How to Build a B2B Lead List by City and Niche (Step by Step)

To build a B2B lead list by city and niche, turn your niche into the words businesses use to describe themselves, pair every keyword with every city you want to sell in, and search a source where those businesses are listed. Then find an email on each company's own website, verify every address, and export only the valid ones, tagged with city and niche so every email you send can be specific.

That's the whole method. The rest of this guide covers each step, and the decisions that separate a list that gets replies from one that gets your domain flagged:

1. Define your niche as searchable keywords
2. Choose your cities
3. Build the city × keyword matrix
4. Pick the source that fits your buyer
5. Clean the company list
6. Find emails on each company's own domain
7. Verify every address
8. Export and segment for outreach
9. Keep the list fresh

To keep it concrete, we'll follow one example from start to finish.

Example (fictional): a two-person web design studio in Denver builds websites for home-service contractors. It wants a list of contractors in the Denver area to pitch.

Step 1: Define your niche as searchable keywords

A niche in your head ("home services", "healthcare", "local businesses") isn't something you can search. A search needs the words businesses use to describe themselves on their listings and websites: roofer, roofing contractor, gutter installer.

Start with how your best customers describe their own business, then widen the list:

• Listing categories. On Google Maps, each business's category appears under its name, and businesses choose it themselves. Look up a few of your existing customers and note their categories: those are proven search terms.
• Everyday words and synonyms. The customer searches "plumber"; the listing says "Plumbing contractor". Search both.
• Sub-niches. "Dentist" also covers orthodontists, cosmetic dentists and pediatric dentists, who buy differently and respond to different messages. Official industry classifications such as NAICS help you see which sub-industries exist.
• The local language. Listings are written in the language of the country. In Germany you search Dachdecker, not roofer; in Spain, fontanero, not plumber.

Then test before you scale. Run one keyword in one city and read the first 20 results. If a good share of them aren't businesses you would sell to (suppliers, schools, franchise head offices, a neighboring trade), the keyword is too broad or too ambiguous. "Dental" is a classic: it also returns dental labs and supply companies.

Example: the studio starts with roofer, roofing contractor and gutter installer. It drops home improvement, because its test search returned hardware stores.

Step 2: Choose your cities

Pick cities where you can deliver and where your proof lands. A case study from Aurora means more to a roofer in Lakewood than to one in Leeds. If you sell by phone or video call, time zones matter too.

• Split big cities. A single search returns a limited number of businesses, so one search for "plumber" in a city of millions only scratches the surface. Search suburbs, districts and neighboring towns separately.
• Don't skip small towns. Their businesses may hear fewer sales pitches than those in the big city, and a local reference carries further in a small market.
• Size the market with official data. In the US, the Census Bureau publishes population totals for cities and towns, and most national statistics offices publish something similar. Sort by population and work down the list.
• Expect overlap. Contractors, cleaners and other service-area businesses work in several towns, so they show up in more than one search. You'll deduplicate in step 5.

Example: the studio picks Denver plus six suburbs: Aurora, Lakewood, Littleton, Arvada, Westminster and Centennial.

Step 3: Build the city × keyword matrix

Each city paired with each keyword is one search. Three keywords across seven cities is 21 searches. Writing the matrix down before you run anything tells you three things: how many searches you need, what the list will cost in time or money, and the order to run them in.

Run your most promising pair first and read the results before running the rest. If the first search is full of the wrong businesses, you've lost one search, not 21.

Overlapping keywords are fine, even useful: roofer and roofing contractor return many of the same businesses, but not all of them. What matters is that you deduplicate afterwards, by website domain, or by name and address for businesses without a website.

If a tool charges per search, check what one search includes, because it changes how you build the matrix. In Glintscout, for example, each city and keyword pair is one search of up to 100 companies (you choose how many), and it costs the same whether it finds 8 companies or 100 (see pricing). A narrow keyword in a small town then costs as much as a broad one in a big city, so in small towns, use the broadest keyword that still describes your buyer.

Step 4: Pick the source that fits your buyer

Where you search decides who you find. Match the source to what you know about your buyer:

Map and directory listings give you breadth. Advertisers and stores give you a reason to write: a business paying for clicks has a marketing budget, and a store running a particular helpdesk has a tool you can offer to replace or improve.

One more choice: a bought database or a live search. A database is instant, but it was compiled at some point in the past; a live search reads the source when you run it. The trade-offs are in lead databases vs live search.

Example: the studio's buyers are local contractors, so it searches Google Maps by city and keyword.

Step 5: Clean the company list

A raw search result isn't a lead list yet. Before you spend time or money on finding emails, remove the rows that can only waste it.

• Directory and social "websites". Many listings give a Facebook page, a booking page or a directory profile as their website. Searching facebook.com for an email finds Facebook's addresses, not the business's. Set those rows aside for phone or social outreach.
• Duplicates. Deduplicate by domain; for businesses without a website, by name plus address.
• Chains and franchises. A franchise location usually links to the brand's site, often a /locations/... page on the head office's domain, so any email you find there reaches head office. Decide whether you sell to the local owner or to the brand, and filter accordingly.
• Closed businesses. Map listings can be marked as permanently closed. Drop them.
• Wrong fits. Suppliers, schools and other businesses that matched the keyword but will never buy.

Businesses with no website at all deserve a list of their own. You can't find an email on a domain that doesn't exist, but if you sell websites or local marketing, they're exactly who you want: reach them by phone or in person.

