Pull your first 120 companies from Google Maps

1. 1

   Start a new research project

   From the dashboard, click New Research. Name the project after the market and the date, not after the client. You will build several versions of this list and you want to tell them apart later.

   

   1

   2

   3

   1. 1Describe the list in plain words. This is the fastest way in and it decides everything downstream.
   2. 2Quick start, when you already know the motion. Upload a CSV, find people, find companies, deep search, or analyse a website.
   3. 3Browse all builders, when you want to pick the source yourself rather than describe it.

2. 2

   Select Google Maps as your source

   Choose Google Maps from the data source options. If you are unsure whether your market has a physical presence, this is the cheapest way to find out.

3. 3

   Describe the business in plain words

   The field asks what business you want to extract. Write it the way you would say it out loud, category then place. Plain phrasing beats clever phrasing: write "bookshops in Japan", not "search for bookshops in Japan".

4. 4

   Several locations in one search share one total

   One request can carry several locations, and they can even carry different searches, like cafes in Paris and clinics in Lyon. What they share is a single total: if the first location fills your row count, the ones after it get what is left, which may be nothing. Split the run when you need a guaranteed number from each city, and keep them together when you just want the total.

5. 5

   Ask for more than you want to keep

   Twenty is the minimum the field accepts. Above that, ask for roughly three times the number of companies you want to end up with, because you are about to cut this list hard. For a 40 row target, ask for 120.

   

   1

   2

   3

   4

   5

   1. 1The price, on the builder itself. Sourcing is the cheapest step in the whole loop.
   2. 2What business do you want to extract. Category and place, in plain words.
   3. 3How many. This is the number that decides whether your filter later has anything to cut.
   4. 4The rule the field does not spell out until you break it: twenty is the minimum.
   5. 5Best practices, from the product itself: one location at a time, city by city, plain phrasing.

6. 6

   Extract, and watch the table fill

   Click Extract. Rows populate in real time. Each one arrives with business name, address, phone number, website, rating, review count, category and opening hours.

7. 7

   Then run it again, differently

   Two more searches, and neither is a repeat. One with a different category phrasing, because solar installers and solar panel companies return overlapping but different sets. One for the next city. Merging them is course 2.3.

What good output looks like

Look at row three. No website and nine reviews. That row will fail the filter in the next lesson, and because you have not enriched anything yet, cutting it costs you nothing and saves you the five credits it would otherwise have consumed.