Database Brand Yourself

This morning I set out to make a map of my ideas. By lunchtime, I had learned something rather important about the relationship between human interpretation and machine intelligence.

I had spent years developing concepts about technology, identity, agency and the future. Some had appeared in books, others in academic papers, presentations, interviews, podcasts and essays. They were increasingly connected in my own thinking, but they were distributed across different places, different formats and different moments in time.

I wanted to bring them together in one place. A visual representation that would allow people to see how they related to one another.

I called it A System of Ideas.

There were twelve concepts, each represented by a node, with connections indicating their relationships. Some concerned how technologies represent us. Others concerned the production of futures, our capacity for agency, or the increasingly complicated relationship between human selves and intelligent machines.

The map was useful because it showed that these were not twelve separate ideas, but parts of a larger inquiry.

But as I worked on it, something else became clear.

The visual map made my world understandable to people. It did not necessarily make that world intelligible to machines.

And I realised that I had been describing this very problem to brands for some time.

From the Database Brand to the Database Condition

In June 2026, I published The Database Brand: How AI Is Changing What a Brand Is.

The argument was that brand identity is moving from something primarily designed and communicated by an organisation to something increasingly assembled by machines from distributed evidence.

A brand might have a carefully crafted identity, a compelling purpose and a consistent narrative. But AI systems encounter it through a different arrangement of information: product details, press coverage, customer reviews, interviews, social signals, behaviours, classifications and countless other attributes distributed across the network.

The Database Brand is the version of the organisation that machines assemble from those signals.

It may or may not resemble the brand the organisation believes it has created.

I had also contributed to Dentsu Creative's Word of Mouth, Word of Machine. At a breakfast meeting before its publication, I argued that consistency of messaging was coming back into fashion. Repeated, recognisable claims would become increasingly important for machine discoverability. But consistency alone wouldn't be enough. Brands would also need to help machines interpret the meaning of those claims by making explicit the relationships between their disparate attributes. It wasn't simply about repeating the same message, but connecting the evidence that gave it meaning.

But my argument was moving beyond discoverability alone.

In developing what I now call The Database Condition, I was trying to describe a wider transformation in the organisation of identity, culture and value.

A database is not simply a databank. It is not just a collection of information. What makes it a database is the structure of relationships between its elements.

And that distinction matters.

The Database Condition describes a world in which attributes become individually addressable, identifiable and potentially attributable to particular entities or sources.

Database Culture describes how cultural material becomes reassemblable, producing different versions and interpretations of supposedly shared experiences.

The Database Self describes how a person becomes recognisable to computational systems through accumulated data and traces, although the resulting representation may be very different from the person as they understand themselves.

And Database Currency, still an emerging concept in my work, raises the possibility that some of those addressable attributes may become increasingly divisible, valuable and exchangeable.

These are not identical processes. But they belong to the same underlying condition.

The world is increasingly represented not through whole things, but through attributes and the relationships between them.

The same person, organisation or event can become different things depending on which attributes are selected, how they are combined or re-combined, and the systems interpreting them.

In my original Database Brand paper, I wrote:

"The question is no longer whether the parts are consistent. It is whether the parts connect."

I had made the argument. Now I was about to discover what it meant to act on it.

I had to database brand myself

As I looked at my map, I realised that I was facing exactly the problem I had been describing to organisations.

My ideas were published. My concepts were defined. I had books, essays, a website, a dictionary, research papers and interviews.

A search engine or language model could potentially find individual pieces of that work.

But could it understand how they connected?

Would it recognise that Database Self and Machine-Readable Self were related but different concepts? That Me:chine developed the inquiry further, beyond computational representation and into the relationship through which machines may participate in the making of the self?

Would it understand that Big Futurity describes the system producing accounts of the future, while Too Much Future describes what happens to individuals when those externally produced futures begin to overwhelm their own sense of possibility?

Would it know that these concepts belonged to the same authored intellectual framework?

I couldn't assume it would.

The very thing I had been advising brands to consider was now staring back at me from my own website.

I needed to database brand my own brand.

Not by inventing a new narrative or repeating my positioning more frequently, but by making the relationships within my intellectual world explicit.

I needed to explain not only what each concept meant, but what it meant in relation to the others.

And I needed to connect those explanations to the original work from which they had emerged.

This was a different exercise from conventional brand storytelling.

A narrative organises meaning for an audience. A database organises attributes and their relationships for retrieval, interpretation and use.

The two are not mutually exclusive. But they are not the same thing.

A machine may encounter a hundred consistent claims about an organisation without understanding how those claims relate to its activities, history, products, people or evidence.

Consistency is not the same as coherence.

And coherence requires more than repetition. It requires relationships.

Three ways of making meaning legible

Working with ChatGPT, I began constructing three complementary representations of my intellectual world.

The first was visual.

I created a map of twelve concepts, with their connections and proximity illustrating areas of relationship, overlap and distinction.

It was designed for people. Someone could look at it and recognise the shape of the thinking without having to read twelve separate essays.

This was the human-scannable version of my intellectual world.

The second was textual.

I created a canonical page for A System of Ideas, defining each concept individually and then describing its relationships with the others.

