Google X-Ray Search Builder

Google X-ray search uses site: and other search operators to find public pages that a platform’s own search may not surface. This builder turns a target site, a list of role titles and an optional location into one copy-ready Google query, then scores how tightly that query is put together. Everything runs in your browser; nothing you type is uploaded.

What is a Google X-ray search?

A Google X-ray search is an ordinary Google query that uses the site: operator to look inside one website, plus extra operators to narrow the page type or the words that must appear. The name comes from the idea of seeing through a platform’s own interface: instead of relying on LinkedIn search, you ask Google which pages on LinkedIn it has indexed and which of those match your terms.

For example, site:linkedin.com/in intitle:"Head of Marketing" "Berlin" asks Google for indexed LinkedIn profile pages whose title contains the phrase and whose page mentions Berlin. It returns public pages only, and only those Google has chosen to index. It does not log in, scrape, or reveal private profiles.

How to use this builder

1. Choose a search target. The tool adds the matching site: filter, including a path such as linkedin.com/in for profiles.
2. Enter your roles, titles or keywords, one per line or separated by commas. Each becomes one alternative in an OR group.
3. Optionally add a location. It is quoted and added as an extra required phrase.
4. Choose where the terms should match: anywhere on the page, the page title with intitle:, or both (strictest).
5. Read the query strength ring and the four component bars. They describe how tightly the query is built, not how many results it will get.
6. Press Copy query and paste the result into Google. Edit the inputs and copy again whenever you refine it.

The operators this builder combines

• site: restricts results to URLs under a domain or path. Built example: site:linkedin.com/in.
• intitle: asks Google to match the phrase in the page title. Google’s matching is not guaranteed to be exact. Built example: intitle:"Head of Marketing".
• Quotes group words into a phrase instead of matching them separately: "United States".
• OR matches either alternative. The builder uses uppercase OR and wraps the group in parentheses: (term one OR term two).
• - excludes pages that contain a term: -jobs.
• Parentheses keep alternatives together so the other filters apply to the whole group: (A OR B).

Spaces between the parts act as an implicit AND: the site filter, the role group, the location and every exclusion all apply to the same result. The term box is deliberately literal. If you type site: or intitle: inside it, that text is quoted as a search phrase instead of being treated as an operator. Use the controls for those two filters so the built query stays predictable.

How the query strength score works

The score is a structural checklist, not a forecast. It rates four parts of the query from 0 to 100 and combines them with fixed weights.

The final number is rounded and mapped to four labels: Strong at 88 and above, Solid from 72 to 87, Thin from 52 to 71, and Weak below 52. The thresholds are a writing convention inside this tool. They are not benchmarks and they do not predict search volume, profile quality or reply rates.

Two things the score cannot judge are the quality of your synonyms and whether the people you find are worth contacting. A query can be structurally perfect and still target the wrong audience.

Worked examples

These examples are produced by the builder itself. They show how the same controls change the query and the structural score.

1. LinkedIn profiles, three marketing titles, United States, page title, one exclusion.

   site:linkedin.com/in (intitle:"Head of Marketing" OR intitle:"VP of Marketing" OR intitle:"Marketing Director") "United States" -jobs

   Score: 96 · Strong. A path-level site filter, three focused alternatives, title matching and a location. The few missing points come from the three-term OR group and the precision score.

2. LinkedIn company pages, two software terms, London, page text.

   site:linkedin.com/company (SaaS OR "B2B software") "London"

   Score: 89 · Strong. The site path and location are strong; page-text matching is the main deduction because the terms can appear anywhere on a company page.

3. GitHub, three technical terms, no location, page title.

   site:github.com (intitle:"machine learning" OR intitle:LLM OR intitle:"developer tools")

   Score: 88 · Strong. A bare-domain site filter cannot separate profiles from repositories, and there is no location filter.

4. No site filter, one term, no location, page text.

   marketing

   Score: 54 · Thin. No target, no location and no title constraint. This is the kind of broad query the builder is meant to improve.

