How to list every bank branch and cash machine in an area
A banks and ATMs in an area list comes from the right place types, searched part by part. When to count and when to list, and how to keep it current.
Part of the guide Location intelligence automation
To build a banks and ATMs in an area list, search a places database for the bank types and the cash machine types separately, inside the boundary you care about, then merge the two and remove repeats. In xMap Flows the Nearby places block does the searching, and the same flow can either hand you the list or count what is around each of your own sites.
This is useful to bank network planners, payment companies, cash-in-transit firms, retailers deciding where a cash machine would help, and researchers studying access to cash. Below: which place types to pick, the difference between counting and listing, a made-up example, the traps, and how to keep the result up to date.
The place types that matter
xMap's places database sorts more than 300 million places into over 5,000 types. Branches and machines sit in different groups, which is helpful, because they answer different questions.
- Bank branches. The banking group has eleven types, among them Bank, Savings bank, Cooperative bank, Building society, Private sector bank and Public sector bank. Which ones appear depends on the country: building societies in some, cooperative banks in others.
- Cash machines. A separate group of four: ATM, Bank or ATM, Coin cashing machine and Crypto ATM. Most people want the first two and should leave the last two out.
- Neighbours you may or may not want. Credit unions, currency exchange services and money transfer services each have their own types. Decide at the start whether they belong in your list.
Two details shape the build. One block takes up to 12 types, and each type is one search per area, so eleven bank types and two machine types need two Nearby places blocks side by side. And a machine inside a branch may be listed once as a branch, once as a machine, or as the combined type, so expect some overlap between the two results.
Counting or listing
Nearby places can pass on two different things, and choosing well saves work.
| You want | Set Pass on to | What comes back |
|---|---|---|
| A list to map, visit or load into a system | The places found | One item per branch or machine: name, type, brand, address, phone, rating, position and distance |
| A number for each of your own sites | Each incoming place, with counts | Your sites, each with how many were found, the nearest one and how far it is |
The two outputs of the Nearby places block.
Listing suits a regulator mapping access, or a security firm planning rounds. Counting suits a retailer with 80 shops who asks which of them have no cash machine within 300 m: the shops stay as the items, each with its count, and the answer is a short sort away. The help article on counts shows the fields. Counting passes on no places, so it is also the lighter of the two.
Covering the whole area
One search returns the 500 nearest places of one type around one point, up to 25 km away. A dense city centre can hold more machines than that, and the step tells you when a search reached its limit. So split the area: put districts or drawn shapes in an Area of interest (up to 8 in a run) and each is searched from its middle to its edge, with only the places inside kept. Every place then carries the district it was found in, which makes the final table easy to total.
A made-up example
These numbers are invented for illustration. A payments company wants every branch and machine in five districts of Lisbon before planning merchant visits.
- An Area of interest holds the five districts, drawn on the map.
- One Nearby places block searches the bank types, another the types ATM and Bank or ATM, both set to The places found.
- Remove duplicates, set to the same place, joins the two results.
- Sort and limit orders them by district and brand, and Show on a map draws them, branches and machines as two colours.
Say the run finds 96 branches and 310 machines, and 22 of the machines turn out to share an address with a branch. The company now knows which streets have a machine and no branch, where its field team should start, and which brand has the densest network in each district.
Mistakes to avoid
- Adding branches and machines together as one total. The overlap at branches inflates it. Report them as two numbers.
- Reading the list as live. The database is refreshed every month. A machine removed last week may still be listed, and one installed yesterday may not be. For anything that opened in the last few days, Map business scan searches online maps on the day.
- Trusting a full search. If a step says it reached its 500 nearest, there are more: make the areas smaller.
- Expecting bank facts. The database knows a place's name, brand, type, address, rating and whether it is open. It does not know cash limits, fees or which machines take deposits.
- Forgetting closed places. They are left out unless you switch off Only places that are open, which you might want when studying branch closures.
Keeping the list current
A list of branches goes stale quietly. An AI employee keeps it fresh without a person rerunning anything, and this is where location intelligence automation earns its keep:
- Schedule runs the search on the first Monday of each month.
- Only new lets through the branches and machines the flow has not found before.
- A second branch of the flow uses What changed to report places found before that are now marked closed.
- Email sends both as one message with a CSV to download.
What lands in the inbox is short: the new machines by district, the branches that closed, and a link to the map. The ready-made template Busy places with the fewest ATMs nearby starts from the counting side instead, and Rival bank branches that closed or moved from the watching side.
Questions
Which place types cover bank branches?
The banking group has eleven types, including Bank, Savings bank, Cooperative bank and Building society. Pick the ones used in your country, or all of them.
Are cash machines listed separately from branches?
Yes. ATM and Bank or ATM are types of their own, so you can list machines without branches. A machine inside a branch may appear in both results, so remove duplicates and report the two as separate numbers.
Can I count the cash machines around each of my shops?
Yes. Feed your shops into Nearby places and set Pass on to Each incoming place, with counts. Each shop comes back with the number found within your distance and the nearest one.
How fresh is the list?
xMap's places database is refreshed every month. A monthly run with Only new shows what was added, and What changed shows places now marked closed.
Does the data say which machines accept deposits?
No. It holds the name, brand, type, address, position, rating and open status of each place, not the services of a machine.
Templates that do this
Read next
More on business and places data · Every article · Templates