.
SOV = (responses mentioning brand / total valid responses) × 100. ChatGPT base: 9,374 responses • Gemini base: 9,421 responses.
Note: The GEO Radar detected 9,700+ distinct entities across both engines. This ranking shows only the 16 defined competitors in Veolia's competitive set.
ChatGPT — SOV Ranking (16 competitors)
Gemini — SOV Ranking (16 competitors)
- Veolia leads both engines by a wide margin. At 22.05% in ChatGPT and 29.05% in Gemini, Veolia appears in roughly one out of every four AI responses. Gemini surfaces Veolia almost one-third of the time.
- The gap to the nearest competitor is large. In ChatGPT, Clean Harbors sits second at 10.30% — a deficit of 11.75 pp. In Gemini, Xylem holds second place at 11.17%, trailing Veolia by 17.88 pp.
- SUEZ and Xylem are the most consistent rivals, ranking in the top 4 of both engines. Together they represent Veolia's primary AI-visible competition across water and environmental services.
- Clean Harbors shows engine divergence: it ranks #2 in ChatGPT (10.30%) but drops to #4 in Gemini (5.80%). This suggests ChatGPT's training data gives more weight to North American hazardous-waste operators.
- The long tail is thin. Beyond the top 6, no competitor exceeds 3.3% SOV in either engine. Brands like Dalkia, Ecolab, SAUR, American Water, and Veralto have minimal AI visibility (<1% SOV).
Brand Impact Score (BIS) is a composite index (0–100) that goes beyond raw visibility to measure the quality of a brand's presence in AI-generated responses. It combines four weighted components: sentiment polarity, positional prominence, mention density, and competitive ranking.
ChatGPT — BIS Ranking (16 competitors)
Gemini — BIS Ranking (16 competitors)
Veolia — BIS Component Breakdown
ChatGPT — BIS Components: Veolia vs Top 5 Competitors
| Brand | Responses | BIS | Sentiment | Position | Mention | Competitive |
|---|---|---|---|---|---|---|
| Veolia | 2,067 | 48.63 | 0.381 | 0.269 | 0.226 | 0.777 |
| ENGIE | 233 | 49.18 | 0.421 | 0.280 | 0.239 | 0.741 |
| Ameresco | 120 | 48.40 | 0.409 | 0.254 | 0.202 | 0.790 |
| Clean Harbors | 965 | 48.28 | 0.289 | 0.280 | 0.188 | 0.863 |
| Cleanaway | 304 | 47.46 | 0.286 | 0.279 | 0.210 | 0.798 |
| Xylem | 732 | 46.92 | 0.400 | 0.252 | 0.214 | 0.711 |
Gemini — BIS Components: Veolia vs Top 5 Competitors
| Brand | Responses | BIS | Sentiment | Position | Mention | Competitive |
|---|---|---|---|---|---|---|
| Veolia | 2,737 | 47.31 | 0.417 | 0.214 | 0.154 | 0.843 |
| Ameresco | 243 | 46.79 | 0.420 | 0.206 | 0.139 | 0.842 |
| Cleanaway | 416 | 46.40 | 0.413 | 0.198 | 0.143 | 0.833 |
| Clean Harbors | 546 | 45.89 | 0.370 | 0.195 | 0.138 | 0.850 |
| ENGIE | 309 | 44.55 | 0.404 | 0.185 | 0.145 | 0.762 |
| Xylem | 1,052 | 42.82 | 0.401 | 0.158 | 0.124 | 0.737 |
- ENGIE edges past Veolia in ChatGPT BIS (49.18 vs 48.63). Despite appearing in only 233 responses (vs. Veolia's 2,067), ENGIE outperforms on sentiment (0.421 vs 0.381), position (0.280 vs 0.269), and mention density (0.239 vs 0.226). Veolia compensates with a stronger competitive score (0.777 vs 0.741).
- Veolia leads Gemini BIS outright at 47.31. Its competitive score of 0.843 is the highest among all 16 competitors, meaning that when Veolia appears, it consistently outranks other brands in the same response.
- Veolia's weakest component is mention density (0.226 ChatGPT, 0.154 Gemini). While the brand appears in many responses (high SOV), those responses also cite multiple other entities, which dilutes Veolia's per-response mention share.
- Clean Harbors has the highest competitive score in ChatGPT (0.863) but the lowest sentiment among the top 5 (0.289). It wins on positioning and competitive ranking but receives more neutral or mixed descriptions.
- The BIS range is compressed among defined competitors. In ChatGPT, the spread from #1 (ENGIE, 49.18) to #6 (Xylem, 46.92) is only 2.26 points. This tight clustering means small improvements in any sub-score can shift rankings.
How Sentiment Is Scored
Every entity mention captured by GEOradar receives a sentiment score on a 1–5 integer scale. The score reflects the tone of the AI-generated text surrounding each mention. All figures on this page use the raw 1–5 scale with no normalization applied.
Very Negative
Negative
Neutral
Positive
Very Positive
ChatGPT — Sentiment Ranking (16 competitors)
Gemini — Sentiment Ranking (16 competitors)
All 16 competitors score between 3.00 and 3.87 — well above the neutral midpoint of 3.0. This is typical for environmental services: AI models frame established operators through a sustainability and public-benefit lens, which suppresses negative language. Sentiment alone does not differentiate brands in this sector. Competitive advantage must come from volume, positioning, and recommendation strength.
Veolia — Score Distribution Across Both Engines
How Veolia's mentions break down by raw sentiment score (1–5) on each AI engine.
ChatGPT (3,292 mentions)
| Score | Label | Count | Share |
|---|---|---|---|
| 5 | Very Positive | 5 | 0.2% |
| 4 | Positive | 2,521 | 76.6% |
| 3 | Neutral | 754 | 22.9% |
| 2 | Negative | 12 | 0.4% |
| 1 | Very Negative | 0 | 0.0% |
Gemini (4,555 mentions)
| Score | Label | Count | Share |
|---|---|---|---|
| 5 | Very Positive | 4 | 0.1% |
| 4 | Positive | 3,822 | 83.9% |
| 3 | Neutral | 726 | 15.9% |
| 2 | Negative | 3 | 0.1% |
| 1 | Very Negative | 0 | 0.0% |
This is one of the sharpest cross-engine divergences in the audit. On ChatGPT, four smaller competitors (Ecolab at 3.87, ENGIE at 3.85, Ameresco at 3.82, Xylem at 3.81) score higher than Veolia, partly because their lower mention volumes are concentrated in positive-framing prompts. On Gemini, Veolia's massive volume (4,555 mentions) does not dilute its sentiment — it leads the entire 16-competitor field at 3.84. Gemini's training data weights Veolia's sustainability narrative more heavily, and the neutral share drops from 22.9% to 15.9%.
Veolia Sentiment by Product Line
Breaking Veolia's mentions by product line exposes a clear drag effect from Hazardous Waste on ChatGPT. Gemini shows the same pattern but with a much narrower gap.
| Product Line | Engine | Avg Sentiment | Mentions | Gauge (1–5 scale) |
|---|---|---|---|---|
| Water Technologies | ChatGPT | 3.85 | 986 | |
| Gemini | 3.86 | 1,408 | ||
| Bioenergy & Efficiency | ChatGPT | 3.85 | 707 | |
| Gemini | 3.85 | 1,307 | ||
| Hazardous Waste | ChatGPT | 3.68 | 1,599 | |
| Gemini | 3.82 | 1,840 |
On ChatGPT, Hazardous Waste scores 3.68 — a gap of 0.17 points below Water Technologies (3.85) and Bioenergy (3.85). This product line accounts for 10 of Veolia's 12 negative mentions on ChatGPT, and its neutral share (31.3%) is more than double that of Water Technologies (14.8%). AI models frame hazardous waste through a compliance and risk lens, producing more cautious, hedged language. On Gemini, the gap nearly disappears: Hazardous Waste scores 3.82 versus 3.86 for Water Technologies. Sector-specific content emphasizing innovation, recovery rates, and circular economy outcomes can shift the ChatGPT framing from risk-neutral to actively positive.
Cross-Engine Summary
Sentiment is a hygiene metric in environmental services — all 16 competitors are positive, and none averages below 3.0 on either engine. The real signal is in the internal gaps: Veolia's Hazardous Waste line underperforms its own Water and Bioenergy lines on ChatGPT by 0.17 points, and ChatGPT frames Veolia more cautiously than Gemini does overall. The actionable target is not "improve sentiment" in general but to close the Hazardous Waste gap on ChatGPT and ensure that engine carries the same positive framing that Gemini already delivers.
What Is Co-occurrence?
Co-occurrence counts how often two entities appear in the same AI response. When a user asks an LLM about water treatment and the model names both Veolia and SUEZ in its answer, that is one co-occurrence event. High co-occurrence means the AI treats two brands as belonging to the same competitive space. This section isolates only the 15 defined competitors from the GEO Radar study. Non-competitor entities (regulators, tech firms, reference sources) are covered in the Entity Ecosystem section.
