The Largest Dataset in This Series – and What It Proves

One hundred companies. Ten industries. One research framework applied consistently across every major sector of the global economy.

This is the cumulative analysis of an independent research series that began in early 2026 and now spans Life & Health Insurance, Global Banking, Industrial Manufacturing, SaaS CRM, Global Pharmaceutical, the Global Legal Industry, the Automobile Industry, Commercial Vehicles, Hospitality & Tourism, and Global Light Vehicle Manufacturers. Each installment examined ten of the world’s largest and most recognized organizations in their respective sector. Each used the AI Visibility Inspector and the Ivica Srncevic Framework to evaluate how confidently AI systems can retrieve, interpret, and cite a brand.

The question driving every installment has been identical: when a buyer, researcher, procurement officer, fleet manager, traveler, or financial decision-maker asks an AI system a question that your company should be able to answer, does your digital presence give AI the structural signals it needs to surface and represent you with confidence?

Across 100 companies and 10 industries, the answer is, with rare exceptions: no.

This article synthesizes the full dataset. It identifies where the structural gaps are deepest, where they are narrowest, which sectors are best and worst positioned, which failure modes recur most consistently, and what the cumulative evidence means for enterprise organizations operating in a search environment where AI-mediated discovery is no longer an emerging trend but the operating condition.

Methodology

The first four installments in this series, Insurance, Banking, Industrial Manufacturing, and SaaS CRM, used the AI Visibility Inspector’s original four-dimension framework: Structural Integrity, AI Extractability, Entity Clarity, and AI Visibility Signals. These produced a composite AI Readability Score on a 0–100 scale.

Beginning with the Pharmaceutical installment, the framework was refined and standardized into four dimensions that better isolated the specific structural signals governing AI retrieval: Structure, Depth, Schema, and Freshness. These produce an AI Retrieval Index score, also on a 0–100 scale.

Both frameworks measure the same underlying reality: how legible, attributable, and citable is a company’s digital presence to AI systems? The scale, the passing thresholds, and the directional findings are directly comparable. Scores below 50 indicate significant structural invisibility. Scores between 50 and 74 represent fair to moderate visibility with material gaps. Scores at 75 and above indicate good to strong AI readiness.

Across the full 100-company dataset, no industry sector average exceeded 75. No sector achieved Grade A as a group. Only individual companies in individual sectors reached or approached the readiness threshold, and they are the exceptions that prove the rule.

The Complete Sector Ranking

Note: SaaS CRM, Industrial Manufacturing, Insurance, and Banking were assessed using the original AI Readability Score framework; Pharmaceutical through Hospitality used the standardized AI Retrieval Index. Both run 0–100 with identical readiness thresholds. Insurance and Industrial averages are derived from the score ranges and performance bands described in their respective reports.

The cross-series average, across all 100 companies and 10 industries, sits at approximately 59. That is Grade C, structurally fair, materially insufficient, and nowhere near the 75-point threshold that constitutes genuine AI readiness.

What the Full Dataset Reveals: Seven Cross-Industry Findings

Finding 1: No Industry Is AI-Ready as a Sector

The single most significant finding in this dataset is the one that applies to every row of the table above: not one sector average has reached 75. Not one industry, assessed as a whole, meets the threshold for good AI readiness.

SaaS CRM leads the series at approximately 73.8, and that figure reflects the presence of HubSpot and Pipedrive at 83, genuinely the strongest performances in the entire 100-company dataset. Without those leaders, the SaaS average would collapse into the mid-to-upper 60s, consistent with the pattern seen across insurance and industrial manufacturing.

At the bottom, Hospitality & Tourism averages 50.7, an industry whose entire commercial model depends on being discovered by travelers, now structurally invisible to the machines those travelers increasingly trust for recommendations.

