FlowsBook a callSign inCreate free accountSign up
Site selection and expansion

White space analysis: finding areas with many residents and few competitors

White space analysis finds where many people live and few businesses like yours serve them. Three ways to run it, from one question to a city-wide grid, and how to read it.

Part of the guide AI employee for site selection

By the xMap team · · 5 min read

A white space analysis looks for parts of a market where demand is high and supply is thin: in location terms, places where many people live and few businesses of your kind are within reach of them. You measure two things on the same map (residents, and competitors) and look for the cells where the ratio of the first to the second is highest. Those cells are candidates to visit, not answers; but they turn a city of a thousand streets into a short list.

This guide is for expansion, franchise and strategy teams, and for anyone opening a second or tenth site of a business people visit from home: childcare, clinics, gyms, pharmacies, grocery, tutoring. It shows three ways to run the analysis in xMap Flows, from a single question to a city-wide grid, where the resident figures exist, a made-up example, and the reading mistakes that send people to the wrong street.

What you need, and where it exists

Two layers, measured over the same areas:

  • Residents from xMap's population grid, through Demographics: people per hexagon of about 0.1 km², with men, women and nine age groups. The grid covers about 100 countries, among them France, Canada, India, Thailand, Egypt and Australia, but not the United Kingdom or the United States. Where it has no data, an area comes back as 0 and the step says so.
  • Competitors from Nearby places: the places of your types around each point, counted, with how many are chains and their average rating.

Where there is no population data, the analysis can still run on supply alone (where are there fewest competitors), but say so in the result: few competitors in an empty industrial zone is not white space.

Three ways to run it

One question: Ask the map

For a first look at one city, ask Ask xMap a plain question such as How many nurseries are there in Bangkok within 5 km of Silom, and where are the gaps? Ask the map counts the places, counts the residents, cuts the circle into nine parts and names the parts where the most people live for each place (north-east, the centre and so on), on a small map. It shows what the answer costs before anything runs. Good for deciding whether a city is worth a fuller look.

A grid over the whole city

For a real short list, measure every small cell. Demographics set to Give back: Every hexagon over the city returns one item per hexagon with its residents. Nearby places, around each incoming place with counts, adds how many competitors are within a set distance of each hexagon. A Calculated field works out people per competitor (residents divided by competitors plus one, so a cell with none does not divide by zero), and Sort and limit keeps the top cells.

A flow that runs every quarter

White space closes as others find it. The same grid, scheduled, shows which gaps are still open each quarter and which a new competitor has taken.

A made-up example in Bangkok

This example is made up. A childcare operator with six nurseries in Bangkok wants three candidate districts for its next openings.

  1. Ask the map gives a first reading for three circles of 5 km. One shows a clear gap in its north-east part; the other two are evenly served.
  2. The grid flow runs over the whole city: residents per hexagon from Demographics, then nurseries and kindergartens within 1.5 km of each of the 100 most populous hexagons.
  3. Children aged 0 to 9 are used instead of all residents, since that is who a nursery serves; the age group is one of the nine Demographics returns.
  4. People per competitor ranks the cells. The top twenty fall into four clusters; one is next to an existing nursery of the operator and is dropped.
  5. The other three clusters go to the property team, each with the count of children, the nurseries already there and their ratings.

Reading the result without fooling yourself

  • Residents are not customers. The grid counts who lives there, not who works there by day or who passes through. A business district may have few residents and many customers: for that, look at Mobility Footfall, which measures people-hours of presence rather than people.
  • Edges matter. A cell at the edge of your circle may look empty because its competitors are just outside the search. Count competitors over a wider radius than the cell.
  • Rating changes the meaning. Five poorly rated competitors leave more room than two excellent ones. Keep the average rating beside the count.
  • Zero can mean no data. A hexagon with 0 residents may be a park, a river, or a place the grid does not cover. Check the step's note.
  • Some gaps have reasons. Rent, zoning, safety and parking explain many empty cells. A site visit is still the last step.

Setting it up as an AI employee

As a flow, the analysis becomes a standing report, part of what location intelligence automation covers:

  1. Schedule: the first day of each quarter.
  2. Area of interest: the city.
  3. Demographics: every hexagon, then Sort and limit to the 100 with the most people.
  4. Nearby places: your competitor types around each, counted.
  5. Calculated field: people per competitor; then Sort and limit to the top 20.
  6. xMap LLM: a short note on each cell; Email and Show on a map as a heat map.

The quarterly email holds the twenty best cells with their residents, competitors and ratings, and a heat map. The templates where to open a nursery, neighbourhoods with many people and few notaries and where to open my next cafe are this analysis for three businesses. For the case of a brand rather than a category, see finding where a brand is missing.

Questions

Can I run a white space analysis in the United States or the United Kingdom?

Not with resident figures, which xMap's population grid does not have there yet. You can run the supply side (where competitors are fewest) and add your own demand data from a table.

How big should each cell be?

The grid's hexagons are about 0.1 km². Count competitors over a wider circle, such as 1 to 2 km for a walk-in business, so each cell reflects what its residents can reach.

Should I use all residents or an age group?

Use who your business serves: children for a nursery or school, older adults for a pharmacy or clinic. Demographics returns nine age groups for each area.

Is the area with the highest people per competitor the best site?

It is the best place to look first. Rent, access, visibility and why the gap exists decide whether it is a good site.

Templates that do this

Read next

More on site selection and expansion · 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.