The emerging market for generative engine optimization is beginning to look less like a standalone SEO specialty and more like a convergence of search, content, reputation and measurement.
Marketing agency Gutenberg is making that convergence the center of a new commercial offering.
On September 1, the company launched Gutenberg AI Visibility, a service that combines SEO, answer engine optimization, generative engine optimization, content and digital PR into a single program designed to improve how brands are discovered, represented, cited and recommended across traditional search and AI-generated answers.
The model is built around a simple premise: ranking a website is no longer the only visibility problem. A company can perform well in conventional search while being absent from an AI-generated shortlist, described inaccurately by a chatbot or consistently cited less often than its competitors.
Gutenberg’s response is to treat those outcomes as one cross-functional marketing problem rather than assigning GEO to a new silo.
It is a strategically interesting proposition. It is also a commercial announcement, not independent evidence that the approach works.
Gutenberg is positioning AI visibility above SEO, AEO and GEO
One of the more useful aspects of Gutenberg’s framework is that it does not present GEO as a replacement for SEO.
The agency assigns different jobs to the disciplines. SEO creates the technical and search foundation that allows a brand and its content to be discovered. AEO structures information so answer-oriented systems can identify useful responses. GEO focuses on whether generative systems understand, reference and recommend the brand.
Gutenberg then places those practices inside a broader AI Visibility layer that also includes content and public relations.
Its AI Visibility service page describes the model as three connected disciplines solving different parts of the same problem: discoverability, answerability and citation authority.
That distinction matters because the GEO market is currently full of overlapping terminology. Some vendors use AEO and GEO almost interchangeably; others present GEO as a new technical optimization layer. Gutenberg is explicitly arguing that no single optimization discipline is sufficient.
Digital PR is being treated as AI infrastructure
The most distinctive part of the service is the role assigned to public relations.
Traditional SEO often views digital PR through the lens of backlinks and domain authority. Gutenberg’s model gives earned media a second function: third-party corroboration that can help establish what a brand is known for across the wider web.
That is relevant to generative systems because an AI answer can synthesize information from sources beyond the company’s own website. A brand can make claims about itself on a product page, but independent coverage, expert mentions, interviews and industry references provide a different kind of signal.
Gutenberg therefore divides the work into three operational layers. Digital makes information technically discoverable and machine-readable. Content creates material worth retrieving and citing. PR builds independent authority around the brand.
The agency summarizes the relationship in a separate explanation of its integrated AEO/GEO model, arguing that these functions need to work against the same visibility objective rather than operating as separate departments.
That is a strategic thesis, not proof of a ranking mechanism. AI platforms do not publish a universal formula assigning weights to technical SEO, content and PR mentions.
Share of Model becomes a core metric
Gutenberg’s service also reflects a broader shift in how agencies are attempting to measure AI visibility.
Traditional SEO dashboards revolve around rankings, impressions, clicks and organic traffic. Those metrics become incomplete when a brand can appear inside an AI-generated response without producing a website visit.
Gutenberg uses a metric it calls Share of Model to measure how often a brand appears or is cited for priority prompts relative to competitors across AI engines.
The idea is analogous to share of voice. Instead of asking what percentage of a search-results page a company occupies, the metric asks how consistently the brand is represented across a defined set of AI-generated answers.
This can help expose competitive gaps that organic traffic alone would miss. A company might maintain stable Google traffic while a rival becomes the brand repeatedly recommended by conversational systems for commercially important questions.
Share of Model is still dependent on methodology. The result changes with the prompts selected, AI engines tested, geography, personalization, model versions and measurement frequency. It should therefore be treated as a defined monitoring metric rather than a universal market-share number.
Citation accuracy measures whether visibility is actually correct
Frequency is only one dimension of AI visibility.
Gutenberg also says it measures citation accuracy and whether AI systems represent the company, its expertise and its offering correctly.
This is an important distinction.
A brand can be visible in an AI answer and still have a problem if the system describes an obsolete product, invents a feature, associates the company with the wrong category or cites a source containing outdated information.
That turns GEO partly into an information-governance problem.
Teams need to know not only whether an AI system mentions them but what it says when it does. Correcting the underlying web information may require changes across product pages, documentation, structured data, third-party profiles and earned media rather than a single SEO edit.
AI referral traffic remains part of the measurement stack
Gutenberg has not abandoned traffic as a metric.
The service measures AI referral performance alongside Share of Model and citation accuracy, and the company says it can connect that visibility with downstream business outcomes and traditional search metrics.
That combination is important because citation volume alone does not demonstrate commercial value.
A brand could increase its appearances in ChatGPT or Gemini while receiving little qualified traffic. Another company could generate relatively few AI referrals but see those visitors convert at a high rate.
Any mature AI visibility program will therefore need to connect upstream visibility with downstream behavior where measurement is possible.
The difficulty is attribution. AI interfaces can influence a buying decision without producing a click, while referrals that do occur represent only part of the user journey. Gutenberg’s announcement describes the measurement ambition but does not publish independent evidence showing how reliably its framework attributes business outcomes to AI visibility.
Competitor benchmarking acknowledges that AI visibility is relative
The service establishes a baseline for how a brand appears across relevant search and AI environments and compares that position with competitors.
This is a useful shift from binary monitoring.
