How to get a list of all restaurants in a city, and how complete it can be
Why one search never returns every restaurant in a big city, how to cover it in parts without duplicates, which types to include, and how to keep the list current each month.
Part of the guide Location intelligence automation
You can build a list of all restaurants in a city that is close to complete, but not with one search. A large city has thousands of restaurants, while a single search returns the few hundred nearest its centre point. The method is to split the city into parts, search each part for every restaurant type you care about, merge the results and remove duplicates, then keep the list fresh with a monthly run that adds new places and flags closed ones.
This guide is for food suppliers, delivery platforms, point-of-sale and booking software sellers, waste and linen services, and anyone else who sells to restaurants or studies them. It shows how big the task is with real figures, how to split a city, which types to include, a worked plan, and the traps that leave holes in the list. The location intelligence automation guide covers the underlying data.
How many restaurants are we talking about?
Those are counts for the type Restaurant within 10 km of the centre, not the whole metropolitan area, and they already exceed what one search hands back. In the Nearby places block, one search is one type around one point, up to 25 km away and the 500 nearest. When a search reaches its 500, the step tells you there may be more. In central London a 500-place search might reach only a few kilometres from its point before it is full.
Decide what counts as a restaurant
Before splitting anything, settle the definition, because it changes the size of the list more than anything else. xMap's places database sorts places into more than 5,000 types. Restaurant is one of them; fast food restaurants, sushi restaurants, Mexican restaurants, kebab shops and many more each have types of their own, and cafes and coffee shops sit in a separate category. xMap's places database lists 2,573 cafes and 1,640 restaurants within 10 km of the centre of Sydney (November 2026): leave out cafes and the list shrinks by more than half.
Write down the types you need. A linen supplier wants sit-down restaurants and hotels; a delivery platform wants fast food and takeaways too; a coffee roaster wants cafes and nothing else. Nearby places lets you pick a whole small group of types at once, such as all the Mexican restaurant types, and up to 12 types a block, each type being its own search.
Split the city into parts
There are three practical ways to cover a big city:
- By district. Put the city's districts in an Area of interest. Nearby places searches each area from its middle to its edge and keeps only places inside it, up to 8 areas in one run. Districts make the list easy to read later, because each place carries the area it came from.
- By points. Draw points across the city with a distance around each, so the circles overlap slightly. Dense centres need points closer together than quiet suburbs.
- By turning on Cover the whole area. For one wide area, Nearby places then searches its middle and six spots around it instead of a single point, so more of the area is reached in one go.
Overlapping parts find some restaurants twice. Remove duplicates, set to the same place, keeps one of each.
A worked plan
This plan is made up, using the London count above as a sense of scale. A wholesaler of kitchen equipment wants every sit-down restaurant in inner London.
- List the inner districts and split them into runs of up to 8 areas each.
- Pick the types: Restaurant plus the cuisine groups the wholesaler sells to, leaving out fast food and cafes.
- For each set of districts, run Nearby places with Keep set to Nearest first and Most places at 500.
- Check each step: wherever a district hit its 500, split that district in two and run it again.
- Remove duplicates across all runs and save the result as a table on the My data page.
- Sort and filter the table by rating, chain or independent, and whether a phone or website is known, to make the call lists.
The finished table will still not be perfect. A place that opened this week may not be in the database yet, and some restaurants are labelled under an unexpected type. Treat it as a strong starting list, then let a monthly run catch what was missed.
Keeping the list current
A list goes stale quickly as restaurants open, change hands and close. Instead of rebuilding it, set up an AI employee that runs monthly over the same districts with Only new after the search, so only restaurants it has not seen before are added, and What changed in a second branch to flag restaurants the flow found before that have since closed or changed phone or website. The template Monthly outlet census of my city follows this pattern and writes to a sheet; Hotels and restaurants to supply, researched adds a short note per place. Each month's note in the Team inbox says how many restaurants were added and how many closed.
Holes to watch for
- One search for the whole city. You get the 500 nearest the centre and nothing from the edges.
- Only the Restaurant type. You miss every place labelled by its cuisine.
- No duplicate removal. Overlapping circles inflate the count.
- Filtering on rating too early. Places with no rating are dropped, and they are often the newest.
- Closed places left in. Nearby places leaves out places marked closed unless you switch that off; keep it on for a working list.
Questions
Is a complete list of every restaurant possible?
Close, but never perfect. No database catches every opening on the day it happens, and some places carry an unexpected type. Covering the city in parts and refreshing monthly gets you near.
Why does my search stop at 500 places?
One Nearby places search returns the 500 nearest places of a type around its point. Split the city into smaller parts so each search stays under that.
Should cafes count as restaurants?
That depends on what you sell. Cafes and coffee shops are a separate category from restaurants in xMap's places database, so add them as their own types if you want them.
Can I export the list to a spreadsheet?
Yes. Save it as a table in My data and download a CSV, or end the flow with the Google Sheets app block to write the rows straight into a sheet.
How do I know which restaurants closed?
Add What changed to the monthly flow: it passes on the places the flow found before whose details changed since, including ones that closed.
Templates that do this
Read next
More on business and places data · Every article · Templates