Competitor Co-occurrence with Veolia — Combined Engines
Bars show the total number of distinct AI responses (ChatGPT + Gemini) where both Veolia and the competitor are mentioned together.
| # | Competitor | Co-occurrences | GPT | Gem | Total |
|---|---|---|---|---|---|
| 1 | SUEZ | 480 | 630 | 1,110 | |
| 2 | Xylem | 207 | 438 | 645 | |
| 3 | Clean Harbors | 397 | 228 | 625 | |
| 4 | Cleanaway | 109 | 244 | 353 | |
| 5 | ENGIE | 78 | 155 | 233 | |
| 6 | Republic Services | 83 | 149 | 232 | |
| 7 | REMONDIS | 60 | 125 | 185 | |
| 8 | Waste Management | 56 | 92 | 148 | |
| 9 | Aqualia | 36 | 82 | 118 | |
| 10 | Ecolab | 29 | 86 | 115 | |
| 11 | Saur | 12 | 50 | 62 | |
| 12 | Ameresco | 12 | 43 | 55 | |
| 13 | Dalkia | 10 | 6 | 16 | |
| 14 | American Water | 4 | 5 | 9 | |
| 15 | Veralto | 2 | 1 | 3 |
Engine-by-Engine Ranking
ChatGPT
| # | Competitor | Co-occ. |
|---|---|---|
| 1 | SUEZ | 480 |
| 2 | Clean Harbors | 397 |
| 3 | Xylem | 207 |
| 4 | Cleanaway | 109 |
| 5 | Republic Services | 83 |
| 6 | ENGIE | 78 |
| 7 | REMONDIS | 60 |
| 8 | Waste Management | 56 |
| 9 | Aqualia | 36 |
| 10 | Ecolab | 29 |
| 11 | Saur | 12 |
| 12 | Ameresco | 12 |
| 13 | Dalkia | 10 |
| 14 | American Water | 4 |
| 15 | Veralto | 2 |
Gemini (1.5 Pro)
| # | Competitor | Co-occ. |
|---|---|---|
| 1 | SUEZ | 630 |
| 2 | Xylem | 438 |
| 3 | Cleanaway | 244 |
| 4 | Clean Harbors | 228 |
| 5 | ENGIE | 155 |
| 6 | Republic Services | 149 |
| 7 | REMONDIS | 125 |
| 8 | Waste Management | 92 |
| 9 | Ecolab | 86 |
| 10 | Aqualia | 82 |
| 11 | Saur | 50 |
| 12 | Ameresco | 43 |
| 13 | Dalkia | 6 |
| 14 | American Water | 5 |
| 15 | Veralto | 1 |
Competitive Proximity Network (ChatGPT)
Concentric rings represent co-occurrence intensity. Veolia sits at the center; competitors closer to the core appear more frequently alongside Veolia in AI responses.
Competitive Tiers
The data reveals three distinct tiers of competitive proximity to Veolia in AI-generated content:
SUEZ leads competitor co-occurrence on both engines by a factor of roughly 2×. This is a direct consequence of the 2022 acquisition: LLM training data is saturated with merger coverage, regulatory filings, and competitive analyses that name both companies together. The implication is twofold. Positive: Veolia inherits context from SUEZ's brand equity in water and waste. Negative: AI models still treat the brands as entangled rather than fully integrated, which may confuse users who expect a single entity.
Several competitors shift rank significantly between engines:
- Clean Harbors ranks #2 on ChatGPT (397) but drops to #4 on Gemini (228). ChatGPT frames hazardous waste management as a primary Veolia context more often.
- Xylem ranks #3 on ChatGPT (207) but rises to #2 on Gemini (438). Gemini draws more heavily on water technology comparisons.
- Ecolab gets 3× more co-occurrence on Gemini (86 vs 29), suggesting Gemini ties Veolia to industrial water treatment contexts more than ChatGPT does.
- Dalkia is notably weak on both engines (10 + 6 = 16 total), despite being a former Veolia subsidiary. AI models no longer strongly connect the two brands.
Three defined competitors barely register in Veolia's co-occurrence data: Veralto (3 total), American Water (9 total), and Dalkia (16 total). AI models do not consider these brands to be in the same competitive space as Veolia. For Veralto and American Water, this reflects genuine market distance. For Dalkia — a former Veolia subsidiary active in energy services — the low co-occurrence suggests successful brand separation but also a missed opportunity if Veolia wants to reclaim that narrative.
Beyond the Competitive Set
When AI models mention Veolia, they also reference a broader ecosystem of entities that are not direct competitors: regulatory bodies, technology companies, reference sources, regional operators, and engineering firms. These co-occurrences reveal how LLMs contextualize Veolia — whether as a regulated utility, a technology partner, or a market-specific operator. This page analyzes the top non-competitor entities that appear alongside Veolia on each engine.
Top 15 Non-Competitor Entities — ChatGPT
| # | Entity | Co-occ. | Category |
|---|---|---|---|
| 1 | Wikipedia | 442 | Reference |
| 2 | Tradebe | 198 | Haz. Waste |
| 3 | EPA | 176 | Government |
| 4 | Acciona | 117 | Regional Op. |
| 5 | Tadweer Group | 102 | Middle East |
| 6 | Stericycle | 94 | Medical Waste |
| 7 | Montrose Environmental | 85 | Env. Services |
| 8 | Fluence Corporation | 82 | Water Tech |
| 9 | IDE Technologies | 80 | Desalination |
| 10 | Urbaser | 65 | Regional Op. |
| 11 | Averda | 63 | Middle East |
| 12 | Triumvirate Environmental | 62 | Env. Services |
| 13 | Schneider Electric | 58 | Industrial Tech |
| 14 | TMA | 50 | Engineering |
| 15 | BEEAH | 48 | Middle East |
Top 15 Non-Competitor Entities — Gemini
| # | Entity | Co-occ. | Category |
|---|---|---|---|
| 1 | EPA | 506 | Government |
| 2 | Fluence Corporation | 262 | Water Tech |
| 3 | Acciona | 186 | Regional Op. |
| 4 | Hydroflux Epco | 185 | Water Tech |
| 5 | Siemens | 147 | Industrial Tech |
| 6 | Clean Earth | 138 | Env. Services |
| 7 | Aquatech | 129 | Water Tech |
| 8 | Averda | 125 | Middle East |
| 9 | Austrans Group | 119 | Australia |
| 10 | Schneider Electric | 111 | Industrial Tech |
| 11 | Enviropacific Services | 106 | Australia |
| 12 | Evoqua Water Technologies | 90 | Water Tech |
| 13 | Pure Environmental | 85 | Australia |
| 14 | SIRC (Saudi Inv. Recycling) | 84 | Middle East |
| 15 | Crystal Clean | 84 | Haz. Waste |
Deep Ecosystem — Gemini (Ranks 16–30)
| # | Entity | Co-occ. | Category |
|---|---|---|---|
| 16 | Environmental Treatment Solutions | 82 | Waste services |
| 17 | Tadweer | 78 | Middle East waste |
| 18 | Metito | 76 | Water / Middle East |
| 19 | ABB | 73 | Industrial tech |
| 20 | Arcwood Environmental | 71 | Waste services |
| 21 | IDE Technologies | 70 | Desalination |
| 22 | Indaver | 69 | European waste-to-energy |
| 23 | MAK Water | 69 | Water / Australia |
| 24 | Tradebe | 69 | Hazardous waste |
| 25 | Alfa Laval | 67 | Water tech / industrial |
| 26 | Ace Waste | 63 | Waste / Australia |
| 27 | WSP | 63 | Engineering consulting |
| 28 | Saltworks Technologies | 60 | Brine treatment |
| 29 | BEEAH Group | 58 | Middle East waste |
| 30 | Empower | 56 | District cooling / UAE |
Ecosystem Cluster Map
Grouping the non-competitor co-occurrences by domain reveals six distinct clusters that shape how AI models frame Veolia:
Wikipedia ranks #1 on ChatGPT with 442 co-occurrences but does not appear in Gemini's top entities at all. This strongly indicates that ChatGPT draws from or references Wikipedia-style content when generating responses about Veolia. The implication: Veolia's Wikipedia page is a high-priority GEO asset for ChatGPT visibility. Any inaccuracies, outdated information, or missing service descriptions on that page will propagate directly into AI-generated answers. Auditing and optimizing the Wikipedia presence should be an immediate GEO action item.
The EPA co-occurs with Veolia in 506 Gemini responses (its #1 non-competitor entity) and 176 on ChatGPT. This confirms a pattern also visible in sentiment analysis: AI models frequently frame Veolia within regulatory and compliance contexts. When users ask about environmental services, Gemini in particular tends to mention both Veolia and the EPA together, which can prime users to think about enforcement rather than solutions and innovation. The gap is 2.9× (506 vs 176), meaning regulatory framing is far more aggressive on Gemini.
Gemini surfaces significantly more regional and specialist entities than ChatGPT. Key differences:
- Wikipedia is #1 on ChatGPT (442) but absent from Gemini's top results. ChatGPT relies more on encyclopedic sources.
- Fluence Corporation jumps from 82 (ChatGPT) to 262 (Gemini, 3.2×). Gemini's water-technology ecosystem is significantly broader.
- Hydroflux Epco (185) and Aquatech (129) appear only on Gemini's top list, absent from ChatGPT's top 15.
- The Australian cluster (Austrans, Enviropacific, Pure Environmental, Ace Waste, MAK Water) is almost entirely a Gemini phenomenon.
- Tradebe shows the opposite pattern: 198 on ChatGPT but only 69 on Gemini. ChatGPT treats Tradebe as a closer peer in hazardous waste contexts.
The two most prominent non-competitor entities are a reference source (Wikipedia) and a regulator (EPA). This tells a clear story: AI models understand Veolia through the lens of publicly available reference content and regulatory frameworks. To shape the AI narrative, Veolia should invest in both: (1) optimizing its Wikipedia presence for accuracy and completeness, and (2) producing thought leadership content that reframes the regulatory context from compliance obligation to proactive environmental leadership. Engine-specific GEO strategies are warranted: regional content optimization matters more for Gemini, while Wikipedia and broad brand messaging matter more for ChatGPT.
Veolia's three core product lines — Water Technologies, Hazardous Waste, and Bioenergy & Energy Efficiency — perform unevenly across AI engines. This section disaggregates the Brand Impact Score by product line to identify where narrative strength and content gaps diverge.
Sentiment +0.432 · Position 0.281
Sentiment +0.428 · Position 0.293
Sentiment +0.331 · Position 0.252
ChatGPT vs Gemini by Product Line
Volume vs Quality: Product Line Scatter
X-axis = total mentions (volume) · Y-axis = BIS (quality). Bubble size reflects response count. Data from ChatGPT run. Scale: X 0–1,800 mentions, Y 45–52 BIS.
