90 Days of KSM™: I Built the Framework, Then Put It Through Its Own Test

On May 5, 2026, I completed the research that introduced the Knowledge Structuring Model (KSM™).

The paper, The Knowledge Structuring Model (KSM™): A Socio-Technical Framework for AI-Mediated Visibility and Citation Authority, was subsequently posted to SSRN on May 15, 2026.

The research began with a deceptively simple question:

“What makes an organization visible, understandable, and authoritative to Artificial Intelligence?”

For more than two decades, organizations optimized websites primarily for search engines. Keywords mattered. Backlinks mattered. Technical SEO mattered. Rankings mattered.

They still do.

But generative AI has introduced another layer of discovery. People increasingly ask AI systems questions and receive synthesized answers rather than simply receiving a list of webpages.

That changes the visibility problem.

Organizations must now consider whether AI systems can find their knowledge, understand what it means, correctly associate it with them, determine whether it is trustworthy, and potentially cite it when constructing an answer.

KSM™ was developed to provide a framework for understanding that problem.

The Knowledge Structuring Model

KSM™ conceptualizes AI Visibility through three interdependent organizational capabilities.

Structured Extractability

Can AI systems access, extract, interpret, and reuse the organization’s knowledge?

Entity Salience

Can AI systems accurately determine who the organization is, what it does, what expertise it possesses, and how its people, products, research, services, and ideas relate?

Citation Authority

Does sufficient credible evidence exist for AI systems to trust and potentially cite the organization’s knowledge?

The multiplicative relationship is important. Strong performance in one dimension cannot completely compensate for serious weakness in another.

An organization can publish excellent information, but if AI systems cannot reliably associate that information with the correct entity, its visibility may remain constrained.

Likewise, a recognizable entity without credible supporting evidence may struggle to establish citation authority.

Then I Asked a Different Question

Publishing a framework is one thing. Operationalizing it is another.

If KSM™ was going to have practical value for organizations, I believed it needed to move beyond conceptual research.

So I decided to apply the framework to something I could observe directly: my own digital knowledge ecosystem.

That included my website, research, publications, professional identity, articles, book, structured data, citations, entity relationships, and presence across the broader digital ecosystem.

“I built the framework—and then made myself the first 90-day implementation case.”

The objective was not to manufacture a predetermined result.

It was to determine where the framework exposed weaknesses, what could realistically be changed, and whether KSM™ could be translated into a repeatable implementation methodology.

What the Initial Audit Revealed

One of the most important lessons was that having substantial expertise and published knowledge does not automatically produce strong AI Visibility.

Knowledge can exist while remaining poorly structured for machine interpretation.

Expertise can exist while entity relationships remain fragmented.

Research can exist without being sufficiently connected to the author, framework, book, organization, or other related knowledge assets.

Authority can exist without being consistently reinforced across the digital ecosystem.

“Publishing knowledge is not the same as structuring knowledge.”

This became one of the central operational lessons of the 90-day experiment.

Turning KSM™ Into a 90-Day Implementation Roadmap

The experiment helped translate the theoretical framework into three practical implementation phases.

Days 1–30: Audit & Baseline

The first phase establishes what AI systems can currently see and understand.

Audit:

Content and Knowledge

• What does the organization actually publish?
• Is important knowledge buried in long-form prose, PDFs, JavaScript applications, disconnected pages, or poorly structured repositories?

Entity Representation

• Are the organization, executives, products, research, services, expertise, and intellectual property consistently represented?

Citation Authority

• What independent sources support important claims?
• Where are citations, research publications, institutional references, media mentions, backlinks, and third-party validation coming from?

Technical Accessibility

• Can machines efficiently access and interpret important content?
• Are structured data, semantic markup, page architecture, metadata, crawlability, and accessibility supporting or hindering extraction?

AI Presence

• How does the organization currently appear across relevant AI-generated answers?

“The purpose of Phase 1 is to establish an AI Visibility baseline.”

Days 31–60: Structure & Strengthen

Once the gaps are visible, the second phase focuses on strengthening the three KSM™ pillars.

Content can be rewritten into clearer, answer-first knowledge structures.

Important concepts can receive dedicated pages.

Structured data can clarify entities and relationships.

Research can be connected explicitly to its author and underlying framework.

Executive profiles can be aligned.

Internal linking can connect related knowledge assets.

Original research, evidence, and defensible expertise can strengthen authority.

Days 61–90: Validate, Measure & Scale

The third phase returns to measurement.

The organization:

• Retests important questions across AI environments
• Reviews entity representation
• Evaluates citation pathways
• Validates structured content
• Documents changes in visibility and representation

Successful interventions can then scale across:

• Topics
• Products
• Executives
• Research
• Services
• Business units
• Geographies
• Organizational knowledge domains

Governance becomes important at this stage because without governance organizations can gradually recreate the fragmentation the implementation was designed to correct.

What Changed During Those 90 Days?

Perhaps the most significant outcome was not one individual ranking, citation, or technical improvement.

It was the evolution of KSM™ itself.

1. Research

   The original KSM™ framework established the theoretical foundation.

2. Book

   AI Visibility, AEO & GEO: Understanding the Knowledge Structuring Model (KSM™) translated the framework for practitioners, executives, marketers, AI professionals, and digital strategists.

3. Assessment Methodology

   The three pillars were translated into measurable dimensions organizations can audit and benchmark.

4. 90-Day Implementation Roadmap

   The research became an operational sequence organizations can use to diagnose, remediate, validate, and scale AI Visibility.

5. Technology

   KSMModel.ai began translating the methodology into an accessible technology and assessment environment.

6. Professional Education

   KSM™ became the foundation for educational and professional-development material around AI Visibility, AEO, GEO, structured knowledge, and enterprise AI strategy.

