A plumbing business can have a busy Google Maps listing and no website link attached to it. For a web design agency, that is a useful place to start research. The next step is to discover whether the business actually lacks a site, has an unlinked one, or simply returned incomplete information.
Those situations call for different work. An unlinked official website might suggest a listing correction. A business whose website cannot be located may deserve further investigation. A listing with an uncertain identity should stay out of an agency's prospect queue until someone resolves it.
This guide turns one local plumber search into a website-verification queue. You can use G Maps Crawl to collect the business records, then review the evidence before deciding which businesses fit your services.
Define a market you can actually review
Start with one query and one area, such as plumbers in North Austin Texas. Write down the service you offer and the question you want the list to answer: “Which local plumbing businesses should we investigate for a website or listing project?”
Define the inclusion rules before inspecting the results. For this example, keep plumbing businesses serving the intended area. Hold supply stores, unrelated contractors, and businesses whose location or identity is unclear for separate review. If the same brand has two distinct branches, keep them as separate locations.
A keyword search is an exploratory sample. It can contain nearby places, adjacent categories, and repeated records. A list of returned businesses cannot establish the number of plumbers operating in North Austin or the percentage of all plumbers without websites.
Keep the first collection small enough that someone can inspect every candidate. A useful initial outcome is five businesses with specific verification tasks, even when a larger search would return more rows.
Collect listing evidence before collecting contacts
In the Online scraper, create a focused search and inspect the available records. Eligible paid Online plans support CSV, Excel, and JSON exports. Online, API, and Extension allowances are separate; choose the product that fits how you plan to work.
Developers can also use the synchronous Search API or MCP server. Its search_google_maps tool returns up to 20 businesses per page. For a bounded first pass, request page 1 and page 2 only if the first page is nonempty. Those two calls can return up to 40 records before deduplication and use up to two API request units, including an empty second page.
Keep extra: false for this task. Website verification does not require additional email or social enrichment. For setup, use the MCP connection guide.
Preserve the business name, category, formatted address, phone, website, business status when available, Maps URL, place identifiers, and original observation timestamp. Include place.fieldStates in the saved JSON so an empty value can be interpreted alongside its collection state.
The search tool does not apply a “no website” filter. Identify records with an empty website field in your saved results, then investigate them. If an agent creates a local spreadsheet, its own file tools do that work; MCP supplies the business records.
Give each missing website a review status
Do not label every empty website cell “needs a website.” Use a separate research column that describes what you know.
| Research status | Evidence required | Next action |
|---|---|---|
| No website returned—unverified | No website URL returned for the listing. | Check the current listing and look for an official site. |
| Official site found | A website matches the business and location. | Record the source and assess whether a listing correction is relevant. |
| Website not located | The defined manual checks did not locate a matching official site. | Keep the checks and their date; decide whether further investigation is worthwhile. |
| Identity on hold | Address, operating status, or business identity is unresolved. | Resolve the identity before qualifying the opportunity. |
These are worksheet labels you create, not G Maps Crawl output fields. Reserve “no website linked” for a current Maps listing you have checked. “Website not located” describes the limits of your research. It does not prove that no website exists.
Verify the business behind the row
Open the returned Maps URL and compare the current listing with the saved record. Note whether a website is now linked and whether the business name, location, and category still match your intended market.
Next, search the business name together with its city or phone number. A website with the same trading name may belong to a different company. Match the site's business identity using location, phone, service area, or other corroborating information. Record uncertainty when the evidence disagrees.
Check whether the only discovered URL is a social profile, directory entry, booking service, or official business site. Those are different forms of online presence. A business using a marketplace intentionally is not automatically missing the functionality your agency sells.
For a confirmed site, inspect the task relevant to your offer: finding emergency contact details, identifying the service area, or requesting an estimate. Document what you actually observed. A missing Maps link alone does not establish a broken contact form, weak search visibility, or lost revenue.
Finally, record who checked the evidence and when. G Maps Crawl can reuse complete Maps details for up to 730 days, while query responses normally have a 24-hour cache window. A newly received response can contain an older website observation. Your verification date belongs in a separate column from place.observedAt.
Work through a small example
The following businesses and findings are fictional. They demonstrate the worksheet, not live search results or customer outcomes.
| Illustrative business | Initial record | Manual finding | Research decision |
|---|---|---|---|
| Example Plumbing A | Website empty; phone available. | A matching official site was located. | Record the site and review the listing link; do not classify as a website-build lead. |
| Example Plumbing B | Website empty; category and address match. | No official site located through the documented checks. | Keep as a candidate for further website research. |
| Example Plumbing C | Website empty; address incomplete. | Two similarly named businesses were found. | Hold until identity is resolved. |
The useful output is the change in what the agency knows. All three rows began with an empty website field, but only one remains a candidate for additional website research.
Turn the findings into a reusable handoff
Keep collection and review columns separate. A practical worksheet includes the source identifier, Maps URL, returned website, source observation date, review status, verified website if found, verification date, evidence URL, and next check.
If you use an AI editor to organize saved records, give it a transformation task rather than permission to expand the search:
Read the saved plumber records. Create a website-verification worksheet.
Keep separate branch locations and flag uncertain duplicates for review.
Preserve source identifiers, URLs, timestamps, and missing-field states.
Mark empty websites as "no website returned—unverified".
Leave manual findings and verification dates empty.
Do not infer website absence, contact businesses, or make additional searches.
After manual review, report counts with their denominators: how many distinct records were in scope, how many had no website returned, and how many of those were checked. Keep unresolved records visible instead of counting them as confirmed opportunities.
Choose the next few businesses based on evidence, service fit, and the work still needed. That gives the agency a queue someone can act on and explain. Run a focused Online search, then use the business-field guide to interpret the records before qualifying them.
