To find B2B leads with Google search operators, combine three parts in every query: a niche term ("commercial kitchen equipment"), a phrase that only an operating business would publish ("trade account" or "our showroom"), and exclusions that remove directories and marketplaces (-site:yell.com). Restrict by country with site:.co.uk or a place name, save the queries that work, and rerun them weekly, keeping only domains you have not seen before.
Generic operator guides are written for SEO link building. This one is for prospecting: the recipes are built to surface company websites, not blog posts, and the method ends with a clean, repeatable list.
Which operators actually work
Google's own help page documents a short list. Everything else is folklore that may or may not behave the same way next month.
| Operator | Documented by | What it does | Prospecting use |
|---|---|---|---|
"exact phrase" | Google, Bing | Matches the words in that order | Buying signals: "trade enquiries", "become a stockist" |
-word or -site:domain | Google (minus), Bing (NOT or -) | Excludes a word or a site | Remove directories, jobs boards, marketplaces |
site: | Google, Bing | Restricts to a domain, domain ending or (on Bing) a directory | site:.co.uk, site:.ie |
filetype: | Google, Bing | Restricts to a file type | filetype:pdf "trade price list" |
before: / after: | Restricts by date the page was updated | Recently updated sites only | |
OR | Bing (documented), Google (works, not on the help page) | Either term | (wholesaler OR distributor) |
intitle: | Bing (documented), Google (works, not on the help page) | Term in the page title | intitle:"contract cleaning" |
inurl: | Google (works, not on the help page) | Term in the URL | inurl:trade-account |
inbody: | Bing | Term in the page body | inbody:"ISO 9001" |
language: and loc: | Bing | Language and country of the page | language:en loc:GB |
Two formatting rules from both engines: no space after the colon (site:example.co.uk, not site: example.co.uk), and one term per intitle: on Bing (chain several if you need them).
Treat undocumented operators as approximate. intitle: and inurl: usually work on Google but can return fewer or looser results than you expect. Check a sample by hand before you trust a query.
Operator recipes by prospect type
The patterns below use fictional niches; swap in your own terms. Each one pairs a niche with a phrase that operating businesses publish and directories rarely do.
Manufacturers
"manufacturer" "made in the UK" "stainless steel" fabrication -site:amazon.co.uk -site:ebay.co.uk
intitle:"manufacturers" "precision engineering" "ISO 9001" site:.co.uk
"our factory" OR "our manufacturing facility" "packaging" site:.co.uk
Wholesalers and distributors
"trade account" "wholesale" "catering equipment" site:.co.uk
inurl:trade "apply for a trade account" "garden furniture"
"UK distributor for" "power tools" -jobs -careers
The wholesalers and distributors guide goes deeper, including how to tell a distributor's site from a retailer's.
Service businesses
"commercial cleaning" "Leeds" "request a quote" -site:checkatrade.com -site:yell.com
intitle:"accountants" "Bristol" "about us" -site:linkedin.com
"structural engineer" "Manchester" "our projects"
Brands looking for stockists (and the stockists themselves)
"become a stockist" "skincare" site:.co.uk
"find a stockist" "craft gin"
A brand's "find a stockist" page is also a ready-made list of retailers in that category.
Buying signals and growth
"we are hiring" "warehouse operative" "distribution" site:.co.uk
"new premises" OR "we have moved" "engineering" after:2026-06-01
"now offering" "electric vehicle charging" "electrician"
Price lists and catalogues
filetype:pdf "trade price list" "plumbing"
filetype:pdf "product catalogue" "industrial fasteners" site:.co.uk
Technology signals
"powered by Shopify" "UK" "homeware"
"proudly powered by WordPress" "joinery"
These footers are optional and often removed, so search finds only some sites. Detecting the platform from the site itself is more reliable; see finding businesses by website technology.
Excluding directories and marketplaces
Directories, marketplaces, review sites, job boards and social networks rank well for almost every niche query. Left in, they fill half your list with pages that are not prospects.
In the query, exclude only the worst offenders. Five to eight -site: terms is plenty. Very long queries are truncated or behave unpredictably, so do not paste a list of 50 domains into Google.
