What's inside

Hiasynth holds one synthetic person for every real resident of Europe. Each one lives in a household, and each household sits in a specific neighbourhood with its own shops, climate and commute. Ask about any slice of it and you get a count, a profile or a persona computed from the whole population, in seconds.

Three layers, one record

Every attribute can filter, group or describe every other, because the three layers are joined rather than averaged. That is what makes "renters in their thirties with children, within walking distance of a tram, who spend above their neighbourhood on eating out" a single question.

The person. Age, sex, education, work and occupation. Income and wealth. Origin and cultural identity. Around sixty attitudes and values: trust, priorities, outlook, politics, how they feel about their life. Media habits, paid subscriptions, brand affinities. How they commute and how their week is shaped. The chance they move, change job, partner or have a child in the next three years.

The household. Composition and size. Gross, net and equivalised income, tax and transfers. Wealth, debt and what servicing it costs. Tenure and housing costs. A complete spend budget across 47 categories, with an online share for each. Cars and their powertrain, heating system, solar, a second home. Financial resilience: months of reserve, what a rate shock or a job loss would do. Emissions per category.

The place. What it is like to live in that square kilometre. Distance to the city centre, the nearest station, grocery shop, school, hospital and fast charger. Restaurants, cafés, bars and shops within walking distance. How the area fills up or empties by day. Land use, building age, climate. Who the neighbours are, from the census grid.

Coverage

One method and one schema for every build, so any two places in the list are answered the same way. Where a country has no source for something, that attribute is left empty and marked, never filled in from a neighbour.

What people use it for

Sizing and locating a market. Describe your customer in attributes and get how many there are, where they are densest, and where they are over-represented against the surrounding population. Rank cities, regions or countries on the same definition.

Understanding a place. Who lives within three kilometres of a site, what the area is like by day, what the households there earn, own and spend. Compare two locations on the same terms.

Reaching an audience. Which channels, subscriptions and media reach a segment, at rates that match measured usage. Which values and messages fit.

Personas with a budget. Sample representative people from any segment. Each comes with a household, a neighbourhood, a wallet and values that make sense together.

Comparing across Europe. One method and one schema for every country, so Lisbon and Luleå are answered the same way. Money is kept in local currency and compared by rank or purchasing power.

How you use it

We've opened the population to AI assistants through an MCP server. Connect it to Claude, ChatGPT, Claude Code or any client that speaks the Model Context Protocol, and ask in plain language. Your assistant picks the tool, runs the query and explains the answer. Getting started takes a few minutes.

Every answer is an aggregate or a synthetic persona. There is no real person behind any row, so results can be used, shared and published without a privacy obligation. Accuracy covers what the numbers can carry and where they run thin.

Need something specific?

Higher volumes, a private deployment, your own attributes joined in, or a tool set built for one job: write to customise@hiasynth.co.