Population data for humanitarian planning beyond the city limits
Find where people live in villages and the edges of towns, hexagon by hexagon, and check the grid covers your country before you plan a single distribution.
Part of the guide Location intelligence automation
Good population data for humanitarian planning tells you how many people live in each small patch of ground, including the villages, farm settlements and fast-growing edges of towns that official tables lump into one district figure. Check first that the grid covers the country and the district you care about; then draw the area, read the residents hexagon by hexagon, and plan distribution points, clinics or water around where people actually are rather than around the nearest town centre.
This is for programme officers, logistics and health coordinators at non-profits and relief agencies, and for district officials preparing a response plan. It explains why district totals mislead outside cities, how to test coverage in five minutes, how to read a heat map of residents, a made-up example, and an AI employee that keeps the figures beside your own list of sites.
Why district totals fail outside the centre
A district table might say 400,000 people. It cannot say whether they live in one market town and a ring of large villages, or scattered across hundreds of hamlets an hour apart. For a food distribution, a vaccination round or a water point, that difference decides how many sites you need, where the trucks go and how far people must walk.
xMap's population grid splits the land into hexagons of about 0.1 km2 each, with residents, men and women, and nine age groups per hexagon. Laid on a map, it shows the shape of settlement: the clusters, the empty ground between them, and the strings of homes along a road or a river. That is the picture a plan needs before anyone picks a site.
Check coverage before you plan
The grid covers about 100 countries, among them India, Egypt, Thailand, France, Canada and Australia, but not the United Kingdom or the United States. Coverage is the first question, not a detail, because a place the grid does not cover comes back as zero, and zero looks the same as an empty field on a map.
- Pick three test points in the district: the main town, a village you know is populated, and a spot you know is empty.
- Run Demographics on each with A point and a 1 km circle.
- Read the step. If the grid has no data there, it says so; the town and the village should both show residents, and the empty spot should show few or none.
- If the known village shows nothing while the step reports data for the town, treat the rural figures with care and lean on your field teams' own counts.
Reading where people live, hexagon by hexagon
Draw the district or the affected area in an Area of interest, or paste its boundary as GeoJSON. Set Demographics to Every hexagon and show the result as a heat map weighted by residents. Three things to look for:
- Clusters away from the centre. A bright patch 15 km from town is a village large enough to deserve its own site.
- Lines of settlement. Homes strung along a road or a canal suggest mobile stops along the route instead of one fixed point.
- Age mix. The 0 to 9 and 60 and over groups matter for nutrition, vaccination and mobility; a cluster heavy in either may need a different service from its neighbour.
An area larger than one search is read in parts of about 170 km2, up to 25 an area, about 4,200 km2 in all. A whole province is too big for one pass: draw the sub-districts you are responsible for and run them as separate areas. Each then comes back with its own totals as well as its hexagons.
Matching people to the sites you already have
A plan is usually a gap between people and services. Upload your existing distribution points, clinics or water points as a table in My data, and give each one the residents within 2 or 5 km with Demographics set to Each incoming place. Sites that serve few people and clusters that no site reaches stand out at once. Nearby places can add public facilities the database lists, such as hospitals or pharmacies, though small rural health posts may be missing and your own lists should win where they disagree.
A made-up example: a river district after heavy rain
Invented, to show the method. A non-profit in eastern India has four food distribution points, all in or near the district town, and wants to know who they leave out after a flood season.
| Made-up area | Residents | Aged 0 to 9 | Nearest current point |
|---|---|---|---|
| District town | 92,000 | 14,500 | In town |
| Cluster north of the river | 31,000 | 5,600 | 11 km |
| Villages along the canal road | 24,000 | 4,300 | 8 to 19 km |
| Southern farmland | 6,000 | 900 | 14 km |
Invented figures to explain the reading; not data about any real district.
The heat map shows that a third of the residents live north of the river, which the current points never reach without a crossing. The team adds one fixed point in the northern cluster and a weekly mobile stop on the canal road, and leaves the thin southern farmland to a monthly visit.
Mistakes that skew a plan
- Reading zero as nobody. Always test coverage first; a missing country and an empty field look alike.
- Planning on residents alone. The grid counts people where they live, not displaced people in temporary shelter after an emergency. Field assessments still lead.
- One huge area. Very large shapes are cut off at the part limit; split them into the units you manage.
- Straight lines for access. A river without a bridge makes 3 km a long day. Check the road network before you call a cluster served.
An AI employee that keeps the picture current
Once the method works for one district, an AI employee can repeat it each month and send the coordinator the gaps, with an approval before anything is shared with partners. The guide to location intelligence automation puts this kind of work in context.
- Schedule: the first Monday of each month.
- My data: your current sites, with names and positions.
- Demographics: residents and age groups within 5 km of each site.
- Sort and limit: the sites serving the most people first.
- Approval by the programme lead, then Email to the field coordinators with the table attached.
Two templates start close to this: Needs assessment for a new district and Impact report for donors: people around each site. The non-profits and humanitarian page lists more.
Questions
Does the population grid cover my country?
It covers about 100 countries, including India, Egypt, Thailand, France, Canada and Australia, but not the United Kingdom or the United States. Test three points in your district before relying on it.
Does it count displaced people in camps or temporary shelter?
No. It counts residents where they usually live. Use your own registrations and field assessments for people who have moved because of an emergency.
How small an area can I read?
Each hexagon is about 0.1 km2, so a village shows as a cluster of a few hexagons and a town as hundreds.
Can I compare my sites with where people live?
Yes. Upload your sites as a table and give each one the residents within a distance you choose, then sort to see which serve the most and the fewest people.
Templates that do this
Read next
More on population and demographics · Every article · Templates