I explained why Database Self and Database Brand share a common underlying logic, how Machine-Readable Self connects to Me, and why Big Futurity and Too Much Future describe different aspects of the relationship between systemic change and personal agency.

I linked the concepts to published articles and other original material.

This was the interpretive layer. It turned a visual collection of connected ideas into an explicit written account of their meaning, development and provenance.

The third was structured.

I added Schema.org structured data to the webpage, identifying A System of Ideas as a creative work, naming its author and publisher, describing its twelve constituent concepts and summarising their relationships.

The structured data did not encode every connection as a formal knowledge graph. The more detailed relational explanation remained in the webpage text. But it provided another layer through which computational systems could identify what the work was, who had created it and what it contained.

I tested the markup with the Schema.org validator. It returned no errors or warnings.

There was no guarantee that every AI system would find the page, interpret it correctly or attribute the ideas faithfully. Structured data cannot provide that guarantee.

But something important had changed.

I had made the meaning of the work more explicit in three complementary forms: visual, textual and structured.

One world of ideas, represented in three different ways.

And in doing so, I had done something that goes beyond optimisation for search or AI discovery.

I had undertaken the work of interpretation myself.

Relationships, not just claims

This was the most important discovery.

A conventional approach to brand communication places enormous emphasis on what an organisation says about itself. Get the narrative right. Repeat it consistently. Make sure everybody is saying the same thing.

These remain valuable disciplines.

But in the Database Condition, an organisation's identity is increasingly assembled from many sources, some of which it controls and many of which it does not.

The machine encounters attributes. It connects them, weighs them, associates them with other entities and uses those relationships to construct an account of what the organisation is.

That means the critical unit of understanding is not necessarily the individual claim.

It is the relationship.

A company's sustainability statement, for example, is one attribute. Its supply chain data is another. Its actions, policies, partnerships, reporting and independent assessments are others.

Repeating a sustainability claim consistently does not establish how those different attributes relate to one another.

A system may make those connections for itself. It may also make them badly, overlook important evidence or produce a misleading account.

The organisation cannot dictate the interpretation, but it can do more to clarify the meaning, provenance and relationships of the evidence available.

That is the difference between simply making claims and making an identity intelligible.

In the age of AI, it is no longer enough to make yourself known. You have to make yourself understood.

And being understood requires more than recognisable attributes. It requires intelligible relationships between them.

In retrospect, I had already formulated this principle in The Database Brand. But mapping my own intellectual world made it tangible.

I could see precisely what was missing when twelve separate concepts were left to stand alone.

It wasn't more messaging.

It was the meaning between them.

Working inside and outside the system

There was another dimension to this experience, and it relates directly to my work on Me:chine.

Me:chine is my concept for the meeting of the machinable and unmachinable dimensions of the self. It examines what happens as intelligent machines increasingly participate in who we become, and how we negotiate the relationship between what can be represented computationally and what remains a matter of human experience, judgement and agency.

Throughout the morning, I had been working inside that very relationship.

I was using AI to help structure my thinking, articulate connections and express my intellectual world in forms machines could interpret. At the same time, I was observing the process from outside it, evaluating what the machine produced, correcting its interpretations and deciding what accurately represented my work.

I was using a machine to help make my own intellectual world machine-readable, while exercising human judgement over what that readability would mean.

I was both participating in and observing the negotiation.

And this was the rather profound realisation. I had been writing about Me:chine, exploring the ongoing relationship between the machinable and unmachinable aspects of ourselves, while simultaneously experiencing that relationship in the act of writing and mapping my own ideas.

I hadn't changed my understanding of Me:chine. I had found myself practising it.

There is a wider implication here for brands, researchers, organisations and individuals. Making something understandable to machines does not guarantee how those machines will interpret it. But we can take greater responsibility for articulating the relationships, distinctions and evidence that give our work meaning.

That does not mean surrendering interpretation to a system. Nor does it mean controlling every interpretation the system might produce.

It means participating more deliberately in the ongoing negotiation between human meaning and machine representation.

The map and the territory

I began the day with twelve concepts and a diagram. I ended it with three different representations of my intellectual world and a practical demonstration of an argument I had been making for some time.

The Database Condition is not simply about the accumulation of attributes. It is about their addressability, their relationships and the different identities or meanings that can be assembled from them.

The same is true of Me:chine. The important thing is not simply what the machine can do, or what remains human. It is the ongoing negotiation between them.

That negotiation played out throughout the process: between claims and evidence, attributes and connections, computational legibility and human judgement.

Every connection required interpretation. Every representation invited another judgement about meaning.

The result was not simply a more discoverable website. It was a demonstration of what happens when the machinable and unmachinable work together, without becoming the same thing.

I hadn't outsourced the meaning of my ideas to a machine. I'd worked with a machine to make that meaning legible.

The machine. Me. Me:chine.

Tracey Follows, 10 October 2026

This essay documents the creation of A System of Ideas, a conceptual map connecting twelve ideas developed through my work on technological change, futurity, identity and human agency. It builds on The Database Brand, The Database Condition, The Next Frontier Is the Self, and the inquiry developed in The Future of You: Five Years On.