Between those poles sits the vast majority of the world’s largest enterprises: performing in the 50–65 range, triggering structural warnings, carrying content that AI systems cannot fully parse with confidence, and ceding retrieval advantage to intermediaries, aggregators, and third-party publications that have made different structural choices.

The pattern is not sector-specific. It is universal. Across insurance, banking, manufacturing, software, pharmaceuticals, law, automobiles, vehicles, and hospitality, the same finding repeats: the world’s biggest brands have not built their digital presence for the machines that now mediate an increasing share of the research, discovery, and consideration journeys that precede every major commercial decision.

Finding 2: Schema Is the Universal Failure Mode

If there is a single technical dimension that explains more of the cross-series underperformance than any other, it is Schema, the structured, machine-readable markup that tells AI systems what a company is, what it makes, and what facts can be attributed to it with confidence.

In the later installments of this series, where Schema was measured as a discrete dimension:

• Commercial Vehicles sector average Schema: 27.0 – the lowest in the entire series
• Pharmaceutical sector average Schema: 34.5
• Automobile (brands) sector average Schema: 36.5
• Hospitality & Tourism sector average Schema: 36.5
• Legal sector average Schema: 36.5
• Light Vehicles sector average Schema: 38.5

The highest Schema score in the entire light vehicle, automobile, commercial vehicle, pharma, legal, and hospitality datasets belongs to General Motors at 80. It is the exception so exceptional it distorts its own sector’s average upward. Remove GM, and the light vehicle Schema average drops to approximately 35, indistinguishable from every other sector.

In the earlier installments, SaaS CRM, Banking, Industrial, Insurance, Schema was captured as part of the AI Visibility Signals dimension, and the finding was identical. HubSpot and Pipedrive stood out for structured signals. The rest of the SaaS field clustered at or below 40 on visibility signals. Across banking, industrial, and insurance, structured signal implementation was described consistently as “the most neglected dimension in the dataset.”

The implication is direct. Without Schema markup, AI systems cannot confirm what a company is. They infer. Inference is uncertain. Uncertainty produces hedged, low-confidence representations. Low-confidence representations do not surface in synthesized AI answers. Brands that have built global recognition over decades become structurally ambiguous to the machines that are now the first audience for the buyers those brands depend on.

This is not a small technical oversight. It is the primary structural gap separating most of the world’s largest companies from genuine AI retrievability.

Finding 3: Freshness Failure Is Near-Universal, and Strategically Catastrophic

In the six later installments that measured Freshness as a discrete dimension, the findings were consistently alarming.

• Commercial Vehicles: seven of ten companies scored zero on Freshness
• Hospitality & Tourism: seven of ten companies scored zero on Freshness
• Automobile (Global Brands): five of ten companies scored zero on Freshness
• Light Vehicles: five of ten companies scored zero or near-zero on Freshness
• Pharmaceutical: Freshness scores were highly polarized, MSD achieved 85, the highest in any single dataset; several others scored zero
• Legal: Several firms achieved notable Freshness scores; Gibson Dunn led at 89, the joint-highest in the series alongside MSD’s 85

Freshness signals, specifically dateModified JSON-LD, tell AI retrieval systems that content is current. Without them, AI systems cannot confirm whether they are reading today’s product specifications, current policy terms, active service portfolios, or updated pricing structures. In practice, the absence of Freshness signals causes AI systems to default toward third-party sources, news publications, review sites, trade media, comparison platforms, that do carry date signals. The brand’s own content becomes, in the AI retrieval hierarchy, a less-preferred source for information about itself.

The sectors where this consequence is most acute are precisely those where Freshness failure is most widespread. A hospitality brand that cannot prove its content is current loses AI retrieval competition for destination queries to OTAs that carry date signals on every listing. A commercial vehicle manufacturer that cannot prove its product specifications are current loses retrieval competition to trade publications that timestamp every article. A pharmaceutical company without Freshness signals cedes pipeline and product queries to healthcare news aggregators.