Google X-ray tips that change results

Use a path, not just the domain

site:linkedin.com/in is usually more useful than site:linkedin.com, because it asks for profile URLs instead of job posts, company pages and help articles. The same idea applies to linkedin.com/company and crunchbase.com/organization.

Keep the OR group short

Two or three closely related titles usually produce a cleaner list than eight loose keywords. If you need a wider net, run separate searches for separate title families instead of one long OR group. It is easier to compare the results that way.

Match in the title when the headline matters

On LinkedIn, the page title includes the person’s name and headline, so intitle: is a useful way to keep the role in the part of the page Google summarises. On GitHub, the title is mostly the profile or repository name, so title matching is less reliable for skills.

Use exclusions for recurring noise

A minus term removes pages that contain it. Exclusions work best when they target a repeated pattern you have actually seen in the results, not a guess. Start with one or two and check whether the list improves.

What X-ray search cannot do

• It cannot see pages Google has not indexed, pages behind a login, or profiles that are private.
• It cannot promise complete coverage. Indexing changes, and Google may show different results to different people.
• It does not return email addresses, phone numbers or verified job titles. A profile snippet is not a verified fact.
• It cannot guarantee relevance. A page can contain a term without belonging to the person or role you want.
• Long queries can be truncated, and repeated automated searching can trigger CAPTCHAs or rate limits.
• The builder reads a leading hyphen on a line as an exclusion. A term that genuinely starts with a hyphen has to be rephrased or left out.
• The target list is fixed. For an unlisted domain, choose Any website and add site:yourdomain.com to the copied query by hand.
• Automated collection at scale can break a platform’s terms or privacy rules. This tool builds a manual query only; it does not scrape.

Privacy and data handling

All parsing and scoring happens in your browser. The page makes no network request, sets no cookie and writes nothing to local storage. The only outbound step is the one you take yourself: pasting the copied query into Google. The query builder cannot see what Google returns.

Frequently asked questions

Is Google X-ray search allowed?

Searching public pages through Google is normal search behaviour. What you do with the results is the part that matters: respect the platform’s terms, robots rules and applicable privacy law, and do not turn a manual search into bulk scraping or unsolicited mass contact.

Why do I get fewer results than expected?

Each operator is another constraint. A path-level site: filter, a long OR group, a quoted location and intitle: all reduce the set. Remove the least important constraint first, usually the location or the title match.

Should I use intitle: or search the whole page?

Use intitle: when the role is likely to be in the page title, as with LinkedIn headlines. Use page-text matching when the term might appear in a bio, description or repository README. Test both; the result count is the honest signal.

How many titles should go in the OR group?

Two or three is a practical starting point. More alternatives widen the query and make it harder to tell which title produced a result. If you need many titles, split them into separate searches.

Does this work with LinkedIn Sales Navigator?

No. This builder produces a Google query, not a Sales Navigator filter set. Sales Navigator has its own fields, limits and account requirements.

Can I use the query on Bing or another search engine?

Not reliably. Bing supports some similar operators, but the syntax and indexing differ. This tool builds Google syntax on purpose and does not claim portability.

Why does the tool mention a 32-word budget?

Long search strings are a known practical risk: they can be cut short or produce unstable results. The builder uses 32 words as a soft guardrail and shows the meter so you can see when a query is getting long. It is not presented as a documented hard limit.

Is my query sent anywhere?

No. The tool is HTML, CSS and JavaScript running on the page. It copies the query to your clipboard only when you press the button.

Related tools

• Boolean Search String Generator — build AND / OR / NOT strings for other platforms.
• LinkedIn Search Query Builder — a Sales Navigator-style alternative.
• Local Business Prospector — turn a niche and a city into a search plan.
• Lead List Quality Scorer — check a list after you have built it.

Next step

Finding the right profiles is one step; turning them into a repeatable outreach system is another. See how mizanursguide.com builds B2B lead-generation campaigns that start with a defined audience and end with a measured follow-up process.