Hazardous Waste dominates in volume with 1,597 mentions across 1,033 ChatGPT responses — more than Water Technologies and Bioenergy combined — yet delivers the lowest BIS at 47.2. The scatter plot reveals a clear inverse relationship: the product line generating the most AI visibility is the one with the weakest brand impact. This signals a narrative quality deficit: AI models associate Veolia with hazardous waste frequently, but frame it through compliance and regulatory language rather than innovation and competitive advantage.
Bioenergy & Energy Efficiency achieves a BIS of 50.1 with only 404 responses and 707 mentions in ChatGPT — the highest quality score but lowest volume. Position score (0.293) leads all product lines, meaning Veolia is mentioned earlier in responses about bioenergy topics. In Gemini, Bioenergy also leads at 48.4 BIS. The opportunity is clear: scaling content production in this vertical could amplify a narrative that already performs well qualitatively. Competitive pressure is lower here — ENGIE (49.5) is the only close challenger.
Top Competitors by Product Line (ChatGPT)
| Product Line | Entity | Responses | Avg BIS |
|---|---|---|---|
| Bioenergy & Efficiency | Veolia | 404 | 50.1 |
| Bioenergy & Efficiency | ENGIE | 218 | 49.5 |
| Bioenergy & Efficiency | Ameresco | 105 | 48.7 |
| Bioenergy & Efficiency | Xylem | 79 | 48.0 |
| Bioenergy & Efficiency | SUEZ | 110 | 44.2 |
| Hazardous Waste | Clean Harbors | 958 | 48.3 |
| Hazardous Waste | Cleanaway | 261 | 47.7 |
| Hazardous Waste | Veolia | 1,033 | 47.2 |
| Hazardous Waste | SUEZ | 187 | 41.9 |
| Hazardous Waste | Republic Services | 195 | 37.2 |
| Water Technologies | Veolia | 630 | 50.1 |
| Water Technologies | Xylem | 635 | 46.8 |
| Water Technologies | SUEZ | 441 | 45.4 |
| Water Technologies | ENGIE | 10 | 44.9 |
BIS by Funnel Stage × Product Line (ChatGPT)
Brand Impact Score averaged across all responses matching each funnel–product intersection. Cell shading: green ≥ 50, amber 48–50, red < 48.
| Funnel Stage | Water Technologies | Bioenergy & Efficiency | Hazardous Waste |
|---|---|---|---|
| Awareness | 49.6 (218 resp.) | 47.7 (129 resp.) | 46.2 (391 resp.) |
| Consideration | 50.2 (209 resp.) | 51.6 (149 resp.) | 48.7 (326 resp.) |
| Decision | 50.5 (203 resp.) | 50.7 (126 resp.) | 46.8 (316 resp.) |
Water Technologies and Bioenergy both strengthen as users move from Awareness to Decision, peaking at 50.5 and 50.7 BIS respectively in the Decision stage. Hazardous Waste follows the opposite trajectory: it weakens from 46.2 (Awareness) through a brief Consideration recovery (48.7) back down to 46.8 at Decision. The Consideration-to-Decision drop of 1.9 points in Hazardous Waste is the most critical gap in the matrix — it means Veolia is being considered but not recommended as the final vendor choice in hazardous waste contexts.
Veolia's AI visibility varies significantly across the four markets studied: United States, Australia, Spain, and the Middle East. This section maps territorial performance using persona-based prompt segmentation to reveal where Veolia's narrative is strongest, where competitors outperform it, and which markets represent the greatest optimization opportunity.
BIS Comparison Across Markets
ChatGPT
Gemini
The Middle East is Veolia's strongest market by a decisive margin: 51.4 BIS in ChatGPT and 48.5 in Gemini, both the highest across all four markets. The advantage stems from two structural factors: (1) fewer local competitors — the Gulf states' environmental services market is dominated by international players, reducing competitive noise; and (2) mega-project visibility — Veolia's high-profile desalination and water-reuse contracts (e.g., Ras Al Khair, Jebel Ali) generate extensive third-party media coverage that AI models absorb. The Middle East desalination market is growing at approximately 8% CAGR, and Veolia's narrative is tightly coupled to this infrastructure expansion.
Spain ranks third with a ChatGPT BIS of 47.5 and a Gemini BIS of 46.4. Despite Veolia's significant operational presence in Spain (water concessions, waste-to-energy facilities), the veolia.es domain generates minimal AI citations. ENGIE leads in Spain (50.2 BIS in ChatGPT) with a stronger digital footprint in Spanish-language energy content. AI models trained on web-crawled data naturally favor domains with high authority signals — and veolia.es lacks the structured data, FAQ schemas, and interlinking depth needed to compete with ENGIE's Spanish-language presence for AI retrieval.
The US is Veolia's weakest market at 47.0 BIS (ChatGPT) and 45.8 BIS (Gemini). This is the most competitive territory in the dataset: Clean Harbors (48.4), Ameresco (47.4), and even Xylem (46.6) all compete for AI attention in the ~$120B US environmental services market. Veolia ranks third behind Clean Harbors and Ameresco in US-specific ChatGPT responses. The US market's density of well-funded, publicly traded competitors with strong English-language content creates a narrative congestion that suppresses Veolia's relative positioning.
Full Market Comparison
| Market | Engine | Responses | Mentions | BIS | Sentiment | Position |
|---|---|---|---|---|---|---|
| Middle East | ChatGPT | 613 | 970 | 51.4 | +0.408 | 0.310 |
| Middle East | Gemini | 808 | 1,350 | 48.5 | +0.412 | 0.236 |
| Australia | ChatGPT | 507 | 785 | 47.8 | +0.374 | 0.256 |
| Australia | Gemini | 804 | 1,345 | 47.8 | +0.438 | 0.210 |
| Spain | ChatGPT | 517 | 850 | 47.5 | +0.367 | 0.257 |
| Spain | Gemini | 571 | 955 | 46.4 | +0.392 | 0.212 |
| United States | ChatGPT | 430 | 685 | 47.0 | +0.366 | 0.240 |
| United States | Gemini | 554 | 905 | 45.8 | +0.417 | 0.188 |
Competitive Ranking by Market (ChatGPT)
Top entities by BIS within each market. Veolia rows are highlighted. Only entities with ≥10 responses shown.
Middle East
| Entity | Resp. | BIS |
|---|---|---|
| ENGIE | 96 | 51.4 |
| Veolia | 613 | 51.4 |
| Xylem | 190 | 48.5 |
| Clean Harbors | 185 | 48.1 |
| SUEZ | 226 | 44.3 |
Australia
| Entity | Resp. | BIS |
|---|---|---|
| Clean Harbors | 156 | 48.4 |
| Veolia | 507 | 47.8 |
| Cleanaway | 303 | 47.5 |
| Xylem | 155 | 46.4 |
| SUEZ | 199 | 44.0 |
Spain
| Entity | Resp. | BIS |
|---|---|---|
| ENGIE | 71 | 50.2 |
| Clean Harbors | 112 | 48.0 |
| Veolia | 517 | 47.5 |
| Xylem | 146 | 46.0 |
| SUEZ | 200 | 44.5 |
United States
| Entity | Resp. | BIS |
|---|---|---|
| Clean Harbors | 512 | 48.4 |
| Ameresco | 82 | 47.4 |
| Veolia | 430 | 47.0 |
| Xylem | 241 | 46.6 |
| SUEZ | 113 | 44.6 |
United States: The US environmental services market is valued at approximately $120B, with intense competition from publicly traded players (Clean Harbors, Republic Services, Waste Management). Veolia's 2022 acquisition of SUEZ's North American assets has not yet translated into proportional AI narrative gains — Clean Harbors leads in BIS by 1.4 points.
Middle East: The GCC desalination market is growing at ~8% CAGR, driven by population growth and water scarcity. Veolia operates the world's largest desalination plant (Ras Al Khair, Saudi Arabia) and holds major contracts in the UAE, Qatar, and Oman. This infrastructure dominance directly feeds AI model training data through project-specific media coverage.
Australia: Veolia competes with local champion Cleanaway (BIS 47.5) in a market worth ~A$15B. The two brands are near-parity in AI perception, with Cleanaway's domestic brand recognition partially offsetting Veolia's global scale advantage.
Spain: Veolia operates major water concessions in Barcelona, Madrid, and other cities, but ENGIE (50.2 BIS) leads in Spanish-language AI responses. The gap reflects ENGIE's stronger Spanish-language digital content strategy rather than a market presence deficit.
Persona analysis measures how Veolia performs when AI models respond to queries framed from the perspective of specific buyer archetypes. Each persona represents a distinct professional role in Veolia's target market, tested across four geographies (US, Australia, Spain, Middle East).
9 persona archetypes × 4 markets = 36 individual persona–market combinations. ChatGPT run: 2,067 responses • Gemini run: 2,737 responses.