This is also the finding with the most asymmetric remediation profile in the dataset. Implementing dateModified JSON-LD on page templates is a development task measured in hours. The retrieval benefit is material and direct. The cost-to-impact ratio of fixing Freshness is, across the full dataset, among the most favorable of any structural intervention available.

Finding 4: Structural Decay Is Systemic, Not Exceptional

Structural Decay, the condition triggered when a page is missing an H1 tag entirely, carries multiple competing H1s, or lacks verifiable date signals, was the norm rather than the exception across every sector where it was measured.

• Commercial Vehicles: 10 of 10 companies triggered Structural Decay warnings, the first time in the series that universal failure was recorded
• Hospitality & Tourism: 10 of 10 companies triggered warnings
• Light Vehicles: 7 of 10 companies triggered warnings
• Automobile (Global Brands): 9 of 10 companies triggered warnings; Volkswagen was the sole exception
• Legal: multiple firms triggered warnings for various structural deficiencies
• Pharmaceutical: multiple companies triggered warnings

The H1 tag is the primary semantic anchor of any web page. It is the signal that tells AI parsers what a page is definitively about. When it is absent, AI must infer topical identity from secondary signals. When it is duplicated, AI encounters competing claims and cannot resolve primary intent. In both cases, the result is the same: imprecise entity representation, reduced citation confidence, and structural disadvantage in AI retrieval.

The causes of H1 failure across this dataset are instructive. In several cases, notably DAF and Daimler Truck in commercial vehicles; Toyota, BYD, and Nissan in light vehicles; Marriott, Royal Caribbean, and MSC Cruises in hospitality, no H1 tag was found at all. These are not small companies. They are category leaders whose global digital teams have not implemented the most basic structural signal in machine-readable content.

In other cases, Ford (3 H1 tags), Suzuki (7 H1 tags), Tripadvisor (2 H1 tags), Iveco (3 H1 tags), the failure mode is the opposite: so many competing H1s that AI parsers cannot resolve which topic is primary. More structural content has produced less structural clarity.

Finding 5: Depth Is Strong Everywhere – and Insufficient Everywhere

One of the most consistent and counterintuitive findings across the full dataset is the relationship between content quality and AI retrievability.

Depth scores, measuring the substantive quality and richness of content, are consistently strong across all ten industries. In the later installments using the AI Retrieval Index:

• Pharmaceutical: multiple companies scored 85–95 on Depth
• Commercial Vehicles: MAN, Freightliner, Peterbilt, Scania all scored 85 on Depth
• Automobile: Tesla scored 90, the highest Depth in that dataset
• Legal: several firms scored 85 on Depth
• Light Vehicles: GM, Ford, Toyota all scored 85 on Depth

In the earlier installments, extractability scores were similarly elevated, the insurance industry clustered consistently in the 70–76 range on AI Extractability, and banking, industrial, and SaaS companies showed comparable content accessibility.

The world’s largest companies write substantial, detailed, professional content. Their digital presences are not thin or low-quality. The content, by almost every human measure, is good.

But across this entire dataset, 100 companies, 10 industries, the pattern is the same: strong Depth or Extractability scores coexist with D-grade or low-C overall scores, because the structural signals that govern whether AI can confidently use that content are absent. A page with a Depth score of 85 and a Schema score of 10 is a page that has invested heavily in writing for humans while leaving the machine layer, the layer that now increasingly determines whether that content surfaces at all, largely unaddressed.

The lesson is direct and uncomfortable for every enterprise digital team that has measured success through content quality metrics: content quality is necessary but insufficient. The machine layer requires its own investment, its own governance, and its own success criteria. Until that investment is made, even the best-written content on the web’s most authoritative domains will fail to surface in AI-synthesized answers with the consistency and confidence that commercial performance requires.

Finding 6: Sector Context Amplifies the Commercial Stakes Differently

The commercial consequences of AI visibility failure vary by sector in ways the scores alone do not fully capture.

