“Communities Create Markets

— The Guild Economy Made Possible by AI —”

1. Discovery

— Markets Are Built Not by “Customers,” but by “Relationships” —

CO Create was not originally founded as an IT company.

Its founder, Masahiro Miura, came from Itochu Nenryo Co., Ltd.
In its early days, the company operated more like a small trading business, leveraging networks in the energy and automotive industries.

But soon after its founding, the world was hit by the COVID-19 pandemic.

The business environment changed dramatically, and the company’s first steps were marked by constant hardship.

In the midst of that turmoil, the people who reached out to support us came from all directions:

investors
corporations
industry associations

and many others.

As we continued the business with support from people across different industries and of vastly different scales, we came to realize something important:

The value of a company is not created by its size, but by the relationships that surround it.

It Is Not Companies but Relationships That Create Economic Activity

During the pandemic, we were approached by a major industry association.

“We want help improving employee retention at the small and midsize companies within our association.”

That request led CO Create to launch a new business:

employee benefits outsourcing.

Drawing on the buyer network we had built during our trading-company days, we assembled a benefits offering that included:

products
services
member privileges

The response exceeded expectations, and we began signing contracts with far more companies than we had originally anticipated.

But as we continued this work, we gradually noticed a deeper structure beneath it.

Who Really Supports an Industry?

As we engaged with the challenges facing industry associations, a very simple structure became visible:

Industry association
↓
Small and midsize businesses
↓
Employees and customers
↓
A network of human relationships

In other words:

industry associations are sustained by small and midsize businesses
and those businesses, in turn, are sustained by their relationships with employees and customers

To solve problems at the industry level, we realized, we had to reach the network at the very edge.

That was when we understood a fundamental truth:

Markets themselves are formed by human relationships.

Discovering the Structure of the Market

From this observation, we arrived at a new hypothesis.

Markets are not created by companies alone.

They are formed through the following structure:

People
↓
Communities
↓
Economic activity

People do not act primarily because of corporations.
They act because of:

people they trust
communities they belong to
values they share

We began calling this structure the

Guild Correlation Economy,
inspired by the merchant guilds of the medieval era.

Not an IT Company, but a Company That Builds Market Structures

To implement this structure, we created

Hapico, a decentralized e-commerce platform.

Hapico is not just another e-commerce website.

It is an infrastructure for a community-based economy—one that forms:

Community
↓
Purchasing network
↓
Economic sphere

This discovery became the starting point for everything CO Create does today.

2. Theory

— An Unexpected Alignment with the Information Science of the AI Era —

Hapico was not originally designed with AI in mind.

It was developed as a decentralized e-commerce system capable of managing layered community structures—parent organizations, sub-organizations, and sub-sub-organizations.

At the time, we were simply focused on three things:

not relying on advertising spend
not depending on dominant platforms
delivering genuinely good products to communities at the very edge

But at one point, a major realization emerged.

Through a Tokyo Metropolitan Government talent exchange program, we welcomed secondees from major corporations such as:

Mitsubishi Heavy Industries
Toshiba Tec

Later, an engineer from Nomura Research Institute also joined us.

When they saw the system, they said:

“This is perfectly aligned with the information science of the AI era.”

Where Decentralized Commerce and AI Engineering Converge

In particular, Hapico’s underlying philosophy aligned closely with three major fields:

network science
behavioral data analysis
vector representation in AI

Network Science

Modern society is increasingly understood as a network:

Person → Person
Person → Organization
Organization → Organization

Google’s search engine, for example, analyzes the network structure of:

Pages
↓
Links
↓
Importance

In other words:

relationships create value.

The same structure exists in economics.

People
↓
Communities
↓
Purchasing

Which means:

community structure = market structure

Behavioral Data Analysis

Traditional marketing relied on attribute-based data such as:

Age
Gender
Occupation

Today, however,

behavioral data

has become far more important.

Amazon, for example, does not primarily recommend products based on demographic attributes.
It recommends them based on purchasing behavior.

Not attributes
but purchasing behavior

In a community-based economy, the following becomes possible:

Community
↓
Behavior logs
↓
Inference of values and preferences

This makes it possible to conduct

values-based market analysis.

AI Vector Analysis

Generative AI handles things like:

Words
Images
Behavior

inside a

vector space.

Similar concepts are placed close to one another.

The same principle can be applied to communities.

Community
↓
Behavioral data
↓
Vector space

As a result,

Communities with similar values

can be clustered automatically.

In other words:

AI can understand the market structure itself.

3. Proof

— More Than a Decade of Building Community Networks in Practice —

In hindsight, our company had been working with—and implementing—this structure for more than ten years.

Today, we have built a network that includes:

numerous communities
a large membership base
ongoing economic activity

Yet for a long time, this structure remained in a state where

monetization lagged behind its underlying value.

The reason is simple.

This model does not truly work until

the network reaches a certain scale.

4. Inflection Point (2026)

In 2026,
we entered a new phase:

the monetization phase of our network assets.

We are doing this through:

collective purchasing
community e-commerce
membership services
behavioral data analysis

In other words, we are building:

Community
↓
Purchasing network
↓
Economic sphere

5. Validation Projects

Today, this model is being validated across three markets.

1) Industry Association Model

(The Purest Form of a Community Economy)

Examples

National Federation of Care Service Providers
Japan Izakaya Association

Industry associations have the following structure:

Shared issues
↓
Collective purchasing
↓
Shared economy

In the care industry, for example, there are common needs such as:

diapers
consumables
staffing
employee benefits

As a result, the following structure emerges:

Industry
↓
Community
↓
Large-scale purchasing network

This is

the purest form of the guild economy.

