FlowsBook a callSign inCreate free accountSign up
Business and places data

How to search multiple business types in one go, and keep the results apart

Cafes, bakeries and juice bars in one search: how types, groups and your own words combine, the limits per block, and how to count or label each kind afterwards.

Part of the guide Location intelligence automation

By the xMap team · · 4 min read

You can search multiple business types in a single step by listing them together: each type is searched on its own around the same point or area, and the results come back as one list in which every place keeps its own type. That last part is what makes it useful. You get one combined list for a call sheet or a map, and you can still split it, count it or colour it by kind afterwards.

Most real questions need more than one type. A coffee supplier sells to cafes, bakeries, hotels and offices with canteens. A cleaning company wants clinics, gyms and offices. A planner counting daily services wants pharmacies, grocers and post offices. This article explains how types and groups work in xMap Flows, where the limits are, and three ways to keep the kinds apart once they are in one list.

Types, groups and everyday words

Nearby places searches xMap's places database, which names more than 5,000 types in about 500 groups. You pick them in the Types of place box, in three ways.

  • Exact types, as the database names them: Coffee shop, Bakery, Juice shop, Pharmacy. Start typing and pick from the list.
  • A whole small group, shown with its size, such as Gas Station (all 6 types). Groups of up to 12 types can be picked as one entry, and they are opened into their types.
  • Everyday words, such as cafes, petrol stations or salons, which the block turns into the matching types. The hint under the box shows what it will search, and names any word that is not in the list.

Up to 12 types fit in one block. Each type is one search for each place or point, so six types around one office is six searches, and a group of six types counts as six. Leaving the box empty is a different thing: it searches every kind of place in one search per point, nearest first, which is handy when you want to know what is around a spot rather than find a particular kind.

Searching online maps for several kinds

Map business scan works the same way with its own list of categories, and also takes your own search words for anything the list lacks, such as a brand or "tile showroom". Each category or word is a separate search in each area, up to 8 areas and 96 searches a run. Its setting Only the kinds searched is on by default: it drops places whose category does not match what you asked for, so a search for clinics does not keep a fortune teller next door.

Three ways to keep the kinds apart

  1. Count each kind around your own places. Set Nearby places to Each incoming place and Pass on to Each incoming place, with counts. With two types or more, every one of your places gets a count per type, such as nearby_cafe_count, as well as the total. This is the right shape for comparing sites.
  2. Group the list into your own kinds. Add Place classifier with a custom list, so Coffee shop, Espresso bar and Tea house all become Cafe and the list can be counted in three or four buckets.
  3. Send each kind to a different person. Add Branch or AI router after the search: cafes to one rep, hotels to another.

A made-up example in Bangkok

This example is invented. A dairy distributor in Bangkok wants new outlets for milk and cream within 3 km of its depot. The buyers it serves are coffee shops, bakeries, dessert shops and hotels. The flow searches those four types around the depot with Most places set to 150, and Only places with a phone or website switched on.

Say the run returns 410 places in one list. Place classifier turns them into Cafe, Bakery, Dessert and Hotel; a sort puts the best rated first within each; and the weekly email shows four short sections instead of one long list. Because each place still carries its database type, the team can see that most of the dessert results are bubble tea shops, a kind it had not planned for, and adds a line for them to its pitch.

Mistakes worth avoiding

  • Using a broad word where a type exists. A word that is not in the list is searched as written and usually finds little.
  • Adding twelve types "just in case". Every type is a search; add only the ones a person will act on.
  • Forgetting duplicates. A place can match two of your types, or two searches can overlap. Remove duplicates by the same place keeps one copy.

Combining types well is a quiet skill behind most good lists in location intelligence automation, and the templates below show it done for welcome offers, outlet censuses and supplier searches.

Questions

How many business types can I search in one block?

Up to 12 in Nearby places. For more, add a second block and join the two lists, removing duplicates.

Can I search every kind of business at once?

Yes. Leave Types of place empty and Nearby places searches every kind around each point, nearest first, each place with its own type.

Do several types cost more than one?

Each type is its own search, so more types mean more searches. Counting around your own places passes on no places, so it costs only the searches.

Can I mix database types and my own words?

In Map business scan, yes: pick categories and add your own search words in the same block. Nearby places searches the database's types, and turns everyday words into them.

Templates that do this

Read next

More on business and places data · 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.