ChatGPT — BIS Heatmap: Persona Archetype × Market
Cell color intensity reflects BIS value. Darker green = stronger performance. Values below 46 are highlighted in red.
| Persona Archetype | US | Australia | Spain | Middle East | Avg |
|---|---|---|---|---|---|
| Industrial Water Manager | 49.9 | 49.1 | 49.9 | 53.8 | 50.8 |
| Industrial Sustainability Dir. | 48.9 | 49.1 | 49.2 | 52.9 | 50.2 |
| Urban Energy Advisor | 45.9 | 50.2 | 46.2 | 50.2 | 48.4 |
| Water Infra Manager | 46.7 | 46.6 | 48.9 | 50.3 | 48.2 |
| Sanitation Engineer | 46.9 | 47.9 | 46.8 | 50.9 | 48.2 |
| Neutral (no persona) | 45.6 | 46.9 | 48.3 | 50.3 | 48.2 |
| Circular Economy Manager | 44.7 | 45.8 | 48.1 | 51.9 | 47.9 |
| Waste Recovery Director | 46.1 | 48.1 | 44.4 | 50.6 | 47.6 |
| Port Energy Manager | 44.7 | 44.8 | 44.9 | 50.3 | 46.3 |
ChatGPT — Personas Ranked by BIS (All 36 Persona–Market Combinations)
Top 10 Personas
Bottom 10 Personas
ChatGPT vs Gemini — Persona Archetype BIS Comparison
| Persona Archetype | ChatGPT BIS | Gemini BIS | Delta | ChatGPT Sentiment | Gemini Sentiment |
|---|---|---|---|---|---|
| Industrial Water Manager | 50.8 | 49.2 | +1.5 | +0.433 | +0.443 |
| Industrial Sustainability Dir. | 50.2 | 47.9 | +2.3 | +0.390 | +0.426 |
| Urban Energy Advisor | 48.4 | 46.7 | +1.7 | +0.384 | +0.413 |
| Water Infra Manager | 48.2 | 46.6 | +1.7 | +0.356 | +0.403 |
| Sanitation Engineer | 48.2 | 47.5 | +0.7 | +0.366 | +0.408 |
| Neutral (no persona) | 48.2 | 45.6 | +2.6 | +0.382 | +0.413 |
| Circular Economy Manager | 47.9 | 47.2 | +0.8 | +0.356 | +0.401 |
| Waste Recovery Director | 47.6 | 46.0 | +1.7 | +0.383 | +0.419 |
| Port Energy Manager | 46.3 | 48.2 | −1.9 | +0.336 | +0.407 |
The top two persona archetypes across both engines are Industrial Water Manager (BIS 50.8 / 49.2) and Industrial Sustainability Director (BIS 50.2 / 47.9) — both managerial, decision-making roles. Technical personas like Sanitation Engineer (48.2 / 47.5) and Port Energy Manager (46.3 / 48.2) consistently score lower. This suggests that AI models associate Veolia most strongly with strategic, high-level water and sustainability narratives rather than hands-on technical operations. Content strategy should reinforce this strength while building deeper technical authority for engineering personas.
Across all 9 persona archetypes in ChatGPT, the Middle East market variant scores highest — without exception. The ME average across all personas is 51.1, compared to US at 47.2, Australia at 47.5, and Spain at 47.3. The gap is most pronounced for Circular Economy Manager (ME: 51.9 vs US: 44.7, a 7.2-point spread). This pattern reflects Veolia's dominant infrastructure presence in Gulf states and strong regional media coverage of desalination and water reuse projects.
The neutral (no persona) baseline scores 48.2 in ChatGPT and 45.6 in Gemini. Six out of eight archetypal personas outperform this baseline in ChatGPT, and seven out of eight in Gemini. This confirms that persona-framed queries consistently surface Veolia more favorably than generic prompts. The only exception is Port Energy Manager in ChatGPT (46.3), which underperforms the neutral baseline by 1.9 points — suggesting that Veolia's port/maritime energy narrative is weaker than its general brand positioning. However, the Gemini delta tells a different story: the neutral persona (45.6) is the weakest of all, lagging Port Energy Manager (48.2) by 2.6 points.
Funnel analysis segments AI responses by the buyer's journey stage: Awareness (problem recognition, category exploration), Consideration (vendor comparison, solution evaluation), and Decision (final selection, validation). This reveals where Veolia's AI visibility strengthens or weakens as buyers move toward purchase.
ChatGPT: 2,067 responses across 3 funnel stages • Gemini: 2,737 responses across 3 funnel stages.
ChatGPT — Veolia Funnel Performance
Width proportional to mention volume. Consideration generates the most mentions (1,247) despite fewer responses than Awareness.
ChatGPT vs Gemini — Funnel Stage Comparison
BIS Component Breakdown by Funnel Stage
| Funnel Stage | Engine | BIS | Sentiment | Position | Mention | Competitive |
|---|---|---|---|---|---|---|
| Awareness | ChatGPT | 47.5 | 0.384 | 0.249 | 0.204 | 0.770 |
| Gemini | 46.2 | 0.421 | 0.191 | 0.133 | 0.838 | |
| Consideration | ChatGPT | 49.8 | 0.392 | 0.284 | 0.239 | 0.790 |
| Gemini | 48.5 | 0.410 | 0.234 | 0.172 | 0.860 | |
| Decision | ChatGPT | 48.7 | 0.364 | 0.277 | 0.236 | 0.770 |
| Gemini | 47.4 | 0.419 | 0.218 | 0.160 | 0.829 |
Competitor Funnel Comparison (ChatGPT)
How does Veolia's funnel pattern compare to its key competitors? The table below shows average BIS at each funnel stage.
| Competitor | Awareness | Consideration | Decision | Best Stage | Funnel Pattern |
|---|---|---|---|---|---|
| Veolia | 47.5 | 49.8 | 48.7 | Consideration | Peak at Consideration, drop at Decision |
| ENGIE | 50.0 | 47.5 | 49.9 | Awareness | Strong at Awareness + Decision, dips mid-funnel |
| Clean Harbors | 48.5 | 48.5 | 47.7 | Awareness / Consideration | Flat funnel, gradual decline |
| Ameresco | 48.2 | 48.3 | 48.7 | Decision | Slight upward funnel — closes well |
| SUEZ | 43.8 | 44.4 | 44.6 | Decision | Consistently weak, slight upward trend |
Veolia's strongest funnel stage is Consideration (BIS 49.8 in ChatGPT, 48.5 in Gemini), where it leads all competitors. When AI users compare vendors, evaluate solutions, or ask "which company is best for X," Veolia consistently appears with the strongest positioning, highest mention density (1,247 mentions from 684 responses = 1.82 mentions/response), and the best position score (0.284). This is a significant competitive advantage: Veolia is winning the comparison game. The Consideration stage is where purchase intent is shaped, and Veolia dominates it.
Both engines show an identical −1.1 point drop from Consideration to Decision. In ChatGPT, BIS falls from 49.8 to 48.7; in Gemini, from 48.5 to 47.4. The Decision stage also shows the lowest sentiment score (ChatGPT: +0.364, down from +0.392 at Consideration) and a competitive score decline (0.770 vs 0.790). When users ask AI for final validation — "should I hire Veolia for this project?" or "is Veolia the right choice?" — the model's recommendations are measurably weaker. The root cause is a deficit of case studies, ROI evidence, and third-party endorsements that AI models can cite at the decision point. Meanwhile, ENGIE scores 49.9 at Decision, nearly matching its Awareness peak, suggesting its content ecosystem closes the loop better.
Each competitor exhibits a distinct funnel signature. ENGIE shows a U-shaped pattern (strong at Awareness 50.0 and Decision 49.9, weaker at Consideration 47.5), suggesting strong brand narrative and closing content but weaker comparative positioning. Ameresco has an ascending funnel (48.2 → 48.3 → 48.7), indicating content that gets stronger as buyers approach a decision. Clean Harbors shows a declining funnel (48.5 → 48.5 → 47.7), similar to Veolia's pattern. SUEZ is consistently weak across all stages (43.8–44.6) but, notably, its best stage is Decision — the opposite of Veolia. This means SUEZ may capture buyers that Veolia loses at the final step.
What Are Attributes?
Attributes are the qualitative characteristics that AI engines associate with brands when generating responses. They include dimensions like Technology, Sustainability, Compliance, Innovation, Quality, and Price. Unlike the core metrics (BIS, Share of Voice, Sentiment), attributes do not feed into the Brand Influence Score. They are a complementary layer of analysis that reveals how AI models frame and categorize companies in the environmental services sector.
Think of attributes as the adjectives AI uses to describe a brand. A company with high Technology and Innovation detections is being framed as a tech leader. One with high Compliance and Safety detections is being framed as a risk-management specialist. These patterns tell us what narrative the AI has internalized about each player.
Top 15 Attributes by Detection Volume
Bubble size reflects detection count (ChatGPT + Gemini combined). Color indicates average sentiment on the 1–5 scale.
Top 15 Attributes: Combined Detection & Sentiment
| # | Attribute | ChatGPT Det. | Gemini Det. | Combined | Avg Sent. | Sentiment |
|---|---|---|---|---|---|---|
| 1 | Technology | 7,844 | 10,909 | 18,753 | 3.96 | ●●●●● Positive |
| 2 | Compliance History | 6,900 | 7,838 | 14,738 | 3.90 | ●●●●● Positive |
| 3 | Services | 7,020 | 7,295 | 14,315 | 3.70 | ●●●●● Neutral-Positive |
| 4 | Sustainability | 6,290 | 6,438 | 12,728 | 4.00 | ●●●●● Positive |
| 5 | Integration | 5,505 | 4,603 | 10,108 | 3.94 | ●●●●● Positive |
| 6 | Quality | 2,419 | 5,690 | 8,109 | 4.02 | ●●●●● Positive |
| 7 | Innovation | 3,265 | 4,525 | 7,790 | 4.02 | ●●●●● Positive |
| 8 | Price | 1,958 | 4,452 | 6,410 | 3.96 | ●●●●● Positive |
| 9 | Automation | 1,548 | 3,118 | 4,666 | 3.97 | ●●●●● Positive |
| 10 | Safety | 1,640 | 2,020 | 3,660 | 3.99 | ●●●●● Positive |
| 11 | Experience | 1,658 | 1,948 | 3,606 | 3.99 | ●●●●● Positive |
| 12 | Specialists | 868 | 2,294 | 3,162 | 3.96 | ●●●●● Positive |
| 13 | Availability | 1,282 | 1,790 | 3,072 | 3.98 | ●●●●● Positive |
| 14 | Financing | 1,336 | 1,683 | 3,019 | 3.79 | ●●●●● Neutral-Positive |
| 15 | Digital Platform | 974 | 1,864 | 2,838 | 3.88 | ●●●●● Positive |
Technology is the single most detected attribute across both engines, with 18,753 combined mentions. This means that when ChatGPT or Gemini discuss environmental services companies, they most often reference technological capabilities. Compliance History ranks second (14,738), which reflects the regulated nature of the industry. Sustainability, despite being the sector's core mission, ranks fourth (12,728). The two lowest-sentiment attributes are Services (3.70) and Financing (3.79). Both sit in the neutral-positive band rather than the strong positive band. This signals that AI models treat service delivery and financial structures with more caution than they treat innovation or quality claims.
