Build a List of Businesses in a City: 5 Ways Compared

The request usually arrives as a number. 500 restaurants in Pune. 300 dental clinics in Leeds. 200 hardware stores across one metro. You need a list of businesses by industry and city, big enough to run outreach against, and you need it this week.

That makes it a counting problem more than a database problem. Each common way to build a list of businesses in a city stops somewhere: a results cap, an export limit, a coverage gap, or a pile of manual hours. Where a method stops matters more than its headline feature.

Most pages on this topic are library guides or pages that sell one method. This one lines up five methods against one target, five hundred businesses of one type in one city, and shows where each one runs out.

Google Maps is where most people start, because a city search for a trade shows real listings with a phone number, address, rating and review count. The trouble is the ceiling per search. Google's own documentation says Text Search (New) "returns a maximum of 60 results across all pages", with at most 20 results on each page.

So five hundred businesses in one city means many separate searches, each covering a smaller piece of the city: neighbourhoods, a grid over the map, or postal codes. Then you need a dedupe pass, because overlapping searches return the same places more than once.

If you use the API, Google lists Text Search Pro at $32.00 and Text Search Enterprise at $35.00 per 1,000 calls in the 0 to 100,000 tier, with free monthly caps of 5,000 for Pro SKUs and 1,000 for Enterprise SKUs. Done by hand, scrolling and copying into a sheet, the money cost is zero and the cost is your time. Maps listings also carry no email addresses, so contacts are a separate job.

Where it breaks for 500: the splitting and the dedupe, not the search itself. You are stitching many partial searches together and checking the overlaps by eye.

Apollo is a company and contact database, not a local map search. Its pricing page advertises "240M contacts & 30M companies" and says you can "select and export up to 10,000 leads in one click", with higher or no limits on higher tiers.

Those are totals across everything Apollo holds. A total doesn't tell you how many dental clinics in Leeds are in it. The sensible test before you pay: run your exact city and industry filter on the free plan and count what comes back. If the count is far below your target, the size of the whole database doesn't help you.

Where it breaks for 500: when your city-and-industry slice is smaller than the number you need. For this job the export limit is rarely the bottleneck. Coverage of your niche is, so check the count first.

Clay's pricing page lists a Free plan with 100 data credits a month, Launch from $167 a month with 3,000 data credits, and Growth at $446 a month with 6,000 data credits.

For a local list, the bigger question is where the businesses come from. Clay is a workflow tool that works on rows from the data sources you connect to it. To build five hundred businesses in one city inside Clay, you still need a source that can produce businesses by type and place, and that source has its own caps and costs. Budget for the Clay plan plus whatever feeds it.

Where it breaks for 500: when the source under the workflow runs out of businesses for your city. The workflow has nothing to work on.

The fourth route is a list someone else already published. Three common versions:

Where it breaks for 500: coverage and noise. Each source gives you a slice. You pay for the list, then pay again in hours removing what doesn't belong.

Mantis starts from the same public map listings as Method 1 and handles the splitting for you. "The area is split into small enough pieces to get past the per-search result cap, and repeats are removed by place identity rather than by name". It shows what it will search and quotes the credits before anything runs, and searching ground already scanned is free.

It also checks each business's website. The live counter on the Google Maps scraper page shows 4,29,819 map listings checked for a website, 1,80,240 verified with no website, and 41.2% of Indian businesses checked with no site. "No website" means the Google listing carried no website link when it was checked. For a public example, the Pune digital marketing leads page shows 142 businesses with no Meta pixel and no Google Ads tag out of 191 websites read, with 97 emails found and 19 named decision-makers, updated 6 November 2026.

The limits matter too. Mantis has no CSV export or CRM sync yet. Leads stay in Mantis, so if you need a file for another tool, a scraper or a manual export is the better route for now. If you need ten thousand rows in a file tonight, a scraper will cost less. Scrapers publish rates of around $1.50 to $4 per 1,000 places, but scraping the Maps interface is against Google's terms, and you clean up the duplicates yourself. Vendor-by-vendor scraper pricing is in Google Maps scrapers compared.

Where it breaks for 500: if the 500 must end up in a spreadsheet or CRM today. Without export, that isn't possible yet.

Sources: Google Maps Platform docs and pricing, Apollo and Clay pricing pages, Yelp developer docs, Kellogg library, MapBusinessOnline, mantisai.in, all checked 13 November 2026. Prices change, so check before you buy.

If you're building the list for a web design agency in particular, how to build a lead list for a web design agency covers which signals to filter on once you have it.

I'm Tarun, founder of Mantis. If five hundred businesses of one type in one city is your job this week, describe it in plain words and Mantis quotes the credits before it runs. New accounts get 500 free credits once at signup, no card. Start here.