A starter set for UK trades and services:
-site:yell.com -site:checkatrade.com -site:trustpilot.com -site:facebook.com -site:linkedin.com -site:indeed.co.uk
For products:
-site:amazon.co.uk -site:ebay.co.uk -site:etsy.com -site:alibaba.com
After collection, suppress the rest. Keep a suppression list of domains (directories, marketplaces, social sites, news sites, government and education) and remove them from your results before you do anything else. A domain list is faster and more complete than any query.
Add positive signals instead of more negatives. A phrase like "our workshop", "trade counter" or "company registration number" appears on business sites and rarely on directory pages. One good positive phrase removes more noise than ten exclusions.
Language and country
Search engines personalise by location, so a query typed in Manchester and one typed in Dublin return different results.
On Google:
- Set the region and the results language in Search settings.
- Restrict the domain ending with
site:.co.uk,site:.ukorsite:.ie. - Many UK businesses use
.com. Run a second version of each query with a UK place name in quotes instead of the domain filter, for example"Sheffield"or"West Midlands". - Use local vocabulary: "joiner" and "chippie" in the UK, "carpenter" elsewhere; "solicitor" rather than "attorney"; "postcode" rather than "zip".
On Bing:
loc:GBrestricts to pages from a country or region, and you can combine codes withOR:(loc:GB OR loc:IE).language:enrestricts to English pages.
Other languages: If you prospect abroad, write the query in the local language. A German distributor is far more likely to publish "Händler" and "Großhandel" than "distributor". Remember that the law on contacting them differs too: Germany generally requires prior consent for advertising email, including B2B.
Making searches repeatable
A one-off search gives you a list. A repeatable search gives you a pipeline.
- Keep a query file. One query per line, with a note of the niche, country and the date you last checked it.
- Record what each query returned. For each result: query, engine, position, URL, title and snippet. The snippet often explains why a result was a false positive.
- Deduplicate by domain. Ten pages from one site are one prospect.
- Apply the suppression list.
- Rerun weekly or monthly. Compare domains against everything you have seen before and keep only new ones.
- Retire queries that stop producing. If a query returns only known domains for three runs, replace it.
Search positions change daily and vary by location and device, so a position is a note about one search at one time, not a ranking metric.
To run the same searches every week without rebuilding them, save them as a workflow in the workflow builder. To get results into a spreadsheet, the Google search results scraper exports them to CSV.
Worked example: commercial laundry suppliers
A fictional firm, Brightwash Services, sells maintenance contracts for commercial laundry machines and wants UK distributors and dealers to partner with. Its first pass:
| Step | Query or action | Why |
|---|---|---|
| Niche plus signal | "commercial laundry" "dealer" OR "distributor" site:.co.uk | Finds companies that describe themselves as dealers |
| Second signal | "commercial laundry equipment" "trade enquiries" | Catches .com sites and different wording |
| Title filter | intitle:"laundry equipment" "service engineers" | Dealers that also service machines |
| Exclusions | -site:amazon.co.uk -site:ebay.co.uk -site:yell.com | Removes the three biggest sources of noise seen in the first run |
| Suppression | Remove marketplaces, hotel booking sites, news and manufacturer head offices outside the UK | Done after collection, by domain |
| Review | Open 20 accepted sites and check each has an address, a service area and an enquiries channel | Confirms the queries find the right type of business |
Of a fictional 140 results across three queries, 60 domains remain after deduplication, and suppression removes 22 more. The remaining 38 are reviewed, the queries that produced mostly noise are rewritten, and the file is saved for a monthly rerun. Next month, only domains not seen before go to review.
The figures are illustrative and your proportions will depend on the niche. Plan for deduplication and suppression to remove a large share of raw results, because several pages per site and many directory pages rank for the same terms.
From search result to qualified lead
A search result is a candidate, not a lead. Before outreach, check each site:
- Is it an operating business? Not a directory, blog, news site, parked domain or holding page.
- Is it the right type? Manufacturer, distributor, retailer or service provider, depending on your offer.
- Is it in the right place? Address on the contact page or in structured data, not just a town name in a blog post.
- Is there a suitable channel? A business email for enquiries or sales, or a contact form.
- Who is it legally? For UK companies, a company number on the site lets you check status on Companies House. See the Companies House lead generation guide.
Where you use AI to fill gaps such as a category or service description, insist on a source URL and passage for every value; see AI lead enrichment with sources.
