Trade area demographics describe the population and households within a location’s relevant market. To compare opportunities, examine the distribution of those households, the number matching your customer criteria, and the geography they can realistically reach. Average income or median age alone cannot establish demand for a store.
Suppose a retailer’s customer research points to a particular household income range. Its expansion team finds two markets with the same average income. Both pass the initial screen, but the retailer still does not know how many households fall within that range.
Better real-world context moves the team from a summary statistic to a relevant population, and then to the commercial conditions around it.
The same average can describe very different markets
Consider two deliberately simplified, hypothetical markets. Each contains 10,000 households. Every household has one of the exact annual incomes shown below, so the means can be checked directly. These are teaching examples, not Factori market findings.
| Measure | Market A | Market B |
| Household income distribution | 5,000 households at $60,000; 5,000 at $120,000 | 8,000 households at $50,000; 2,000 at $250,000 |
| Total households | 10,000 | 10,000 |
| Mean household income | $90,000 | $90,000 |
| Share with income of at least $150,000 | 0% | 20% |
| Households with income of at least $150,000 | 0 | 2,000 |
In Market A, the mean is halfway between $60,000 and $120,000. In Market B, the weighted mean is 80% of $50,000 plus 20% of $250,000. Both equal $90,000, yet the number of households above the example threshold is entirely different.
That does not make Market B the better location. The threshold itself needs support from customer research, and income does not establish purchase intent. It does show why the average cannot answer a question about the size of a particular segment.
Using the median can help describe a skewed distribution, but it still supplies one summary value. To count households in an income range, use the relevant distribution. The US Census Bureau, for example, publishes household income bands in ACS table B19001, separately from median household income in B19013.
The same principle applies to age and household composition. A median age does not tell you how many families with young children live in a catchment. Choose the measure that answers the business question.
Audience concentration and audience size answer different questions
A high percentage can indicate a strong local concentration while concealing a small total audience.
Take a separate hypothetical comparison. Here, the team is evaluating households with annual income of at least $150,000 across two catchments:
| Measure | Smaller catchment | Larger catchment |
| Total households | 20,000 | 80,000 |
| Share meeting the income criterion | 35% | 18% |
| Households meeting the criterion | 7,000 | 14,400 |
The smaller catchment has the higher concentration. The larger catchment has more than twice as many households meeting the criterion. Neither result is a forecast of how many will visit or buy.
The calculation is straightforward: total households multiplied by the relevant household share. Both figures must refer to the same population, geography, and period. Multiplying a household percentage by a resident count changes the unit and produces a misleading result.
Use concentration to understand how common the audience is. Use count to understand its potential scale. Compare reachability, costs, and customer behavior before deciding which market deserves investment.
Separate demographic percentages do not reveal a combined audience
An attribute list is not the same as a segment you can measure.
Suppose a dataset reports that, among the same 10,000 households, 40% rent their homes and 30% have income above $100,000. Multiplying 40% by 30% gives 12%, or 1,200 households. That calculation assumes the two characteristics are independent.
The supplied percentages do not establish independence. With those totals alone, the number of high-income renter households could be anywhere from zero to 3,000. Those are mathematical bounds, not estimates of what is likely.
To narrow the answer, the team needs a suitable cross-tabulation or a validated model of the overlap. A cross-tabulation reports the attributes together for the same underlying population. Even then, the estimate may have sampling uncertainty or reporting limits.
When evaluating aggregated people data, ask which attributes can be analyzed jointly. If the required overlap is unavailable, keep it as a missing input or an explicitly labeled assumption. Do not present a product of unrelated percentages as a measured segment.
Choose the geography before ranking the audience
The relevant population depends on where customers can reasonably travel and what other destinations they might choose. A postal area, city boundary, radius, and drive-time catchment can contain very different populations.
For a store, start with the format and likely journey. A walk-up convenience offer and a destination furniture store are unlikely to draw from identical areas. For an existing location, customer and visit-origin evidence can help assess the catchment. A proposed location needs assumptions informed by comparable stores and access conditions.
Use trade area analysis to make those boundaries explicit. Test whether the market ranking changes under plausible alternatives. If a small boundary change reverses the result, the team should investigate that sensitivity before treating the ranking as settled.
Also keep the population definition clear. Resident households, daytime workers, observed visitors, and a retailer’s customers are different groups. Nearby residential demographics do not establish the demographics of everyone who visits a store.
A smaller map cell is not necessarily a better estimate
Demographic information is often collected or estimated for standard geographic areas. To describe a custom catchment, an analytical system may allocate portions of those estimates to the new boundary. That process adds assumptions about where people and households are distributed.