In SaaS CRM, where the entire go-to-market model depends on digital discovery, no distributor network, no dealer relationship, no trade show presence substitutes for being found online, the structural gaps documented in the analysis represent the highest proportional commercial exposure. A SaaS company invisible to AI is a company that has lost its primary acquisition channel.

In Legal, the research journey for outside counsel increasingly runs through AI systems. A managing partner asking ChatGPT for the leading firms in a specific practice area receives a synthesized answer. Firms not represented in that answer do not appear in the consideration set of the most AI-fluent buyers, precisely the buyers who tend to drive the largest mandates.

In Pharmaceutical, clinical decision-makers, procurement researchers, and healthcare administrators are increasingly using AI to navigate complex product and pipeline comparisons. Schema-undeclared products and Freshness-invisible pipelines miss the retrieval moment that precedes the evaluation shortlist.

In Hospitality, the OTA displacement effect is the most operationally visible consequence in the dataset. Brands that have spent billions trying to drive direct bookings, to reduce OTA commission dependency, and to build direct customer relationships are, through structural AI visibility failure, systematically handing AI retrieval advantage to the intermediaries they have been trying to disintermediate. Booking Holdings at 63 outperforms Marriott International at 35 in this dataset not because it is a better hotel, it is not a hotel at all, but because it has made better structural decisions about machine readability.

In Commercial Vehicles and Industrial Manufacturing, the buyer is a procurement professional or technical engineer making a multi-year, multi-million-euro decision. These buyers research inside AI systems before they engage a distributor or sales representative. Brands invisible at that research stage do not make the shortlist. Brands that do not make the shortlist cannot be selected regardless of product quality.

The commercial consequence is not hypothetical. It is the same consequence that has played out in search engine rankings for two decades, except that AI retrieval operates on structural eligibility rather than link authority, and the remediation path is different, the implementation requirements are specific, and the competitive window for establishing retrievability advantage is open now, not in five years.

Finding 7: The Gap Between Leaders and the Field Is a Strategic Signal

Within each sector, the spread between the top performer and the sector median is as important as the average itself. And across the full dataset, that spread carries a consistent message: the companies that lead their sectors have made deliberate structural choices that their competitors have not.

General Motors at 75 in both the Automobile and Light Vehicle datasets stands twelve to twenty points above most of its competitors. The primary differentiator is a Schema score of 80, the highest in both datasets by a significant margin. GM did not outscore its competitors on content quality or structural organization. It outscored them on machine-readable entity declaration. That single dimension, implemented more thoroughly than any peer in the dataset, produces a double-digit retrieval advantage.

HubSpot and Pipedrive at 83 in SaaS CRM lead a field where the next group clusters in the upper 70s. Their advantage comes from structured content architecture, clear entity signals, and FAQ schema implementation, the same structural investments that produce AI retrievability. The companies that followed them in implementation will follow them in AI visibility.

DLA Piper at 72 in Legal leads a field that averages 60.9. The twelve-point advantage over the sector mean comes from a combination of strong Schema (65, the highest in the legal dataset), reasonable Freshness, and clean structural signals. A buyer asking AI to compare leading firms for a cross-border M&A mandate encounters DLA Piper’s representation first.

MSD at 73 in Pharmaceutical, with a Freshness score of 85, the highest in any pharma dataset, leads a field that averages 57.8. The fifteen-point gap is substantially explained by a single structural implementation: verifiable content dates. The company that tells AI systems its content is current is the company that gets cited when AI answers questions about current treatments, pipeline products, and ongoing clinical research.

In every case, the leader is not the company with the largest marketing budget or the most extensive content team. It is the company that has made a specific structural decision, Schema, Freshness, H1 clarity, entity signal implementation, that its competitors have not made. And that decision, because of its technical nature, is replicable by any competitor willing to make it.

