Identity Is Always a Trinity

Why the same three-layer structure governs how algorithms, organisations, and people process identity

What you’re about to read

If you run a business, you’ve probably wondered why some brands seem to dominate AI recommendations while others - sometimes bigger, sometimes better - get ignored or hedged. This article explains why. It’s a theoretical framework… but it’s grounded in 25 billion data points and a decade of watching what actually makes algorithms trust (or distrust) a brand. The practical implication is simple: there’s a specific order in which you must build your digital presence, and most businesses get it backwards. Understanding the structure beneath the tactics will save you years of wasted effort and show you exactly where to focus. The theory takes ten minutes. The payoff is super valuable.

I have spent 27 years studying how algorithms understand identity. Since founding Kalicube in 2015, my team has tracked 25 billion data points across 73 million brand profiles, watching three families of system - Knowledge Graphs, Large Language Models, and Search Engines - evolve from simple tools into the world’s most influential identity processors.

One observation has become undeniable: every one of these systems processes identity the same way. Not similarly. The same way.

That forced a bigger question. What if the pattern is not about algorithms at all? What if identity itself - human, brand, algorithmic - follows a universal structural law?

I believe it does.

Three Layers. Always.

Every identity system I have studied operates across exactly three dimensions:

Fact - what is solidly, verifiably true. The canonical layer. Slowest to change, hardest to fake, most expensive to build. Everything else depends on it.

Narrative - the constructed story of authority and position. Where the question ‘who is best?’ gets answered. It is comparative, judgment-driven, and shaped entirely by the quality of evidence behind it.

Surface - perception. Vast, volatile, immediately satisfying but often unreliable. It is what most people obsess over, and what they have the least direct control over.

These are not equal. They are stratified. Fact is the foundation. Narrative is built on top. Surface is the consequence. That order is not a preference. It is an architectural requirement.

The Algorithmic Evidence

The three algorithmic systems that now mediate brand identity - what I call the Algorithmic Trinity, a framework I developed in 2024 - are three expressions of this same structure.

Knowledge Graphs deal in solid fact. Structured, canonical, verified. They are the institutional memory of the algorithm - slow to update, difficult to manipulate, and the foundation upon which everything else is built.

Large Language Models deal in conversation and judgment. They weigh evidence, construct narratives, compare entities, and decide who to recommend. They are the narrative layer - rich, contextual, ruthlessly responsive to the quality of evidence behind any claim.

Search Engines deal in fast results for immediate satisfaction. They are the surface layer - vast, volatile, responsive to signals both meaningful and meaningless.

Here is the insight that took me years to see clearly: these are not one-to-one mappings. They are a matrix. Every dimension exists in every system. Knowledge Graphs have narrative elements. LLMs deal in facts. Search Engines display authoritative content.

But each dimension dominates in one system:

	Knowledge Graphs	LLMs	Search Engines
Fact (U)	DOMINATES	exists	exists
Narrative (C)	exists	DOMINATES	exists
Surface (D)	exists	exists	DOMINATES

That dominant diagonal is where each dimension has its greatest leverage. But every cell in the matrix is active. Optimisation that targets only the diagonal misses the full picture. Optimisation that ignores the diagonal wastes effort.

The Same Structure Governs Human Identity

Strip away the algorithms entirely. Look at how human identity actually works. The same architecture appears.

Raw truth - who you are at your most brutally honest. Unedited, unperformed, unmanaged. The factual bedrock of self.

Constructed narrative - who you are trying to be. The intentional professional identity you build with proof, positioning, and evidence of competence.

External perception - what others actually believe about you. Vast, volatile, shaped by signals both accurate and distorted.

Same three layers. Same stratification. And the same dominant diagonal when you examine how identity is processed in different contexts. Close relationships operate primarily on raw truth - the factual layer. Professional networks operate primarily on constructed narrative - the authority layer. Public reputation operates primarily on surface signals - the perception layer.

Consider someone who builds a strong public presence (Surface) without substance behind it. The moment they enter a professional context that evaluates evidence (Narrative), the facade cracks. And the moment they enter a close relationship that demands authenticity (Fact), it collapses entirely. The build order is structurally required: you have to know who you actually are before you can construct an authentic narrative, and you need the narrative and the proof before public perception can accurately represent you.

This is not a metaphor. The same structural law governs both systems. That is what makes it a theory, not an analogy.

Convergence Is the Mechanism of Trust

Trust - algorithmic or human - is what happens when all three layers say the same thing.

When a brand’s Knowledge Graph facts, LLM narrative, and search results all converge on the same conclusions, AI stops hedging. It does not say ‘claims to be the leading authority.’ It says ‘is the leading authority.’ The verification detector is satisfied.

When a person’s raw truth, professional narrative, and public reputation all converge, people experience them as authentic. There is no gap to sense, no performance to see through.

Divergence is how distortion works. When your facts and your narrative do not match, AI hedges. When your narrative and your perception do not match, people sense something is off. When your facts and your perception have decoupled entirely, you have a reputation crisis - or an AI hallucination problem.

Identity convergence is not perfection. It is coherence. The goal is not that every system says the same words, but that every system arrives at the same conclusions from different evidence.

And critically: convergence at a high level of all three dimensions beats excellence in any single one. A lesser-known entity with tight alignment between Fact, Narrative, and Surface will consistently generate more algorithmic trust than a famous entity whose layers are fragmented. I have seen this pattern thousands of times across the 73 million profiles we track. A client of ours, James Dooley, won a £540,000 contract over a larger competitor specifically because the decision-maker saw a “night and day difference” after spending 30 minutes comparing their online identities. Coherence converts.

What Follows

If identity is triadic, stratified, and subject to convergence as the mechanism of trust, then several things follow:

The build order is non-negotiable. Fact → Narrative → Surface. You cannot shortcut this. Not for brands, not for people, not for algorithms. Every methodology that starts at the surface - traditional SEO, social media hacks, ‘personal branding’ tricks - is building on sand.

AI systems are identity architectures, not search engines. Once you understand the triadic structure, you stop ‘optimising for AI’ and start building coherent identity. The optimisation is a consequence, not a tactic.

Convergence beats fame. Size and visibility do not determine algorithmic trust. Alignment does. This is measurable, actionable, and profoundly good news for any entity willing to do the work.

The same methodology works for everything. You are not optimising for three different systems. You are optimising three dimensions of identity, each of which resonates across all three systems but has maximum leverage in one.

From Theory to Practice

This theoretical framework - what I have been calling the Identity Convergence Theory - is the intellectual foundation beneath everything we build at Kalicube. It is the reason The Kalicube Process™, our proprietary methodology for implementing a holistic, brand-first digital marketing strategy, works the way it does: building Understandability (Fact) first, then Credibility (Narrative), then Deliverability (Surface), across all platforms simultaneously rather than optimising for systems one at a time.

I have written a separate, detailed article explaining exactly how The Kalicube Process implements this theory in practice - including the UCD Framework, the build sequence, and how we measure convergence across the Algorithmic Trinity. You can read it here.

Start at Depth

I have spent nearly three decades asking how algorithms understand identity. The answer turned out to be both simpler and more profound than I expected.

Identity is always triadic. It is always stratified. Trust is always the product of convergence between Fact, Narrative, and Surface. And the build order is always the same.

Start at depth. Build through narrative. The surface corrects itself.