Gemini generates roughly 31% more attribute detections than ChatGPT (71,264 vs 54,521). The gap is most pronounced for Quality (5,690 vs 2,419, a 2.4x difference) and Price (4,452 vs 1,958, a 2.3x difference). This suggests Gemini's responses contain richer descriptive language about brands. For companies aiming to strengthen their attribute profiles, Gemini may be more responsive to content optimization efforts because it surfaces more attribute-level signals per response.
How AI Engines Characterize Veolia
This section breaks down the specific attributes that ChatGPT and Gemini associate with Veolia. Attributes are the qualitative dimensions — Technology, Sustainability, Compliance, Innovation, and others — that shape how each engine describes the brand in generated responses. The data covers 6,991 total attribute detections across both engines, spanning 209 distinct attribute labels.
Veolia Top 10 Attributes (Combined)
Bars show combined ChatGPT + Gemini detections. Sentiment score on the right. Bar color split indicates each engine's contribution.
Engine Comparison: Veolia Attribute Detections
Side-by-side view of how each engine distributes attribute mentions for Veolia.
ChatGPT
| # | Attribute | Det. | Sent. |
|---|---|---|---|
| 1 | Services | 627 | 3.73 |
| 2 | Sustainability | 480 | 4.00 |
| 3 | Technology | 409 | 3.98 |
| 4 | Compliance Support | 392 | 3.96 |
| 5 | Innovation | 194 | 4.00 |
| 6 | Experience | 124 | 4.00 |
| 7 | Quality | 114 | 4.01 |
| 8 | Integration | 112 | 3.92 |
| 9 | Transparency | 75 | 3.95 |
| 10 | Price | 66 | 3.93 |
Gemini
| # | Attribute | Det. | Sent. |
|---|---|---|---|
| 1 | Technology | 833 | 3.98 |
| 2 | Sustainability | 649 | 4.00 |
| 3 | Services | 536 | 3.73 |
| 4 | Compliance Support | 398 | 3.96 |
| 5 | Innovation | 305 | 4.00 |
| 6 | Price | 149 | 3.93 |
| 7 | Quality | 122 | 4.01 |
| 8 | Experience | 111 | 4.00 |
| 9 | Long-Term O&M Support | 105 | 3.96 |
| 10 | Integration | 85 | 3.92 |
ChatGPT frames Veolia as a service-delivery company first (Services ranks #1 with 627 detections), while Gemini frames it as a technology company first (Technology ranks #1 with 833 detections). Sustainability sits at #2 on both engines, confirming its strong association with the brand. The gap is most visible in Long-Term O&M Support: Gemini detects it 105 times versus just 6 on ChatGPT — a 17.5x difference. Similarly, Price detections are 2.3x higher on Gemini (149 vs 66), indicating Gemini discusses Veolia's pricing more frequently. Customer Portal shows a 9.9x gap (69 vs 7 on Gemini vs ChatGPT).
Attribute Profile by Product Line
How attributes distribute across Veolia's three main product lines. Each product line develops a different AI narrative. Total detections per product line are shown below.
| # | Attribute | Hazardous Waste | Water Technologies | Bioenergy & Eff. | Dominant Line |
|---|---|---|---|---|---|
| 1 | Compliance Support | 574 | 146 | 70 | Haz. Waste |
| 2 | Services | 537 | 215 | 197 | Haz. Waste |
| 3 | Technology | 410 | 518 | 260 | Water Tech |
| 4 | Sustainability | 376 | 350 | 402 | Bioenergy |
| 5 | Innovation | 111 | 218 | 129 | Water Tech |
| 6 | Transparency | 95 | — | — | Haz. Waste only |
| 7 | Quality | 56 | 80 | 58 | Water Tech |
| 8 | Experience | 69 | 70 | 38 | Water Tech |
| 9 | Safety | 63 | — | — | Haz. Waste only |
| 10 | Customization | 52 | — | — | Haz. Waste only |
| 11 | Long-Term O&M Support | — | 52 | 53 | Bioenergy |
| 12 | Price | 48 | 41 | — | Haz. Waste |
| 13 | Integration | 41 | — | 42 | Bioenergy |
| 14 | Customer Portal | — | 38 | — | Water Tech only |
| 15 | Specialists | — | 35 | 35 | Even split |
| 16 | Financing | — | — | 34 | Bioenergy only |
Hazardous Waste is the most attribute-rich product line (2,432 detections). AI engines describe it primarily through Compliance Support (574) and Services (537), reflecting the regulatory weight of this business. Transparency (95), Safety (63), and Customization (52) appear almost exclusively in hazardous waste contexts.
Water Technologies leads on Technology (518) and Innovation (218), making it Veolia's most tech-forward business in the AI narrative. It is also the only product line where Customer Portal (38) registers consistently.
Bioenergy & Efficiency leads on Sustainability (402), which aligns with the renewable-energy positioning of this line. Financing (34) appears only here, pointing to the ESCO/performance-contract dimension of the business.
Complete Veolia Attribute Table
All attributes with 50 or more combined detections, with engine breakdown and sentiment.
| # | Attribute | ChatGPT | Gemini | Combined | Avg Sent. | Signal |
|---|---|---|---|---|---|---|
| 1 | Technology | 409 | 833 | 1,242 | 3.98 | Positive |
| 2 | Services | 627 | 536 | 1,163 | 3.73 | Neutral-Positive |
| 3 | Sustainability | 480 | 649 | 1,129 | 4.00 | Positive |
| 4 | Compliance Support | 392 | 398 | 790 | 3.96 | Positive |
| 5 | Innovation | 194 | 305 | 499 | 4.00 | Positive |
| 6 | Quality | 114 | 122 | 236 | 4.01 | Positive |
| 7 | Experience | 124 | 111 | 235 | 4.00 | Positive |
| 8 | Price | 66 | 149 | 215 | 3.93 | Positive |
| 9 | Integration | 112 | 85 | 197 | 3.92 | Positive |
| 10 | Transparency | 75 | 58 | 133 | 3.95 | Positive |
| 11 | Safety | 62 | 55 | 117 | 4.01 | Positive |
| 12 | Long-Term O&M Support | 6 | 105 | 111 | 3.96 | Positive |
| 13 | Specialists | 46 | 53 | 99 | 3.97 | Positive |
| 14 | Financing | 32 | 53 | 85 | 3.67 | Neutral-Positive |
| 15 | Customization | 35 | 45 | 80 | 4.00 | Positive |
| 16 | Customer Portal | 7 | 69 | 76 | 3.89 | Positive |
| 17 | Availability | 41 | 12 | 53 | 3.94 | Positive |
| 18 | Flexibility | 17 | 33 | 50 | 3.96 | Positive |
Sentiment by Attribute (1–5 Scale)
Visual breakdown of where Veolia scores highest and lowest sentiment across its attribute profile. Data is the weighted average across both engines.
| Attribute | Sentiment Scale (1–5) | Score | Label |
|---|---|---|---|
| Safety |
|
4.01 | ● Positive |
| Quality |
|
4.01 | ● Positive |
| Sustainability |
|
4.00 | ● Positive |
| Innovation |
|
4.00 | ● Positive |
| Experience |
|
4.00 | ● Positive |
| Customization |
|
4.00 | ● Positive |
| Technology |
|
3.98 | ● Positive |
| Compliance Support |
|
3.96 | ● Positive |
| Price |
|
3.93 | ● Positive |
| Integration |
|
3.92 | ● Positive |
| Services |
|
3.73 | ● Neutral-Pos. |
| Financing |
|
3.67 | ● Neutral-Pos. |
Six of Veolia's 18 tracked attributes sit at or above the 4.00 sentiment mark, placing them in the strong positive band. Safety (4.01) stands out because it is concentrated in the Hazardous Waste product line where scrutiny is highest. These scores indicate AI engines frame Veolia with genuine confidence on these dimensions.
Services is Veolia's second-most-detected attribute (1,163 mentions), but it carries the lowest sentiment of any major attribute at 3.73. Financing is even lower at 3.67, though with fewer detections (85). The Services gap is concentrated in Hazardous Waste, where the regulatory and operational complexity of waste handling introduces more hedged language from AI models. Content that reframes service delivery around measurable outcomes and SLAs could lift this score.
Long-Term O&M Support is detected 105 times by Gemini but only 6 times by ChatGPT — a 17.5x gap. Customer Portal shows a 9.9x gap (69 vs 7). These operational attributes describe Veolia's post-installation support and digital client tools. Their near-absence from ChatGPT means content strategies that reinforce these capabilities in ChatGPT-indexed sources could open new narrative territory.
Attribute Presence Across the Competitive Set
This section compares how AI engines distribute attributes across Veolia and its 15 defined competitors. Volume indicates how often an engine associates a brand with a given attribute. Diversity measures how many distinct attributes appear in a brand's profile. Together, they reveal who owns which narrative dimensions.