The Cross-Industry Leaderboard: Individual Company Standouts

Across the full 100-company dataset, a handful of individual performances stand out as the clearest examples of what AI readiness looks like at enterprise scale, and what it unlocks.

General Motors (75) is the most instructive case in the dataset. Assessed in both the automobile and light vehicle analyses, GM’s Schema score of 80 is the structural differentiator that no competitor in either dataset has matched. It is the only company in either analysis to achieve Grade B, and its performance demonstrates that the 75-point threshold is reachable for a global automotive brand, which means every other automotive brand in this dataset has a structural gap, not a capability gap.

HubSpot and Pipedrive (83 each) are the highest-scoring companies in the entire dataset. They demonstrate that the 75-point threshold is not only reachable but exceedable, and that SaaS companies with disciplined content architecture and entity signal implementation can achieve AI readiness as a consistent organizational capability rather than a one-off technical fix.

MSD (73) is the pharma sector leader and the strongest argument in the entire dataset for Freshness signal investment. A Freshness score of 85 against a sector average approaching zero is not the result of better content, MSD’s content is comparable to its competitors. It is the result of implementing the machine-readable date signals that allow AI systems to confirm currency. That implementation separates MSD from every other pharmaceutical company in the analysis.

DLA Piper (72) and A&O Shearman (71) are the legal sector leaders, and their combined performance above 70 is the closest the legal industry comes to approaching readiness. DLA Piper’s Schema score of 65 is the highest in the legal dataset and reflects a structured data investment that its competitors have not matched.

Booking Holdings (63), while not approaching the readiness threshold, leads the hospitality dataset by ten points over its nearest OTA competitor and by a much wider margin over the hotel chains and cruise lines it intermediates. It is the clearest example in the dataset of structural advantage translating into AI retrieval dominance in a sector where the company is itself an intermediary, and where the brands being intermediated have the content assets, property portfolios, and domain authority to compete, but have not made the structural investments to do so.

The Cross-Industry Patterns: What Always Fails, What Sometimes Works

After 100 companies and 10 sectors, the data is unambiguous on which structural elements drive consistent failure, and which, when present, drive consistent outperformance.

What always fails:

Schema implementation, particularly at levels below 50, is the dominant predictor of underperformance across the entire dataset. No sector with an average Schema below 40 has achieved a sector average above 60. Schema failure is the single most reliable predictor of AI retrieval underperformance in this series.

Freshness signal absence is the second-most consistent failure mode. Sectors where Freshness averages approach zero, commercial vehicles, hospitality, automobiles, are also the sectors with the lowest overall averages. The correlation is structural: AI systems that cannot verify content currency default to third-party sources that can.

H1 absence or fragmentation appears in every sector’s dataset and is the most acute form of Structural Decay. The company with no H1 is a company AI parsers cannot anchor to a primary topic. The company with seven H1s is a company AI parsers cannot resolve. Both outcomes produce the same retrieval failure.

What sometimes works:

High Depth or Extractability scores are necessary but not sufficient. They are present in virtually every strong performer in this dataset, but they are also present in many D-grade performers. Depth without Schema, Freshness, and H1 clarity does not produce AI readiness. It produces well-written content that AI systems cannot confidently surface.

Strong Structure scores are similarly necessary but not sufficient. Multiple companies in this dataset achieved Structure scores of 100, perfect architectural organization, while recording overall scores in the low 50s or high 40s. Structure enables retrieval eligibility. It does not confer it.

The companies that outperform their sectors combine adequate scores across all four dimensions. No single dimension compensates for catastrophic failure in another. GM’s 75 comes from 100 on Structure, 85 on Depth, 80 on Schema, and 4 on Freshness, a balanced profile where even a low Freshness score is offset by strength everywhere else. HubSpot’s 83 comes from similarly distributed investment. The sector laggards, by contrast, typically show strength in one or two dimensions and collapse in the others.