7. Industry Applications

   The framework began being explored across different organizational environments rather than remaining limited to general digital marketing.

AI Visibility Was Only the Beginning

KSM™ originally focused primarily on an external problem:

“Can AI find, understand, trust, and cite organizational knowledge?”

But enterprises are increasingly deploying AI internally.

Retrieval-Augmented Generation, enterprise copilots, knowledge assistants, multi-agent architectures, and Agentic AI systems must work with internal organizational knowledge.

These systems require knowledge they can:

• Retrieve.
• Interpret.
• Relate.
• Validate.
• Reason over.
• Act upon.

This led me to extend the original KSM™ research toward Enterprise Knowledge Readiness (EKR).

My subsequent research, Beyond AI Visibility: Extending the Knowledge Structuring Model (KSM™) to Enterprise Knowledge Readiness (EKR) for Agentic Artificial Intelligence, examines whether the organizational capabilities identified through KSM™ have broader significance as enterprises prepare knowledge for autonomous AI.

Enterprise Content Is Becoming Enterprise Knowledge

One of my strongest conclusions from the first 90 days is that organizations need to reconsider what they call “content.”

A research paper is not simply content.

A methodology is not simply content.

Executive expertise is not simply content.

A validated case study is not simply content.

A proprietary dataset is not simply content.

A product specification is not simply content.

These are knowledge assets.

When properly structured, connected, attributed, supported, and governed, they can become part of an organization’s machine-readable knowledge architecture.

What This Means for Marketing Leaders

Marketing organizations should not abandon SEO. They should expand beyond it.

AI-mediated discovery increasingly requires coordination across:

• SEO
• AEO
• GEO
• Structured data
• Content strategy
• Corporate communications
• Reputation management
• Digital PR
• Research
• Knowledge management
• Entity management
• AI strategy

KSM™ provides an architecture for connecting those activities around a common objective:

“Building knowledge that machines can understand and trust.”

What This Means for Enterprise AI Leaders

For CIOs, CTOs, CDOs, Chief AI Officers, knowledge leaders, and enterprise architects, the implications extend further.

AI performance will increasingly depend on the quality of the knowledge environment surrounding the model.

A sophisticated model operating over fragmented, contradictory, inaccessible, poorly governed enterprise knowledge remains constrained by that knowledge.

KSM™, EKR and Warrant Clearing

This progression also connects with my research with Jabran Chaudry on Warrant Clearing for Artificial Intelligence Agents.

KSM™ focuses on preparing knowledge so machines can discover and interpret it.

EKR extends that challenge into the enterprise knowledge environment.

Warrant Clearing addresses another critical dimension: maintaining relationships among claims, evidence, provenance, context, and justification as autonomous reasoning progresses.

“KSM™ prepares enterprise knowledge for machine consumption. Warrant Clearing preserves the integrity of that knowledge throughout autonomous reasoning and execution.”

The future AI enterprise needs knowledge that is structured, identifiable, attributable, trustworthy, governed, and usable.

What Comes After the First 90 Days?

The next stage includes:

• Continued empirical testing
• Industry-specific KSM™ applications
• Enterprise AI Visibility benchmarking
• Development of EKR
• Knowledge governance
• Agentic AI readiness
• Implementation playbooks
• Assessment methodologies
• Continued research into trustworthy autonomous AI

The objective is not to replace SEO, AEO, GEO, knowledge management, or AI governance.

It is to provide a framework connecting them around an emerging organizational challenge:

“How do we structure knowledge for a world in which Artificial Intelligence increasingly mediates what people discover, what organizations know, what machines trust, and eventually what autonomous systems do?”

90 Days Later

Ninety days ago, KSM™ began as an attempt to answer a research question.

Today, it is developing into a framework for approaching a much broader transformation.

“AI Visibility is not simply a marketing problem. It is a knowledge problem.”

“Enterprise AI readiness will increasingly become a knowledge-readiness problem.”

The organizations that understand this early have an opportunity to move beyond optimizing individual pieces of content and begin building something far more durable: a trusted enterprise knowledge architecture for the AI era.

Continue with the research

Underlying research

1. Bowen, A. Q. (2026). The Knowledge Structuring Model (KSM™): A Socio-Technical Framework for AI-Mediated Visibility and Citation Authority. SSRN Electronic Journal. SSRN 6721140 · https://doi.org/10.5281/zenodo.20766821
2. Bowen, A. Q., & Chaudry, J. I. (2026). Warrant Clearing for Artificial Intelligence Agents: Maintaining Justified Claims Across Recursive Agentic Synthesis. SSRN Electronic Journal. SSRN 6899418 · https://doi.org/10.2139/ssrn.6899418
3. Bowen, A. Q. (2026). Beyond AI Visibility: Extending the Knowledge Structuring Model (KSM™) to Enterprise Knowledge Readiness (EKR) for Agentic Artificial Intelligence. SSRN Electronic Journal. SSRN 7113181 · https://doi.org/10.5281/zenodo.21680112

About the author

Dr. Anthony Q. Bowen is an executive AI strategist, researcher, author, educator, and creator of the Knowledge Structuring Model (KSM™) and Enterprise Knowledge Readiness (EKR).

His research examines AI Visibility, structured enterprise knowledge, Agentic Artificial Intelligence, AI governance, digital transformation, knowledge architecture, and the evolving relationship between organizations and intelligent systems.

He is the author of AI Visibility, AEO & GEO: Understanding the Knowledge Structuring Model (KSM™) and co-author with Jabran Chaudry of research addressing Warrant Clearing for Artificial Intelligence Agents.

Executive portfolioORCIDSSRNLinkedIn

Continue reading