How to extend your customer data with public data

You already know a lot about your customers. Their names, their postal codes, what they bought and when. What that record does not tell you is where the next thousand customers are, or how many people in a region look like the ones you already serve. That gap is where a lot of marketing spend goes to audiences that were never a strong match.

Public data closes the gap. Government statistics agencies publish detailed figures about who lives in every part of the country: income, education, age, language spoken at home, housing, and internet access. Joining that to what you already hold turns a list of past buyers into an answer to a harder question: where else do people like this live, and how many of them are there? This is the everyday use of public data for a marketing team.

What most teams start with

Most teams plan regional campaigns from a blend of client data, internal knowledge, and platform targeting. The customer list is solid on the people who have already bought. Beyond that, it goes quiet. You can see that your best customers cluster in a handful of postal areas, but the list alone does not say what those areas have in common, or which other areas share the same make-up. So the lookalike step happens inside an ad platform, using the signals it holds.

What public data adds

Start with the areas your best customers already live in. For each one, public data such as the American Community Survey, the annual household survey the US Census Bureau has run since 2005, describes the resident population in detail: median household income, education level, age distribution, language spoken at home, and how many households have a home broadband connection. Cambium AI reads those figures at the local-area level, so the description fits the neighbourhoods you actually sell into and not a state average.

Say you run a regional homeware brand with 6,000 customer records, and your strongest sales come from a dozen postal areas. Read the public data for those areas, and a pattern usually shows up: households in a particular income band, a high share of owner-occupied homes, and a median age in the late thirties to mid forties. That profile is the thing the customer list was missing. It describes the buyer you have already converted, in figures anyone can check.

From profile to audience

Once you have the profile, you can find every other area in the region that matches it and count the households there. That is your lookalike audience, sized in real households. The same reading shows the gaps: areas that fit the profile where you have no customers yet. Those are the places a regional campaign should reach first.

The county-level work Cambium AI publishes shows the method in the open. How income and poverty split across the counties of a single state (see the Minnesota county income and poverty breakdown), or how education levels change from one county to the next (the most and least educated counties in America), uses the same public data you would read to profile and size your own audience. The CEO demographics post reads one narrow slice of the population the same way.

Check the profile before you spend on it

A profile is only worth acting on if you can trace where each figure came from. Before a lookalike audience drives a budget, put it through the same test you would apply to any segment: what is the data source, how large is the sample, what is the geographic resolution, and when was it last updated? Those questions are set out in the guide on how to verify a persona before you spend on a campaign. Because public data is published and documented, every figure has an answer.

This is where a public-data profile earns its place in a real decision: it describes a population you can defend in the room where the budget gets signed off, with a source under every figure. For a fuller walk-through of using the survey for research, see the guide on using the American Community Survey for market research.

The short version

Your customer data tells you who you already have. Public data tells you who else looks like them, where they are, and how many. Put the two together, and a regional campaign starts from evidence rather than a guess about which areas to target.

Get in touch to learn how we can enrich your data with public data.