TM
A World Model of Any Population
Populus builds a faithful replica of the people you need to understand: a market, your members, a workforce, or a whole country. Every modeled person sits in a household and a real place, with the correlations intact. Add any credible data source, bring in your own, and run a decision forward to see what drives the outcomes. The person who owns the decision can question the answer, change the assumptions and run it again. Test the move before you make it.
Trusted by
The Population Behind Your Data
Most data describes the people you have already reached. Populus models the whole population they belong to: the people in your data, and the people who aren’t.
People modeled in the United States today, at individual resolution
Individual-level representation behind every aggregate, with the correlations right
Rebuilt every year against the newest census and survey data
From question to decision-grade answer
The Population: Every Person, In Place
Privacy locks away the data that describes individuals, so only fragments get published: aggregates, marginals, samples. Populus rebuilds a coherent population out of those fragments, using a method peer-reviewed and in use for fifteen years. Every modeled person is consistent with everything known about where they live: age, income, race, education, household structure, health, industry.
The reconstruction is the part that travels. It works wherever reliable census and survey data exists, so a population can be a country, a market, or one organization’s people. The United States is live today at full depth, and new populations are built on demand.
Any dataset can give you the right number of diabetics. Populus puts them on the right incomes, in the right households, in the right neighborhoods, all at once.
Add Any Credible Data to Every Person
Name a national survey and Populus projects it onto the full population, with the statistical decisions built into the framework rather than made fresh each time. National health surveys are live today, including BRFSS, NHANES and HINTS. Upload your own research and the same machinery applies, so your sample of 10,000 becomes insight about entire markets. Every source added makes the next question cheaper to answer.
Your Data Becomes a Population
Import your customers, members or employees and they become a population inside the same world, which you can enrich with everything Populus already knows. Where your data is better than ours, it replaces ours. Where we need to check our work, it calibrates the model against what you have actually observed. Export anywhere: Snowflake, Databricks, Postgres, or local files. Enterprise deployments run inside your own cloud, so your data never leaves your environment.
One platform for exploration, unlimited destinations for analysis.
Run the World Forward
This layer is built on the framework that modeled epidemics for public health agencies. Test a launch, a price, a coverage change, or a shock. Populus runs thousands of futures under explicit assumptions and ranks the factors that separate the good outcomes from the bad ones, so you know what to control and what to watch. See what a coverage change does to a book of business before an actuary commits to the assumption.
We don’t predict the future. We show you what drives the differences between futures.
Question the Answer and Run It Again
Built for Decision-Makers and the Teams That Support Them
Size Any Market
Segment by demographics, geography, or your own attributes, and size any population in seconds, from national markets to neighborhood niches.
See It Instantly
Explore population patterns through intuitive maps, charts, and dashboards.
Push to Your Stack
Export enriched populations straight to Snowflake, Databricks, or a local file.
Test Before You Act
Run interventions and what-if scenarios. Get the range of outcomes and the ranked drivers, not a single guess.
Privacy by Construction
Modeled people, not masked people. There are no real identities to expose, and no way for anyone to look one up.
Trust the Answer
Peer-reviewed methods, transparent assumptions, and outputs your own data scientists can validate.