Automation and the search engines' rules
Typing queries by hand is ordinary use. Automating it is different. Google's terms prohibit "using automated means to access content from any of our services" in violation of machine-readable instructions such as robots.txt files, and Google says its Custom Search JSON API is closed to new customers, with existing customers given until 1 January 2027 to move to an alternative. Search engines also detect and challenge automated traffic.
If you automate, keep it modest: low volumes, delays between requests, no attempts to evade challenges or blocks, and a clear understanding of the terms that apply to you. Third-party platforms may restrict data collection or automated actions. Before running a workflow, check the platform's rules and make sure you have permission for the actions you configure.
How Map Extract helps
Map Extract's SERP scraper is a Windows desktop module for running the method above as saved campaigns across Google, Bing and Brave. The Google search scraper page describes the same module focused on Google; it uses the same engine and filters.
- Campaign filters that match the method. Keywords (one per line), required terms that must appear in the result title or description, language, location, allowed domain endings such as
co.ukanduk, website types (business, manufacturer, distributor, wholesaler, retailer, service provider, and types to exclude such as directory, marketplace, blog, news, government and social) and your own suppression list. - Organic results only. Ads and sponsored results are skipped, redirect wrappers are unwrapped, and popular platforms and government domains are excluded by default.
- Context kept with each result. URL, title, snippet, engine and position on that results page at that time, plus related search suggestions, so you can see why a result was accepted or rejected.
- Optional AI help. With a configured Gemini key it can suggest keywords and suggest domains to suppress based on your stated intent. You review both before use.
- Straight into website research. Accepted sites go to the website data extractor, which reads public pages (respecting robots.txt), and records business contacts, company numbers, structured data and site type.
- Repeatable runs. The workflow builder saves the source order and settings so you can rerun the recipe. The AI business finder can turn a plain-English brief into niche and location suggestions for you to review.
What it does not do: it does not guarantee result counts or rankings, it does not get around search engine challenges (pages that need a challenge are recorded, not counted as results), and it does not decide whether a site is a good prospect for you. Delays and per-keyword limits are settings you control; start small. Search sources are off by default in the workflow and you enable them when you need them.
Public business and social results the app collects are uploaded automatically to Map Extract's servers while it is licensed and running; the privacy notice explains this.
Key takeaways
- Build each query from a niche term, a phrase only operating businesses publish, and a few exclusions.
- Rely on documented operators (quotes, minus,
site:,filetype:,before:/after:); treatintitle:andinurl:on Google as approximate. - Exclude a handful of directories in the query and suppress the rest afterwards by domain.
- Set country and language, and run a
.comvariant with place names for UK targets. - Save queries, deduplicate by domain and rerun on a schedule, keeping only new domains.
- Automated searching is subject to the search engines' terms; keep it modest and never evade blocks.
Frequently asked questions
Which Google search operators are best for finding leads?
Quotes for exact phrases, the minus sign to exclude words or sites, and site: to restrict to a domain or ending such as .co.uk are documented by Google and reliable. intitle: and inurl: are widely used and usually work but are not on Google's help page, so treat their results as approximate. Combine a niche term, a buying-signal phrase and a few exclusions.
How do I stop directories showing up in Google results?
Exclude the biggest offenders with -site: (for example -site:yell.com -site:facebook.com) and add terms that only operating businesses use, such as 'trade account' or 'our workshop'. Keep the exclusion list short in the query, because very long queries get truncated, and remove the remaining directories, marketplaces and social sites from your results afterwards with a suppression list.
Is it allowed to scrape Google search results?
Google's terms prohibit using automated means to access its services in violation of machine-readable instructions such as robots.txt, and the official Custom Search JSON API is closed to new customers. Manual searching is fine. For anything automated, read the terms that apply to you, keep volumes low and paced, and accept that the search engine may block or challenge automated traffic.
How do I search Google for businesses in one country only?
Set your search region and language in Google's search settings, then add site:.co.uk (or the relevant country ending) to restrict the domain. Many UK businesses use .com, so also try queries with a UK town, county or 'UK' in quotes instead of the domain filter. On Bing, the documented loc:GB and language:en keywords do the same job.
Sources
Written by the Map Extract editorial team. Found an error? Email [email protected] and we will check and correct it. How we research guides: editorial policy.