The Grand Aggregate: 100 Companies, One Conclusion

Across 10 industries, 100 companies, and the full span of this research series, the evidence converges on a single finding:

The world’s largest enterprises have not built their digital presence for machine audiences.

They have built exceptional content ecosystems, sophisticated navigation architectures, and high-investment brand communications, all optimized for human users and traditional search engine ranking. The machine layer, the structured signals that enable AI systems to identify, attribute, and confidently cite these brands, has been left largely unaddressed.

This was a defensible position in 2022, when AI-mediated discovery was an emerging experiment. It was a manageable gap in 2023 and 2024, when AI systems were supplementary to search rather than primary. It is not defensible in 2026, when the research journeys that precede the world’s most significant commercial decisions, fleet procurement, pharmaceutical sourcing, legal counsel selection, financial product choice, vehicle purchase, hotel booking, are increasingly beginning inside AI systems, producing synthesized answers that draw from structurally eligible sources.

The companies in this dataset that have achieved structural AI readiness, or approached it, are not necessarily the largest, the most prestigious, or the best-resourced in their sectors. They are the ones that have made specific, implementable, structural decisions about Schema markup, Freshness signals, and H1 clarity. Those decisions are not proprietary advantages. They are replicable choices. And they are available to every company in every dataset in this series, today.

The window for first-mover advantage in AI retrievability is open. The data shows it has not yet been claimed by most industries. The organizations that claim it earliest will not simply improve a technical metric. They will occupy the structural position from which AI-mediated buyer journeys begin, a position that, once established, compounds in visibility, attribution confidence, and competitive advantage with every query an AI system answers on their behalf.

Key Takeaways Across All 10 Industries

The full dataset of 100 companies produces the following cross-series conclusions:

No sector is AI-ready as a whole. The highest sector average in the series is SaaS CRM at approximately 73.8, a figure still below the 75-point readiness threshold, and artificially elevated by two category outliers.

Schema is the universal gap. Across every sector where it was measured discretely, average Schema scores clustered between 27 and 40. The single company with a Schema score above 70 in the six later datasets is General Motors at 80. Every other company is below the threshold at which AI systems can attribute content with full structural confidence.

Freshness failure is strategically catastrophic. Six of the ten installments documented near-total or total Freshness failure across the majority of their datasets. The sectors with the worst Freshness profiles are also the sectors with the lowest overall scores. The remediation cost is minimal. The return is direct.

Structural Decay is the norm, not the exception. Only one company in the automobile dataset, Volkswagen, avoided triggering a Structural Decay warning. In commercial vehicles and hospitality, the warning was universal: ten of ten. In every sector, structural decay at the H1 or date-signal level is the dominant technical failure mode.

Depth cannot compensate for structural failure. The most counterintuitive finding across all 100 companies is that content quality is broadly strong, and broadly insufficient. The machine layer has not received equivalent investment to the content layer. Until it does, the content layer’s value remains partially unrealized.

The leaders are replicable, not exceptional. General Motors, HubSpot, Pipedrive, MSD, DLA Piper, these companies have achieved their relative positions not through capabilities their competitors lack, but through structural choices their competitors have not made. Those choices are available to every organization in every dataset, today.

The commercial window is open, but it is not permanent. AI-mediated discovery is compounding in market share across every sector analyzed in this series. The organizations that address their structural gaps earliest will establish AI retrievability as a durable competitive position. The organizations that defer will address these gaps in a more competitive environment, as their most forward-thinking competitors establish the reference points against which their own AI representations are compared.

This cumulative analysis synthesizes findings from ten independent installments covering: Life & Health Insurance, Global Banking, Industrial Manufacturing, SaaS CRM, Global Pharmaceutical, Global Legal, Automobile Industry, Commercial Vehicles, Hospitality & Tourism, and Global Light Vehicle Manufacturers. All assessments use the AI Visibility Inspector and the Ivica Srncevic Framework.