Esri documents how data apportionment is used to calculate demographic information for custom areas and notes limitations when summarizing below the original geography. Inspect the source resolution and allocation method; do not judge precision from how detailed a map looks.
A comparison worth trusting should retain:
- The population being described, such as residents or households.
- The source period and original geography.
- The method used to estimate values for the requested area.
- Missing, suppressed, or modeled fields.
- Available uncertainty measures and important source changes.
When comparing ACS estimates, the Census Bureau advises considering margins of error. A small difference between two estimates may not support a confident ranking. Other data sources have their own uncertainty and comparability requirements.
Connect the population to evidence of demand
Demographics help a team identify a plausible audience. To evaluate a retail opportunity, add evidence about how that audience might interact with the location.
For a premium homeware retailer, the useful questions could include whether comparable stores serve similar catchments, which destinations compete for shopping trips, whether the premises are accessible, and how much demand may overlap with the existing network. Household income is one input to that analysis.
Mobility data can add context about activity and journeys. Place data can identify nearby competitors, anchors, and complementary businesses. First-party transactions and customer research help test whether the external conditions are associated with the outcome the retailer cares about.
Preserve the limits of each source. Joining an area’s demographics to visit-origin data produces information about origin areas; it does not prove the attributes of each visitor. Likewise, several overlapping catchments cannot simply be added together to calculate an unduplicated audience.
The goal is a better-supported market comparison. A sales forecast requires additional modeling and validation.
A practical framework for comparing trade areas
Build a short decision brief for each candidate using the same definitions:
| Question | Evidence to include |
| How large is the relevant audience? | Count, denominator, attribute definition, and uncertainty |
| How concentrated is it? | Share of the local population and comparison with a relevant benchmark |
| Can we measure the required segment? | Joint distribution, validated model, or an explicit gap |
| Is the geography appropriate? | Catchment method, travel assumptions, and boundary sensitivity |
| What supports demand? | Comparable-store outcomes, access, activity, competition, and network overlap |
This brief gives the team something more useful than a “best demographics” score. It explains why a market merits investigation and which missing evidence could change the decision.
Give AI the context behind the demographic fields
The same discipline matters when an AI agent prepares the comparison. Provide the population unit, source period, geography, and field definitions. Make clear which values are estimates and which attribute combinations the data actually supports.
A good request is: “Compare these catchments using both audience counts and shares. Check that denominators match. Use joint distributions for combined segments where available, and flag missing overlaps. Explain which differences are supported by the data and which need further validation.”
Better context makes the analysis more specific while reducing the need to fill gaps with assumptions. The model still needs to apply the definitions correctly.
Build the market comparison with Factori
Factori People Data provides population and audience context for market planning. Teams can use available demographic and socioeconomic fields alongside other real-world datasets and their own business information. Factori’s MCP server provides compatible AI assistants with access to its workspace.
Start with the markets and the customer question. Check the available attributes, joint breakdowns, geographic resolution, and source periods before choosing the analysis. Field availability and suitability need to be established for the market under review.
Bring Factori the markets you are comparing, your audience criteria, and the decision you need to make. Explore the population context available, then connect it to the places, movement, and business evidence that make the comparison useful.
About Factori
Factori is a leading global location intelligence company that provides unmatched data insights to help businesses better understand the physical world:
Factori datasets are governed, privacy-safe, and structured to join seamlessly with your existing workflows across SQL, data warehouses, BI tools, and ML pipelines. With over 90B+ location signals collected every day across 150+ countries, Factori delivers broad market coverage and reliable location intelligence at scale. Datasets are available via APIs, raw data, the Factori platform, and MCPs to support different use cases and markets.
FAQ
What are trade area demographics?
The population and households within a location’s relevant market. Read them by distribution, matching-household counts, and reachable geography, not by averages alone.
Why can two markets with the same average income be different?
The distributions differ. In the worked example both average $90,000, but one has 0% of households above $150,000 and the other has 20%.
What is the difference between audience concentration and audience size?
Concentration is the share of the local population that matches; size is the total count that matches. They answer different questions.
Can you multiply two demographic percentages to size a segment?
No. That assumes independence. Use a cross-tabulation or validated overlap model; otherwise the true count can range widely (in the example, 0 to 3,000).
Does a smaller map cell mean a more accurate estimate?
No. Apportioning data to custom areas adds assumptions. Check the source resolution and method, and consider margins of error.