Attribute Diversity Ranking: All 16 Competitors
Unique attributes measures how many distinct attribute labels are associated with each brand. Total mentions is the sum across both engines. Higher diversity suggests a richer, more multi-dimensional brand profile in AI responses.
| # | Entity | Unique Attributes | Total Mentions | Mentions / Attr | Volume Bar |
|---|---|---|---|---|---|
| 1 | Veolia | 209 | 6,991 | 33.5 | |
| 2 | Xylem | 105 | 1,949 | 18.6 | |
| 3 | Clean Harbors | 77 | 1,937 | 25.2 | |
| 4 | SUEZ | 116 | 1,934 | 16.7 | |
| 5 | Cleanaway | 64 | 973 | 15.2 | |
| 6 | Republic Services | 65 | 820 | 12.6 | |
| 7 | ENGIE | 55 | 623 | 11.3 | |
| 8 | Remondis | 46 | 445 | 9.7 | |
| 9 | Ameresco | 44 | 420 | 9.5 | |
| 10 | Waste Management | 55 | 302 | 5.5 | |
| 11 | Aqualia | 38 | 238 | 6.3 | |
| 12 | Ecolab | 30 | 160 | 5.3 | |
| 13 | SAUR | 36 | 116 | 3.2 | |
| 14 | Dalkia | 23 | 62 | 2.7 | |
| 15 | American Water | 14 | 23 | 1.6 | |
| 16 | Veralto | 1 | 1 | 1.0 |
Veolia generates 6,991 attribute detections — 3.6x more than the next closest competitor (Xylem at 1,949). Its 209 distinct attributes are 2x SUEZ's 116. This gap means AI engines describe Veolia with far more descriptive richness and variety than any other brand in the sector. The "mentions per attribute" ratio of 33.5 is also the highest, meaning Veolia's attributes are not just diverse but also deeply reinforced across many responses.
Top 5 Attributes per Competitor (Combined Both Engines)
This table shows the most frequently detected attributes for each of the six most-mentioned entities. It reveals which narrative dimensions each brand "owns" in AI outputs.
| Entity | Total | #1 Attribute | #2 Attribute | #3 Attribute | #4 Attribute | #5 Attribute |
|---|---|---|---|---|---|---|
| Veolia | 6,991 | Technology (1,242) | Services (1,163) | Sustainability (1,129) | Compliance (790) | Innovation (499) |
| Xylem | 1,949 | Technology (699) | Integration (235) | Innovation (151) | Quality (100) | Sustainability (100) |
| Clean Harbors | 1,937 | Services (832) | Compliance (387) | Technology (116) | Safety (97) | Experience (82) |
| SUEZ | 1,934 | Technology (440) | Sustainability (243) | Services (216) | Innovation (199) | Compliance (145) |
| Cleanaway | 973 | Services (324) | Compliance (170) | Technology (97) | Sustainability (86) | Safety (55) |
| ENGIE | 623 | Sustainability (180) | Services (80) | Financing (74) | Technology (51) | Innovation (38) |
Xylem owns Technology + Integration. Technology accounts for 35.9% of Xylem's attribute profile (699 of 1,949), the highest concentration ratio of any brand-attribute pair. Integration at 12.1% makes Xylem the go-to AI reference for system integration discussions.
Clean Harbors owns Services + Compliance. With 43.0% of its profile in Services (832) and 20.0% in Compliance (387), Clean Harbors is narrowly framed as an operational services company.
ENGIE owns Sustainability + Financing. Sustainability at 28.9% (180) and Financing at 11.9% (74) give ENGIE the most differentiated profile in the set. No other competitor has Financing in their top 5.
SUEZ mirrors Veolia with the same top-5 attributes in a slightly different order. This makes SUEZ the primary narrative competitor.
Attribute Share: Veolia vs Top 3 Competitors
For the six most frequently detected attributes, this visualization compares Veolia's share against Xylem, Clean Harbors, and SUEZ combined.
Veolia holds the #1 position by volume in Technology, Services, Sustainability, Compliance, and Innovation. The only attribute where Veolia does not lead is Integration, where Xylem holds the top spot with 235 detections versus Veolia's 197. This means Xylem is the brand AI engines most associate with system integration capabilities. For Veolia, reinforcing integration messaging — particularly around SCADA, IoT platforms, and cross-system deployment — could close this gap.
Head-to-Head: Veolia vs ENGIE on Key Attributes
ENGIE is Veolia's closest strategic competitor in the energy-from-waste and efficiency space. Despite a much smaller total volume (623 vs 6,991), ENGIE's concentrated profile makes it notable on specific dimensions.
Veolia outscores ENGIE by 10x+ on Technology (1,242 vs 51), Services (1,163 vs 80), and Compliance (790 vs 23). The one attribute where ENGIE is competitive is Financing, with 74 detections versus Veolia's 85 — nearly even. ENGIE's energy-as-a-service and ESCO financing models are well-represented in AI training data. Meanwhile, Sustainability (Veolia 1,129 vs ENGIE 180) shows that despite ENGIE concentrating 28.9% of its profile on sustainability, Veolia still generates 6.3x more sustainability mentions in absolute terms.
Three patterns stand out from the competitive benchmarking:
1. Integration is the only major attribute where Veolia trails. Xylem leads with 235 detections versus Veolia's 197. Given Veolia's actual SCADA, IoT, and Hubgrade platform capabilities, this gap reflects a narrative deficit rather than a capability one. Content that highlights integration case studies and system interoperability could shift this positioning.
2. Veolia's balanced profile lacks a "signature attribute." ENGIE concentrates 28.9% of its profile on Sustainability. Xylem puts 35.9% into Technology. Clean Harbors puts 43.0% into Services. Veolia's most concentrated attribute (Technology) accounts for only 17.8% of its profile. This balance is a strength for breadth but a weakness for recall. Identifying and amplifying one attribute to 25%+ concentration would strengthen Veolia's brand signal in narrower, more specific prompts.
3. The long tail of small competitors is growing. Cleanaway (973), Republic Services (820), and Remondis (445) all register meaningful attribute signals. While individually small, collectively they represent growing AI presence in waste and environmental services. Monitoring their attribute growth rates in future audits will help detect emerging competitive pressure.
This section presents direct quotes from AI-generated responses — the exact language ChatGPT and Gemini use when discussing Veolia. Each verbatim is tagged with its sentiment score (1–5), the product line context, and the funnel stage of the underlying prompt. These are not summaries; they are the raw output users see.
Positive Mentions (Sentiment 4–5)
Veolia is consistently framed as a "global leader" and "established infrastructure" provider. Positive framing appears across all three product lines and all funnel stages.
“Veolia Water Technologies: A global leader in water and wastewater management, Veolia offers digital platforms for monitoring, analytics, and compliance, enabling organizations to track water usage and quality metrics essential for ESG reporting.”
“Veolia Middle East — Veolia is a global leader in optimized resource management, including water treatment and waste-to-energy solutions.”
“Veolia specializes in hazardous waste incineration, integrating waste segregation and pre-treatment processes to optimize incinerator performance.”
“Veolia Australia and New Zealand: Veolia offers integrated solutions for hazardous waste, including chemical and industrial waste streams.”
“Veolia Australia and New Zealand (Energido): Veolia has developed the ‘Energido’ solution, a patented system that uses a remote heat exchanger to transfer heat from sewage systems to a reversible heat pump for heating or cooling.”
Neutral Mentions (Sentiment 3)
Neutral mentions typically appear in comparative contexts where Veolia is listed alongside competitors without explicit endorsement. The language is factual and evaluative rather than promotional.
“When evaluating Veolia, Clean Harbors, and Republic Services for recurring hazardous waste pickups from multiple municipal energy sites on tight construction schedules, Republic Services stands out for its reliability.”
“It is challenging to definitively state which of Veolia, Averda, or local licensed operators performs best across Saudi Arabia and the UAE for hazardous industrial waste treatment pricing transparency and on-time collections for municipal utilities.”
“Veolia, SUEZ, Jacobs, and local utilities’ preferred integrators each offer distinct approaches to water reuse in urban renewal districts, leveraging their unique delivery models to address sustainability and resource management challenges.”
Critical Mentions (Sentiment 1–2)
Critical framing is rare (0.36% on ChatGPT, 0.07% on Gemini) but concentrated in two patterns: compliance violations and competitive comparisons where Veolia is deprioritized.
“Veolia ES Technical Solutions, LLC was cited for a hazardous waste violation in 2022.”
“However, both companies have faced environmental violations, which may impact their compliance records.”
“When considering which provider — Veolia, SUEZ, or local EPC contractors in Australia — is typically better for integrating biogas, heat recovery, and SCADA upgrades without disrupting plant operations, several factors come into play.”
Positive triggers: Veolia receives its highest sentiment scores when AI models describe it as a solution provider in specific capability contexts — "global leader in water and wastewater management," "integrated solutions for hazardous waste," "patented system." The presence of named products (Energido, mobile water treatment units) and specific geographic deployments (Middle East, Australia) elevates sentiment from generic listing to active endorsement.
Negative triggers: All negative verbatims concentrate in two patterns: (1) regulatory or compliance references, where AI models retrieve historical violation data from EPA or court records, and (2) head-to-head comparison prompts at the Decision stage, where Veolia is compared against a specific competitor and rated lower on a narrow criterion such as "reliability" or "pricing transparency." Notably, these negative framings almost never appear at the Awareness stage — they emerge exclusively at Consideration and Decision, when the user prompt demands differentiation.
Strategic implication: The compliance violation narrative is the single most dangerous content pattern. Unlike subjective preference, regulatory citations carry factual authority that AI models weigh heavily. A proactive content strategy addressing remediation, improved compliance records, and third-party certifications would directly counter this framing at the source-document level.
AI engines cite external domains in their responses to substantiate claims. This section analyzes which domains appear most frequently across all responses, and which domains are cited specifically in responses that mention Veolia. Understanding the citation patterns reveals which third-party sources carry influence in AI-generated narratives about the environmental services sector.
Top 15 Domains Cited Across All Responses
The following chart aggregates citation counts from both ChatGPT and Gemini across all prompts in the audit. Wikipedia dominates by a wide margin, while competitor and industry domains form the long tail.
