How to find areas with older residents, hexagon by hexagon
Map the share and the number of people over 60 in small hexagons across a city, pick out the clusters worth acting on, and know what the age data cannot tell you.
Part of the guide Location intelligence automation
To find areas with older residents, split the city into small cells, count the people aged 60 and over in each, and look for places where both the share and the number are high. A population grid of hexagons of about 0.1 km² gives exactly that, in nine age groups, so you can map the over-60s, or the over-80s alone, street block by street block rather than district by district.
Home care agencies, senior living developers, councils planning day centres, pharmacies choosing delivery zones and charities placing volunteers all start here. This article covers the two numbers to read together, how to set the run up, how to turn a map of hexagons into a short list of areas, and the things people often assume the data says that it does not.
Share and count: read both
The share of over-60s in a hexagon can mislead. A hexagon with 40 residents, 20 of them over 60, shows 50%. A hexagon with 2,000 residents and 400 over 60 shows 20%, and holds twenty times as many older people. Ranking on the share alone sends you to quiet edges of the city; ranking on the count alone sends you to the densest towers, where older people may be a small minority.
- High share and high count: the clearest areas with an older population. Start here.
- High count, ordinary share: large, mixed neighbourhoods. Many older people, but services shared with everyone else.
- High share, low count: small pockets, often a care home or a block of retirement flats. Worth a look on the ground before planning anything.
A simple rule that works in most cities: keep hexagons with at least a few hundred residents, then rank by the number of people over 60. That drops the noisy, nearly empty cells without losing the dense ones.
Setting up the run
- Draw the part of the city you care about with Area of interest. A whole city works if it fits; an area larger than about 4,200 km² needs to be drawn in parts.
- Add Demographics with Give back set to Every hexagon. Each hexagon comes back with its centre, residents, the nine age groups, seniors_60_plus and seniors_pct, up to 5,000 hexagons a run.
- Add a Filter: population at least 300, or whatever floor suits your city.
- Add a Calculated field for the over-80s if that is your group: age_80_plus divided by population.
- Send it to Show on a map as a heat map weighted by seniors, and to a sheet or email as a table sorted by count.
From hexagons to areas you can name
A heat map of 3,000 hexagons is good for a first look and bad for a decision. Teams need a handful of places with names. Two ways to get there:
- Look at the map and draw shapes around the three to five warm clusters, then run Demographics again with Totals for each area. Each drawn shape comes back as one row with its residents and over-60s.
- Ask xMap LLM, set to the whole list, to group neighbouring hexagons with high counts and describe each cluster from the figures, without adding any number of its own.
A worked example, made up
Say a home care agency in Lyon wants to know where to recruit carers so they live close to clients. It draws the city and its inner suburbs, asks for every hexagon, keeps those with 300 residents or more, and ranks by people over 60. The top of the list falls into four clusters: two older residential districts, one riverside area with several retirement residences, and one suburb built in the same decade, whose first residents have aged together. The agency puts its recruitment adverts in those four areas and gives each coordinator one cluster.
What the data does not tell you
- Coverage. The grid covers about 100 countries, among them France, Canada, India, Thailand, Egypt and Australia, but not the United Kingdom or the United States.
- Age groups are ten years wide. "60 and over" mixes active 62-year-olds with people in their nineties. Use the 70 to 79 and 80 and over groups when frailty matters.
- Residents only. Older people who come into an area for a hospital or a market are not counted where they visit.
- Care homes. A single large residence can light up a hexagon. Check what is there with Nearby places, searching for nursing homes and retirement homes.
Put next to what is already there (clinics, pharmacies, community centres, bus stops), a map of older residents becomes a map of gaps, which is where most of the decisions are made. For the wider method, see location intelligence automation.
Questions
What size are the hexagons?
About 0.1 km² each, on the H3 grid at resolution 9: roughly a few street blocks in a dense city.
Can I map only people over 80?
Yes. Every hexagon carries age_80_plus. Use it directly, or work out its share of residents with a Calculated field.
Can the data show older people who live alone?
No. It counts residents by age group and sex only. Household make-up, health and income are not part of it.
Why does one small hexagon show a very high share?
Usually because few people live there, or a care home sits in it. Set a minimum number of residents and check what is there before acting on it.
Templates that do this
Read next
More on population and demographics · Every article · Templates