Simply asking whether ChatGPT mentions a company can produce a misleading sense of success. The more commercially relevant question may be whether the company appears more or less frequently than competitors when buyers ask category, comparison or recommendation questions.
Competitive monitoring can also reveal a source problem. If one competitor is repeatedly recommended, marketers can investigate which third-party publications, product pages, datasets or authoritative references appear around those recommendations.
That turns AI monitoring from passive reporting into a source-discovery process.
Human review is a deliberate part of the product
Gutenberg describes itself as operating a human-in-the-loop model.
AI can support analysis and monitoring, but the company says strategic, editorial, digital and PR professionals retain oversight around positioning, accuracy, authority and brand safety.
That is particularly relevant in a service whose goal includes changing what machines say about a company.
Automating the production of hundreds of thin pages in an attempt to manufacture AI visibility could create exactly the opposite outcome: weaker content, inconsistent facts and reputational risk.
Human review does not guarantee quality, but Gutenberg is making editorial judgment part of the service architecture rather than treating GEO as a fully automated content-generation workflow.
The agency explicitly refuses to guarantee AI citations
One line in the launch announcement deserves more attention than the marketing language around it.
Gutenberg says it does not guarantee specific rankings or citations from third-party AI platforms.
That is an important boundary for a young optimization industry.
Marketers do not control the retrieval systems, ranking systems or generated responses of ChatGPT, Gemini, Perplexity or Google AI Overviews. Those products change frequently, and the same prompt can produce different sources over time.
An agency can improve crawlability, content quality, factual consistency and external authority. It can measure whether visibility changes afterward. It cannot credibly promise that an independent AI platform will cite a particular page on demand.
Gutenberg’s own service FAQ makes the same point, saying no credible partner can guarantee an AI citation.
The integrated model is plausible, but Gutenberg has not published proof that it works
This is where the launch needs to be separated from evidence.
Gutenberg’s announcement explains what the service does and why the agency believes Digital, Content and PR should work together. It does not publish controlled experiments demonstrating that clients using the service receive more citations than comparable brands that do not.
There are no independent tests of Gutenberg AI Visibility in the launch material, no before-and-after client dataset and no verified case study showing a specific increase in Share of Model, AI referrals or revenue attributable to the service.
That is not unusual for the launch of an agency offering. It simply limits what can be concluded.
The service should currently be evaluated as a methodology and commercial proposition, not as a scientifically validated GEO system.
Gutenberg’s own statistics should not be confused with independent benchmarks
The agency publishes extensive educational material around AI search, including definitions, statistics and recommendations about content structure, authority and citations.
Some of those claims draw on external research; others are Gutenberg’s own interpretations of how brands should approach AI visibility.
For buyers evaluating the service, the distinction matters.
A vendor’s glossary or methodology page can be useful for understanding its framework, but it is not independent validation of the vendor’s effectiveness.
The strongest evidence would be transparent methodology, reproducible measurement and client outcomes that can be independently scrutinized.
Those are not part of the September 1 launch announcement.
The service reflects a larger change in the agency market
Even without outcome data, Gutenberg’s launch is notable because of how it packages the problem.
SEO agencies are adding GEO. PR firms are talking about AI citations. Content agencies are restructuring pages for answer extraction. Analytics vendors are building prompt-monitoring dashboards.
Gutenberg is attempting to put all of those activities under one operating model.
That could appeal to companies discovering that AI visibility issues cross departmental boundaries. The technical team may control schema and crawling, the content team controls product information, communications manages third-party narratives, and analytics measures referral behavior.
If each team optimizes independently, the company can end up with conflicting descriptions of the same product across the web.
An integrated program at least creates the possibility of one source-of-truth strategy across those functions.
GEO is becoming less about tricks and more about information architecture
The broader lesson from Gutenberg’s positioning is that AI visibility is increasingly being framed as an organizational problem rather than a collection of prompt-engineering tricks.
A company needs crawlable information, clear answers, authoritative content, accurate entities, consistent product facts and credible external references. It also needs a way to monitor whether AI systems reflect that information correctly.
None of those requirements is entirely new.
SEO has long cared about crawlability and relevance. Content marketing has long cared about expertise. PR has long cared about third-party credibility. Analytics has long cared about attribution.
Generative AI changes the surface where those disciplines meet.
The real test will be whether integrated visibility produces measurable business outcomes
Gutenberg AI Visibility is available immediately across the agency’s markets for B2B, B2C and B2G organizations in technology, financial services, healthcare, retail, professional services and other research-heavy sectors.
The launch gives marketers a defined framework: measure current AI visibility, compare it with competitors, improve technical and content foundations, build external authority, monitor citation accuracy and track AI-driven referrals and business outcomes.
That is a coherent model.
What remains unproven is the part that matters most commercially: whether brands using Gutenberg’s integrated approach reliably gain more accurate citations, stronger Share of Model and better business results than they would through existing SEO, content and PR programs.
Until case studies or independent testing emerge, the service should be judged on the quality of its execution and measurement rather than on promises about what AI engines will do.
Gutenberg itself sets the correct expectation by refusing to guarantee rankings or citations.
That may be the most credible part of the launch. In an industry rushing to sell GEO as a new optimization discipline, Gutenberg is betting that AI visibility is better treated as the combined output of search, content and reputation — measured together, but still subject to platforms no marketer controls.