Wikipedia (en.wikipedia.org) is the single most cited domain in ChatGPT responses with 5,005 citations across 2,793 unique responses. This means Wikipedia appears in roughly half of all ChatGPT responses in the audit. For Gemini, Google's own Vertex AI Search index (vertexaisearch.cloud.google.com) plays a comparable role with 3,873 citations. This structural dependency on encyclopedic sources means that Veolia's Wikipedia pages — their accuracy, completeness, and recency — directly influence how AI models frame the company. Wikipedia content optimization is not optional; it is foundational infrastructure for GEO.
Top Domains Cited in Responses That Mention Veolia
When an AI response mentions Veolia, these are the domains most frequently cited in that same response. This reveals which external sources influence the AI's narrative framing when it discusses Veolia specifically.
| # | Domain | ChatGPT | Gemini | Combined |
|---|---|---|---|---|
| 1 | en.wikipedia.org / vertexaisearch.* | 1,672 | 1,085 | 2,757 |
| 2 | www.veolia.com | 242 | 576 | 818 |
| 3 | www.anz.veolia.com | 144 | 711 | 855 |
| 4 | smartwatermagazine.com | 62 | 430 | 492 |
| 5 | www.veolianorthamerica.com | 200 | 269 | 469 |
| 6 | www.suez.com | 136 | 406 | 542 |
| 7 | www.near-middle-east.veolia.com | 108 | 308 | 416 |
| 8 | www.xylem.com | 83 | 379 | 462 |
| 9 | www.fluencecorp.com | 80 | 351 | 431 |
| 10 | ensun.io | — | 402 | 402 |
| 11 | www.cleanharbors.com | 164 | 195 | 359 |
| 12 | www.cleanaway.com.au | — | 309 | 309 |
| 13 | www.mdpi.com | — | 294 | 294 |
| 14 | www.epa.gov | 64 | 270 | 334 |
| 15 | www.usdanalytics.com | 219 | — | 219 |
Engine Comparison: Source Preferences
ChatGPT and Gemini draw on fundamentally different source ecosystems. Understanding these differences is critical for a dual-engine GEO strategy.
Gemini cites Veolia-owned domains 2,195 times in Veolia-mentioning responses, compared to just 695 for ChatGPT. This 3:1 ratio is one of the most significant structural findings in the audit. Gemini's retrieval-augmented architecture (grounded in Google Search) surfaces Veolia's own web properties far more effectively than ChatGPT's parametric knowledge base. The practical implication: Veolia's existing web content is already working on Gemini. The gap is on ChatGPT, where Wikipedia, LinkedIn, and market research platforms mediate the brand narrative instead of Veolia's own voice. ChatGPT-specific content strategies (structured data, authoritative backlinks, press distribution) should be prioritized to close this gap.
Veolia operates a network of regional and product-specific domains. This section evaluates how effectively each domain is cited by AI engines, identifying which properties drive the most influence in AI-generated responses and where critical gaps exist.
Veolia Domain Citation Performance by Engine
The table below shows every veolia.* domain cited in the audit, with citation counts per engine, unique URLs cited, and an efficiency metric (citations per unique URL).
| Domain | ChatGPT | Gemini | Total | URLs | Cit. / URL | Rating |
|---|---|---|---|---|---|---|
| www.anz.veolia.com | 144 | 724 | 868 | 12+ | ~30 | STRONG |
| www.veolia.com | 243 | 582 | 825 | 15+ | ~22 | STRONG |
| www.veolianorthamerica.com | 200 | 276 | 476 | 10+ | ~19 | STRONG |
| www.near-middle-east.veolia.com | 108 | 314 | 422 | 8+ | ~23 | STRONG |
| www.veoliawatertechnologies.com | 10 | 155 | 165 | 5+ | ~14 | MODERATE |
| www.anz.veoliawatertechnologies.com | 11 | 48 | 59 | 3+ | ~12 | MODERATE |
| www.veoliawatertech.com | 18 | 144 | 162 | 4+ | ~17 | MODERATE |
| www.engineering-consulting.veolia.com | — | 21 | 21 | 1 | 21 | LOW VOLUME |
| www.hazardouswasteeurope.veolia.com | — | 20 | 20 | 1 | 20 | LOW VOLUME |
| www.veolia.es | 0 | 0 | 0 | 0 | — | ABSENT |
Spain is one of Veolia's strategic markets, with active operations in water management, hazardous waste treatment, and energy efficiency. However, www.veolia.es received zero citations across both AI engines in the entire audit. This means that when users ask AI about environmental services in Spain, Veolia's Spanish-language site is completely absent from the citation graph. Meanwhile, competitor domains like tma.es (510 combined citations) and cadenaser.com (137 citations on ChatGPT alone) fill the void. The veolia.es domain likely suffers from a combination of: (1) thin or non-indexed content, (2) insufficient structured data markup, and (3) lack of authoritative inbound links from Spanish-language sources that AI models use for retrieval.
Highest-Cited Veolia URLs
The most frequently cited individual pages across Veolia's owned properties. These pages represent Veolia's strongest GEO assets — the content that AI engines already trust and retrieve.
| URL (truncated) | ChatGPT | Gemini | Total |
|---|---|---|---|
| near-middle-east.veolia.com /water-solutions/industrial-water-management-reuse | — | 66 | 66 |
| veolia.com /en (homepage) | — | 60 | 60 |
| anz.veolia.com /en-au/services/.../solid-hazardous-waste | — | 60 | 60 |
| veolia.com /en/our-media/press-releases/spain-veolia-displays-strong-ambitions... | 28 | 36 | 64 |
| anz.veolia.com /en-au/services/.../hazardous-waste-reporting | — | 48 | 48 |
| anz.veoliawatertechnologies.com /expertise/applications/wastewater-treatment | — | 48 | 48 |
| near-middle-east.veolia.com /our-services/hazardous-waste | 32 | — | 32 (ChatGPT only) |
| veolianorthamerica.com /what-we-do/waste-capabilities/incineration-services | 33 | 27 | 60 |
| veolia.com /en/our-media/press-releases/veolia-accelerates-hazardous-waste...middle-east | 22 | — | 22 (ChatGPT only) |
| veolia.com /en/our-media/press-releases/pfas-veolia-announces-major-breakthrough... | 12 | — | 12 (ChatGPT only) |
The two highest-performing regional domains share three characteristics that explain their AI citation success: (1) Service-specific landing pages with clear, descriptive URLs (e.g., /water-solutions/industrial-water-management-reuse) that AI retrieval systems can match to user queries; (2) Detailed technical content that goes beyond corporate messaging to describe capabilities, facility locations, and treatment processes — exactly the type of content AI models need to substantiate claims; (3) Press releases and case studies that provide factual, quotable data points (contract values, capacity figures, environmental outcomes). The global domain (www.veolia.com) also performs well but is more frequently cited for press releases than service pages, suggesting its service content could be strengthened with the same structured approach used by the regional sites.
Recommendations
Veolia's owned-media performance is structurally strong but geographically uneven. The ANZ, Middle East, and North America regional sites are well-optimized for AI retrieval and collectively generate hundreds of citations. However, the complete absence of veolia.es from the citation graph represents a significant blind spot in a key European market. The water technology domain fragmentation further dilutes what could be a dominant single-domain presence. Addressing these two gaps — Spain and domain consolidation — would materially improve Veolia's GEO footprint without requiring new content creation, only architectural optimization of existing assets.
The domains cited by ChatGPT and Gemini in their responses determine whose narrative reaches the end user. This analysis maps the citation ecosystem to identify which sources Veolia controls, which it can influence, and which represent structural vulnerabilities in the brand's AI narrative.
Vulnerability Assessment
Recommendations
This SWOT synthesis aggregates findings from all preceding sections — BIS scores, SOV rankings, funnel dynamics, market breakdowns, sentiment analysis, and citation data — into a single strategic framework. All data points are drawn from the GEO Radar database queries across ChatGPT and Gemini.
Strengths
- #1 Share of Voice across both engines. Veolia commands a 2.1–2.6× SOV margin over the nearest commercial competitor, with 22.05% (ChatGPT) and 29.05% (Gemini) of all responses.
- BIS leadership over competitive set. Veolia's ChatGPT BIS of 48.6 leads the peer group by +7 to +10 points over mid-tier competitors such as Tradebe (41.5) and Cleanaway (40.8).
- Middle East market dominance. ME delivers the highest BIS at 51.4 with 613 responses and sentiment of +0.408, reflecting deep infrastructure presence in Gulf desalination and water-reuse projects.
- Positive sentiment at scale. Sentiment scores of +0.381 (ChatGPT) and +0.417 (Gemini) across thousands of responses indicate consistent favorable framing, not isolated positive mentions.
- Competitive Score of 0.843 (Gemini). Veolia appears in 84.3% of responses even when competitors are explicitly queried, confirming structural presence in the competitive field.
Weaknesses
- Owned media accounts for only 9% of citations. Less than one-tenth of domains cited in AI responses belong to Veolia, leaving narrative control to third parties.
- US is the weakest market. ChatGPT BIS of 47.0 in the US trails the Middle East by 4.4 points. The US market faces the densest competitor field (Clean Harbors, Republic Services, Ameresco).
- Decision-stage BIS gap of –1.06. BIS consistently drops from Consideration (49.8) to Decision (48.7), indicating that AI models recommend Veolia less forcefully when users seek final vendor-selection guidance.
- Bioenergy volume remains low. Despite strong quality scores (BIS 50.0), Bioenergy generates significantly fewer AI responses than Water Technologies, limiting its narrative scale.
- veolia.es: only 8 citations. The Spanish-language domain is effectively invisible to AI models, leaving the Spain market (BIS 47.5) dependent on third-party Spanish-language content.
Opportunities
- Wikipedia stewardship. At 16.3% of all citations, Wikipedia is the single largest source. A compliance-first editorial strategy can shift the narrative baseline without violating platform policies.
