← All articles

SEO & Websites

Separate Brand Searches From New-Customer Discovery Before Judging SEO

Classify search queries without treating brand clicks as new demand or nonbrand searches as proven new customers, using a worked sample.

By , Co-Founder & CEOUpdated 4 min read
Matching search tokens and varied discovery shapes occupy two separated trays.

An SEO report says search clicks doubled. Most of the extra searches contain your company name after a successful referral campaign. The increase matters, but it does not establish that more people discovered your service through a generic search.

Separate the visible query mix before deciding what to do next. Brand and nonbrand groups answer different questions, and neither group proves whether a visitor was a new customer.

Classify the queries you can actually see

Consider a hypothetical company called Cedar Repair. Its name contains common words, so a broad filter for “cedar” would misclassify some searches. Start with recognizable business names, approved variants, and known product names, then keep ambiguous terms in a separate group.

Sample queryClassificationInterpretation limit
cedar repair companyBrand-likelyCould reflect prior awareness; does not establish why the searcher knows the company.
cedar repair phone numberBrand-likelyCould be an existing customer or a new prospect following a recommendation.
boiler repair bristolNonbrand service queryUseful service intent signal; does not prove the searcher is new or qualified.
cedar repairAmbiguous until reviewedCould refer to the business or to repairing cedar material.
Queries absent from the tableUnknown or unreportedDo not allocate them to the nonbrand group to balance the total.

The labels are an editorial classification. Search Console also offers a branded/nonbranded filter for eligible sites, but Google notes that it may misclassify queries and is unavailable for sites with low impressions. Review the terms whichever method you use. Neither method provides a verified customer identity.

Keep the unknown portion visible

Search Console's performance documentation explains query and page groupings, while its query guidance describes omitted and truncated rows. Some query data does not appear in the table. Chart totals and exported rows can also differ through aggregation and data limits.

Suppose the hypothetical report shows 100 total clicks and 80 clicks in the visible classified query rows: 50 brand-likely, 20 nonbrand, and 10 ambiguous. Keep the other 20 outside those groups. Do not call the nonbrand total 50 by assigning everything you could not classify to discovery.

Use a consistent period, property, search type, and filter when comparing groups. Save the classification terms so a new analyst does not change the meaning of “brand” halfway through the trend.

Choose an action each group can support

For brand-likely searches, check whether people reach the expected page and can find contact or service information. Strong brand demand still deserves a usable website. A company can waste referrals if its search result or page misrepresents the offer.

For nonbrand service queries, inspect which pages appear and whether they answer the buying need. A repair query reaching an unrelated article gives you a specific page decision to investigate. A single impression does not establish enough demand to commission a large content series.

For ambiguous queries, review the destination and surrounding query patterns before deciding. If the company name overlaps a common task, keep that uncertainty visible instead of forcing a confident split.

Connect the query story to inquiry evidence

Ask whether qualified inquiries changed during the same period, using your own defined inquiry stages. Search queries cannot tell you which booking closed. Our SEO lead measurement guide covers those definitions.

In the report, write a conclusion the data supports: “Most visible click growth involved brand-likely searches; service-query evidence remains limited. Improve the service destination and continue measuring qualified inquiries.” That gives the owner a useful action without pretending to know every visitor's history.

Avoid mixing query and page totals

When the owner asks which page supports discovery, examine the page view as a separate analysis with its stated filters. Do not add page impressions to query impressions or treat the two tables as independent customer groups. Save the report setup alongside the classification so another analyst can reproduce the same view. A clear statement of what each total includes lets the business compare periods without inventing missing customers or claiming that every row describes a different person.

For help interpreting search data and improving the service pages behind it, explore our SEO services.

SEO growth

Want this done for your business?

See how the SEO Growth Partnership works — we audit your site, improve listings, and publish content for the services and locations you want customers to find.

Luke Thompson

About the author

Luke Thompson

Co-Founder & CEO, The Ops Guide

Luke Thompson is Co-Founder and CEO of The Ops Guide. He writes about practical uses of AI, workflow automation, custom software, and SEO for growing businesses.

About The Ops Guide

The Ops Guide helps businesses improve how work gets done through custom software, workflow automation, AI systems, and SEO services.

The Ops Guide Brief

Get AI news you can use.

The developments worth your attention, what they change at work, and what to do next. Choose a daily briefing or a weekly selection.

Free to subscribe. Confirm by email. Unsubscribe any time.