Step 6: Find emails on each company's own domain

The most reliable email for a business is the one published on its own website. Check the contact page, the footer, the about or team page, and the legal pages (imprint, privacy policy, terms). In the EU and the UK, e-commerce rules require online service providers to publish an email address, so European business websites often have one on a legal page.

Which address to pick depends on the size of the company:

• Small businesses, the bulk of any local list: the published general address (info@, office@, hello@) is usually read by the owner or whoever runs the office. That's a fine first contact.
• Larger companies: a named person who owns the problem you solve beats a shared inbox. In Europe, a named work address is personal data under the GDPR, which raises the bar for using it.
• Skip support@, careers@, privacy@, noreply@ and technical addresses such as postmaster@ and abuse@. A pitch sent there lands in a ticket queue or a filter, not in front of a buyer.

One address per company is enough for a first email. Writing to four people at a five-person business reads as a mass campaign, not a conversation.

The full set of methods, including email patterns and what to do when a company only has a contact form, is in how to find a company's email address.

Step 7: Verify every address

Addresses go stale. People leave, businesses change domains or email providers, and contact pages keep old addresses for years. Every bounce tells mailbox providers that you send to a list you don't maintain, and a poor bounce record hurts the reputation of your whole domain, not just your outreach.

Verification checks an address without sending an email: the format, whether the domain accepts mail, and whether the mail server accepts that particular mailbox. Each address ends up with one of four verdicts:

A common rule of thumb is to keep hard bounces under 2% of what you send, and verifying before every campaign is how you stay there (more in what a good email bounce rate is). How the check works is explained in how to verify an email address without sending an email, and catch-all emails explained covers whether to send to addresses the server can't confirm.

Example: the studio verifies its list and exports only the valid addresses. The catch-all and unknown ones wait for a re-check.

Step 8: Export and segment for outreach

Export only valid addresses, and keep the context that makes each email specific. A useful lead list has at least these columns:

Then segment. A segment is a group that can get the same email without it reading like a mass email: roofers in Aurora with no website, or dentists in Leeds running Google Ads. The city lets you reference their market, the niche lets you talk about their work in their words, and the signal gives you a reason to write now. Cold email templates for local businesses show how to turn a signal into a first line.

Send each segment in small batches, watch which ones reply, and grow the matrix where the replies come from: more cities for a niche that works, more niches in a city that works.

Before the first send, set up your sending domain (SPF, DKIM and DMARC) and learn the thresholds. Google's email sender guidelines tell senders to keep the spam rate reported in Postmaster Tools below 0.1% and never let it reach 0.3%, and Yahoo's sender best practices set the same 0.3% ceiling. The practical setup is in cold email deliverability rules.

Step 9: Keep the list fresh

A lead list starts aging the day you build it. Businesses open, close, rebrand and change domains; staff move on; mailboxes get shut down. Three habits keep a list usable:

1. Re-run your matrix on a schedule. New businesses are new leads, and one that just opened needs almost everything: a website, a booking system, an accountant. Every few months, run your searches again and add what's new.
2. Re-verify before each send. If an address was verified more than a few weeks ago, check it again. A verification costs less than a bounce.
3. Keep a suppression list. Everyone who opts out, bounces hard or complains goes on a do-not-contact list, and every new list is checked against it before you send. In many places this is also the law: the US CAN-SPAM Act, for example, requires you to honor opt-out requests within 10 business days (FTC compliance guide).

Playbooks by what you sell

The nine steps work for any offer. These guides apply them to three common sellers, with the niches, signals and emails that fit each:

Common mistakes

• A niche that's really a category of categories. "Local businesses" isn't a niche; "roofers in the Denver suburbs" is.
• One search per big city. You'll see a fraction of the market. Split it into districts and suburbs.
• Emails from the wrong domain. An address found on a directory or social page may belong to the platform. Use addresses on the business's own domain.
• Guessing instead of verifying. A guessed firstname@ address looks plausible and bounces anyway.
• One email for the whole list. If a message would work for any business in any city, it won't work well for any of them.
• Treating the list as permanent. Re-run, re-verify and suppress.

FAQ

How many leads should my first list have?

Enough to test one message on one segment. A common rule of thumb is 100 to 200 verified contacts per message you test: enough to tell a message that works from one that doesn't before you scale it. Build a few segments of that size rather than one huge list.

Can I build a B2B lead list for free?

Yes, by hand: search a map or directory, copy each business into a spreadsheet, and visit each website for an email. It costs time instead of money, a few minutes per business once you include the website visit. The part that's hard to do by hand is verification, which means checking each address against its mail server.

Should I buy a lead list instead?

A bought list is faster, but it's a snapshot from whenever it was compiled, and you rarely know when each address was last checked. If you buy one, verify it before you send, and check that its categories and cities match what you would have searched yourself.

Can I target people by job title this way?

Not directly. Searching by city and niche finds businesses and their business emails, not people by title. For small businesses that's rarely a problem, because the owner usually reads the main inbox. For larger companies, add a step to find the right person.

Is it legal to email the businesses on my list?

It depends on where they are. The US CAN-SPAM Act allows commercial email to businesses if you follow its rules, such as honest headers and subject lines, identifying the message as an ad, your postal address and a working opt-out. The UK's PECR lets you email companies without prior consent, but not sole traders or some partnerships (ICO guidance). Germany and Spain are stricter: email advertising generally needs prior consent, even between businesses. The details are in is cold email legal? This is general information, not legal advice.