FlowsBook a callSign inCreate free accountSign up
Population and demographics

Comparing cities by people per business, fairly

Residents per dentist, gym or bakery is a fast way to rank cities for expansion. How to measure every city on the same footing, and why water, centres and commuters skew it.

Part of the guide Location intelligence automation

By the xMap team · · 5 min read

To compare cities by people per business, measure every city on exactly the same footing: the same shape and size of area around a comparable centre, the same kinds of business counted the same way, and residents counted from the same population source. Divide residents by businesses in each city, then rank. The ratio is only fair when the method is identical; most misleading city rankings come from one city measured over a wider circle, a centre placed on a harbour, or a commuter hub compared with a dormitory town.

This is for expansion teams choosing the next city, franchise developers ranking regions, investors screening markets and consultants writing country studies. It covers how to set the comparison up, a worked example using real business counts for Toronto and Sydney, the traps between cities that do not exist within one city, and an AI employee that keeps the ranking current.

Setting up a fair comparison

  1. Choose the kind of business exactly: dentists, not "health". A type in the places database means the same thing in every city.
  2. Choose one area rule and use it everywhere: a 10 km circle around each city centre, or every district you draw, or a drive time. Never mix them.
  3. Agree what the centre is. A city hall, a main station or the middle of the business district give different circles. Pick one rule, and move a centre that falls on water or a park onto land.
  4. Count residents for the same area with Demographics, and businesses with Nearby places or the counts on a market page.
  5. Divide residents by businesses. Add residents per square kilometre beside it, so dense and spread-out cities can be read together.

A worked example: dentists in Toronto and Sydney

Both cities have population figures in xMap's grid, and both have a market count of dentists. xMap's places database lists 830 dentists within 10 km of the centre of Toronto and 768 within 10 km of the centre of Sydney (November 2026).

830Dentists, Toronto
768Dentists, Sydney
4.7Average rating, both cities
Dentists within 10 km of each city's centre, xMap's places database, November 2026; the ratings are of the few hundred nearest each centre.

On counts alone the cities look almost the same. The ratio needs residents for those same two circles, which you would read from Demographics. To show the arithmetic, suppose (these population figures are invented) the Toronto circle held 1,250,000 residents and the Sydney circle 950,000. Toronto would come out at about 1,506 residents per dentist and Sydney at about 1,237. Sydney would look more crowded with dentists.

Now look at the map before believing it. Toronto's circle runs out over Lake Ontario on its southern side, and Sydney's is cut by the harbour and the ocean. Water holds no residents and no clinics, but it shrinks the land in each circle by different amounts, so the residents per square kilometre of land, not of circle, are what make the two comparable. And both centres pull in office workers who visit a dentist near work: neither circle's residents are the whole customer base.

Traps that only appear between cities

Listing depth differs by country

A places database can list businesses more completely in one country than another, and small practices are more often missing where fewer of them are online. When one city looks badly underserved, check a second kind of business there too. If everything looks scarce, suspect the listing before the market.

The total and the nearest are different groups

On market pages, the total is every place within 10 km, but chains, independents, ratings and openings are worked out from the few hundred nearest the centre. People per business must use the total, never the nearer group.

Commuter cores and dormitory towns

A city whose centre fills with commuters every day has more businesses per resident than one whose people work elsewhere. Comparing a capital's core with a suburb town will always flatter the town. Compare cores with cores, and whole metro areas with whole metro areas.

Different habits

In some cities people shop at big stores far from home; in others they walk to small shops. The same people-per-business figure can mean a gap in one city and normal life in another. Treat the ranking as a shortlist for a closer look, not as a verdict.

Keeping a city ranking up to date with an AI employee

In xMap Flows a monthly ranking is one AI employee:

  1. Schedule: the first day of each month.
  2. Area of interest: the cities, each a circle of the same size, for example Paris, Lyon, Toronto, Montreal, Mumbai, Bangkok and Cairo.
  3. Demographics: totals for each area.
  4. Nearby places: the kind of business you care about, with counts for each area.
  5. Calculated field: residents_per_business = round(population / max(nearby_count, 1)).
  6. Sort and limit, highest first, then Email or Make a report.

What lands in the inbox: a table of cities ranked by residents per business, with the residents, the count and residents per square kilometre beside each, and a note when a city's figures moved since last month. A city outside the covered countries simply comes back with no population, and the step says so, rather than a made-up number.

Questions

Can I compare London or New York this way?

Not with xMap's population grid, which does not cover the United Kingdom or the United States. You can still compare business counts there, or bring residents from another source into a table and mark where they came from.

What size of area should I use for each city?

One that fits how people use the business. A 10 km circle around the centre works for a first screen of big cities; drive times or drawn districts are better for a final choice.

Why do two cities with the same count rank differently?

Because the residents in the same circle differ, often because water, parks or industry take up part of the area. Look at residents per square kilometre of land as well.

Is a high number of people per business always a gap?

No. It can also mean people shop elsewhere, the listing is thin, or the business is spread in a way the circle misses. Use it to choose where to look closer.

Templates that do this

Read next

More on population and demographics · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.