2) Corporate Community Model

(Turning Customer Networks into Economic Networks)

Example

Enearc Kansai
Enepan Club

Companies inherently possess:

Customers
↓
Relationships
↓
Networks

But many companies operate only through one-off interactions:

Advertising
↓
Product sales

Our model introduces a different structure:

Customers
↓
Community
↓
Economic activity

This makes it possible to transform customer networks into economic networks through:

member benefits
collective purchasing
new services

In short, it enables

the monetization of customer networks.

3) B2C Model

(Formation of a Self-Sustaining Community)

New Business

Driver Life Plus

Until now, our businesses had relied on existing communities such as:

Associations
Corporations

But in those cases,

control over community formation remained external.

So we launched a new model that creates the following structure autonomously:

Individuals
↓
Occupational community
↓
Economic sphere

The logistics industry is particularly well-suited to this because it has:

a large population
many shared life and work challenges
a strong sense of occupational community

This makes the following structure especially viable:

Drivers
↓
Community
↓
Collective purchasing

6. Why Driver Life Plus?

This project is not simply a new e-commerce business.

It is

a self-sustaining model of the community economy.

Until now, community formation depended on:

Associations
Corporations

Driver Life Plus seeks to build it autonomously as:

Individuals
↓
Community
↓
Economic sphere

In other words,

it is the prototype of the guild economy.

7. What These Three Models Demonstrate

The current validation efforts can be summarized as follows:

Model	Role
Industry associations	Large-scale communities
Corporate customers	Customer networks
B2C	Self-sustaining communities

All three are validating the same underlying structure:

People
↓
Communities
↓
Economic networks

8. Scale Model

— The Blue-Collar Co-op Concept —

The care-sector purchasing platform and Driver Life Plus are not simply e-commerce businesses.

They are

collective purchasing models built on occupational communities.

Their structure is as follows:

Occupation
↓
Community
↓
Collective purchasing
↓
Economic sphere

This resembles the traditional consumer cooperative model, which is based on geographic communities.

But our target is not geography.

It is

occupational communities.

Addressable Market

Japan’s blue-collar workforce includes approximately:

Occupation	Population
Logistics	3 million
Construction	5 million
Manufacturing	10 million
Care work	2 million

Total:

approximately 20 million people

This represents a massive market.

Even at Just 5% Membership Penetration

20 million × 5%
= 1 million members

If each member uses

1,000 yen per month

in purchases or services:

1 million × 1,000 yen
= 1 billion yen per month

That would create

an economic sphere worth 12 billion yen annually.

Network Effects

A defining feature of this model is that

the more communities it includes, the more valuable it becomes.

Occupational communities
↓
Purchasing network
↓
Lower product prices
↓
Member growth

This creates a classic

network effect.

The Next Form of the Cooperative

Traditional co-ops were built like this:

Region
↓
Collective purchasing

But in modern society,

occupational communities

can, in some cases, form even stronger bonds than geographic ones.

Driver Life Plus represents the culmination of our philosophy:

Occupation
↓
Community
↓
Economic sphere

It is a project to build a

blue-collar cooperative for the AI era.

9. Why the Community Economy Becomes Possible Now

— A Technological Shift: From Attribute-Based Analysis to Vector-Based Analysis —

In building Driver Life Plus,
we made a major technological shift.

That shift was:

from attribute-based analysis to AI-driven vector-based analysis.

The Limits of Attribute-Based Big Data

For many years, the IT industry relied primarily on data such as:

Age
Gender
Occupation
Region

These are

attribute-based data points.

But this approach cannot adequately explain human behavior or values.

That is because purchasing behavior is deeply influenced by

Community
Relationships
Shared values

—in other words, by

relational data.

Google Proved the Power of Relational Data

Google, one of the world’s greatest success stories in big data, solved this problem early on.

Its search engine dramatically improved search quality by analyzing:

Pages
↓
Links
↓
Importance

This is

relational data.

The key idea is this:

analyze relationships, not just attributes.

That same idea later evolved into:

search engines
social networks
generative AI

AI Made the Analysis of Relationships Practical

Modern AI handles:

Words
Images
Behavior

vector spaces.

This is, fundamentally,

a way to treat relationships mathematically.

In other words, AI can now directly analyze structures such as:

People
↓
Behavior
↓
Values

The Era in Which the Guild Economy Becomes Feasible

The model we had long envisioned—

Community
↓
Relationships
↓
Economic sphere

—was, for many years, extremely difficult to reproduce technologically.

But with AI-driven

vector analysis,

we can now work with:

Community
↓
Behavioral data
↓
Value clusters

Which means:

we have entered an era in which even non-Big Tech companies can handle relational data at scale.

Technology Has Finally Caught Up with the Market Structure

What we have pursued for years—

the community economy

—has only now become truly implementable because of advances in AI.

Its structure is:

Community
↓
Behavior
↓
AI analysis
↓
Market formation

This is

the technological foundation of the guild economy.

10. The Next Phase

As these validation efforts advance,
communities will no longer function merely as groups.

They will begin to function as

economic units.

And from there, the following market structure will emerge:

Community
↓
Economic network
↓
AI analysis

11. Final Vision

We are moving away from an economy in which giant platforms dominate individuals,

toward an economy formed through:

Community
↓
Network
↓
AI

That is,

a community economy.

CO Create is building

the infrastructure for that community economy.