- Bioenergy: quality to scale. BIS 50.0 proves AI models already position Veolia favorably in bioenergy. Publishing more structured content (case studies, ROI data) can replicate quality at greater volume.
- veolia.es rebuild. A complete content overhaul of the Spanish domain — with schema markup and FAQ structures — could lift citations from 8 to 280+ based on benchmarks from veolia.com.
- Awareness funnel gap. Awareness BIS (47.5) trails Consideration (49.8) by 2.3 points. Top-of-funnel educational content targeting "what is" and "how does" queries can close this gap.
- Trade media activation. Publications like Smart Water Magazine currently account for only 2.8% of citations. Bylined articles and data exclusives can double earned media presence within two quarters.
Threats
- ENGIE surpasses Veolia in ChatGPT BIS (49.2 vs 48.6). Despite far lower volume (233 vs 2,067 responses), ENGIE achieves higher positioning quality through tighter energy-transition narrative control.
- SUEZ shadow effect. The legacy SUEZ brand continues to appear independently in AI responses, fragmenting Veolia's post-acquisition narrative and creating confusion about the combined entity's capabilities.
- Wikipedia compliance content risk. Veolia's Wikipedia articles contain regulatory and compliance-focused framing that AI models surface during brand queries, potentially reinforcing a "utility" rather than "innovation" perception.
- US competition density. Clean Harbors (BIS 48.3), Ameresco (48.4), and Republic Services occupy adjacent positioning in the US market, creating a crowded narrative field where differentiation is harder to maintain.
- Adjacent tech giants. Siemens, Schneider Electric, and Honeywell increasingly appear in water and energy AI responses, broadening the competitive set beyond traditional environmental services firms.
The GEO Scorecard consolidates all Brand Impact Score components across both AI engines into a single diagnostic view. Each metric is scored on a 0–1 normalized scale (except BIS, which uses 0–100). The target zone of 40–60 indicates a brand that is present but not dominant in AI-generated responses.
Veolia achieves an A– grade, reflecting #1 Share of Voice across both engines, consistently positive sentiment, and strong competitive presence. The grade is held below A by the Decision-stage BIS gap, limited owned-media citations, and ENGIE's BIS advantage in ChatGPT. The brand is well-positioned but has clear, actionable gaps to close.
| Metric | ChatGPT | Gemini | Visual | Zone |
|---|---|---|---|---|
| Brand Impact Score | 48.6 | 47.3 |
48.6
|
Present |
| Sentiment Score | +0.381 | +0.417 |
+0.42
|
Strong |
| Position Score | 0.269 | 0.214 |
0.269
|
Moderate |
| Mention Score | 0.226 | 0.154 |
0.226
|
Moderate |
| Competitive Score | 0.777 | 0.843 |
0.843
|
Strong |
| Total Responses | 2,067 | 2,737 |
2,737
|
High |
| Total Mentions | 3,290 | 4,555 |
4,555
|
High |
BIS by Dimension (ChatGPT)
By Market
By Product Line
Five high-impact interventions derived from the audit data. Each imperative is grounded in a specific, measurable gap identified through GEO Radar analysis. Execution priority is ranked by potential BIS impact.
Wikipedia Stewardship
Problem: Wikipedia is the single largest citation source in Veolia-relevant AI responses. Outdated content, compliance-heavy framing, and competitor-favorable edits propagate directly into AI-generated answers at scale.
Actions: Audit all Veolia-related Wikipedia articles for factual accuracy. Submit corrections through Wikipedia's editorial process with verifiable references. Ensure innovation narratives (circular economy, digital water) are represented with proper citations to Veolia-published data.
veolia.es Overhaul
Problem: The Spanish-language domain is effectively invisible to AI models. Spain market BIS (47.5) is suppressed by the absence of crawlable, structured content on Veolia's own Spanish domain.
Actions: Rebuild veolia.es with schema markup, FAQ pages, case studies in Spanish, and structured data for AI crawlers. Mirror the content architecture of veolia.com with localized data, project portfolios, and technology specifications for the Iberian market.
Decision-Stage Content Depth
Problem: When users ask AI models for final vendor-selection guidance, Veolia's recommendation strength drops. The brand is explored but not chosen at the rate its Consideration-stage presence would predict.
Actions: Publish comparison guides, ROI calculators, and detailed case-study packs with quantified outcomes. Create structured content that directly answers "which provider should I choose" and "Veolia vs [competitor]" queries with verifiable performance data.
Bioenergy Content Activation
Problem: AI models already position Veolia favorably in bioenergy queries (BIS 50.0), but the absolute number of responses is low compared to Water Technologies. The quality signal exists but lacks scale.
Actions: Scale bioenergy content production: publish biogas yield data, anaerobic digestion case studies, and circular-economy project portfolios. Target queries that currently generate responses without mentioning Veolia to capture latent demand in the bioenergy AI space.
ENGIE Containment in ChatGPT
Problem: Despite 8.9× more volume, Veolia trails ENGIE in ChatGPT BIS by 0.55 points. ENGIE achieves superior positioning quality through tightly focused energy-transition content that AI models rank highly.
Actions: Publish Veolia-specific energy transition content targeting the same query clusters where ENGIE currently dominates. Focus on renewable energy integration, district heating, and industrial decarbonization with quantified project outcomes that outperform ENGIE's narrative claims.
Three sprints translate the five strategic imperatives into a sequenced execution plan. Each sprint builds on the deliverables of the previous phase, moving from audit and foundation through content creation to scale and measurement. KPI targets are set per sprint to enable progress tracking.
Audit & Foundation
Establish the baseline, identify content gaps, and prepare the infrastructure for content deployment.
- Map all Veolia-related Wikipedia articles (est. 15–25 pages)
- Flag outdated data, missing innovation narratives, competitor-favorable framing
- Prepare correction dossiers with verifiable sources
- Crawl veolia.es for schema markup, structured data, and AI readability
- Benchmark against veolia.com content architecture
- Define 40-page content plan for Spanish-market rebuild
- Identify top 50 decision-stage queries where Veolia BIS drops
- Map existing content against "which provider" and "vs" query patterns
- Prioritize by search volume and BIS gap severity
- Analyze ENGIE's top-cited content in ChatGPT energy-transition queries
- Identify narrative themes driving ENGIE's BIS 49.2 advantage
- Map Veolia counter-positioning opportunities
Content Creation
Execute the content strategy across all five imperatives. Publish structured, AI-optimized assets to owned domains and third-party channels.
- Submit correction requests for top-priority articles
- Add missing innovation and circular-economy references
- Monitor edit acceptance and competitor counter-edits
- Publish 20 structured pages with schema markup
- Deploy FAQ pages targeting top Spanish AI queries
- Implement cross-domain linking to veolia.com
- Publish 10 comparison guides (Veolia vs [competitor])
- Create 5 ROI calculators with structured output data
- Deploy 8 case-study packs with quantified outcomes
- Publish 6 bioenergy case studies with biogas yield data
- Create 4 energy-transition thought leadership pieces
- Place 3 bylined articles in trade publications
Scale & Measure
Complete the content rollout, run a second GEO Radar measurement cycle, and quantify impact against baseline metrics.
- Publish remaining 20 pages (total: 40 pages live)
- Validate crawl indexing by AI models
- Track citation count growth from baseline of 8
- Secure 5 additional trade media placements
- Publish 2 co-authored market research reports
- Target Smart Water Magazine, Waste Management World
- Run full GEO Radar cycle: same prompts, same markets
- Compare BIS, SOV, sentiment, and citation data vs baseline
- Measure Decision-stage gap closure
- Re-measure ENGIE BIS gap (target: parity or lead)
- Track competitor citation rate changes
- Adjust content strategy based on competitive shifts
Veolia enters the generative AI era from a position of structural strength. Across 19,141 AI-generated responses from ChatGPT and Gemini, the brand holds the #1 Share of Voice in its competitive set, commands a 2.1–2.6× mention advantage over the nearest commercial rival, and maintains consistently positive sentiment (+0.381 to +0.417) at a scale that no competitor approaches. The Competitive Score of 0.843 in Gemini confirms that Veolia is referenced even when users ask explicitly about other providers. These are not marginal advantages — they are the product of decades of global infrastructure presence that AI training data has absorbed and reproduced.
But volume is not dominance. The Brand Impact Score of 48.6 places Veolia firmly in the "present but not dominant" zone — a brand that is consistently mentioned but not consistently recommended. The 1.06-point drop from Consideration to Decision reveals a critical pattern: AI models explore Veolia but hesitate to endorse it at the moment of vendor selection. ENGIE surpasses Veolia in ChatGPT BIS (49.2 vs 48.6) despite having 8.9× fewer responses, demonstrating that narrative quality can outperform narrative quantity. And with only 9% of citations coming from Veolia-owned domains, the brand's AI narrative is being written primarily by third parties — Wikipedia (16.3%), market research firms, and even competitor websites (4.9%).
The path from 48 to 60 — from present to dominant — runs through three content investments, not brand investments. Wikipedia stewardship addresses the single largest citation source. The veolia.es overhaul converts an invisible domain (8 citations) into a functioning AI content asset. Decision-stage content depth closes the funnel gap where Veolia's recommendation strength falters. These are not multi-year brand transformation programs. They are executable content operations, deliverable within a 90-day sprint cycle, measurable through the same GEO Radar infrastructure that produced this audit.
The path runs through three content investments — not brand investments — executable in 90 days. Wikipedia stewardship, owned-domain rebuild, and decision-stage content depth are the levers that convert Veolia's volume advantage into positioning dominance. The data is clear, the gaps are specific, and the roadmap is actionable.
Data collection: March 2026 • Engines: ChatGPT & Gemini
Total prompts analyzed: 19,141 • Markets: 4 • Product lines: 3 • Competitors: 16