Branded vs Non-Branded Keywords: Why the Split Matters for SEO Strategy
- Filed November 10, 2026
- Keyword research, guide 1 of 6
Branded searches come from people who already know the brand: they type its name, a misspelling of it, or the name of one of its products. Non-branded searches come from people looking for something the brand offers without naming it. The two groups behave differently and answer different business questions. Google’s own description of the difference: branded queries “typically lead to higher-ranking pages from your site and result in higher click-through rates, whereas non-branded queries offer organic growth” (Google’s announcement of the branded queries filter). Mixing them in one traffic number makes a site look healthier, or weaker, than it is.
Why the split matters
- Brand demand reflects awareness, which is driven by advertising, PR, word of mouth and product success as much as by SEO.
- Non-brand demand reflects discoverability: whether people who do not know the brand find it when they search for a problem, product category or question.
- The two move independently. A TV campaign can lift branded clicks while non-brand visibility falls. A report that combines them would show growth and hide the loss.
- They set different expectations. Google describes branded queries as typically leading to higher-ranking pages and higher click-through rates, so brand CTR is not a benchmark for non-brand queries.
The Search Console branded queries filter
Google introduced a branded queries filter in the Search Console Performance report on November 20, 2025, and announced on March 11, 2026 that it is available to all eligible sites. According to Google:
- it splits query data into Branded and Non-branded views, across web, image, video and news search types;
- it uses “an internal, AI-assisted system”, not a regular expression, and includes the brand name in all languages, typos and queries that refer to a unique product or service of the site;
- “some queries may occasionally be misidentified”;
- it is available only for top-level properties (not URL path or subdomain properties) and for sites with enough queries and impressions;
- it “has no effect on how Google Search ranking works”.
Google also added an Insights card showing the split of total clicks between branded and non-branded traffic.
Building your own classification
Where the filter is unavailable (URL path or subdomain properties, sites without enough queries and impressions) or where you need a classification you control, build one with the query filter’s regular expression option.
A branded query classification example
Imagine a company called Brightwater Coffee that sells the “Ember” grinder and the “Tidal” espresso machine.
| Query | Branded? | Reason |
|---|---|---|
| brightwater coffee | Yes | Brand name |
| brightwatter coffee | Yes | Misspelling |
| brightwater ember review | Yes | Brand and product |
| ember grinder | Yes, probably | Unique product name; check that "ember" is not a generic term in the category |
| tidal espresso machine price | Yes | Product name |
| best burr grinder | No | Category query |
| how to dial in espresso | No | Informational |
| brightwater vs breville | Yes | Brand comparison; worth tracking separately as a comparison subgroup |
| coffee near me | No | Generic local |
An illustrative regular expression for Search Console’s “Custom (regex)” query filter:
brightw[a]?t+er|bright water|bemberb.*(grinder|review)|btidalb.*(espresso|machine)Test it on an export of queries and in the filter itself before relying on it, since a pattern written for another regex engine may behave differently there, and record it in the report’s definitions.
Edge cases to decide in advance
- Generic words as product names. If a product is called “Ember”, queries for “ember” alone may be unrelated. Include only combinations with category terms.
- Comparison queries (“brand vs competitor”) show brand awareness but also active shopping; consider a third group.
- Competitor brands are non-branded for you but are useful to track as a separate group.
- Employee and support queries (“brightwater login”, “brightwater jobs”) are branded but not marketing outcomes; consider excluding them from performance analysis.
Interpreting the two groups
| Signal | Branded | Non-branded |
|---|---|---|
| Clicks rising | Awareness growing (which can come from non-SEO activity) | SEO discoverability improving |
| Clicks falling | Awareness or demand falling; or competitors bidding on the brand; or results features taking clicks | Lost visibility, competition, demand shift |
| CTR | Typically higher, in Google's description | Lower; compare against your own history |
| Position | Typically higher-ranking pages, in Google's description | The actual competitive picture |
Note that query tables omit anonymized queries, so branded plus non-branded rows will not add up to the chart total when you build your own regex classification.
Reporting
- Report non-branded clicks and impressions as the primary SEO visibility measure.
- Report branded clicks as an awareness indicator, with context from marketing activity.
- Keep the classification method stable month to month, and note the date when switching from a custom regex to Google’s filter, since the two classify differently.
Brand classification checklist
- [ ] Check whether the Search Console branded filter is available for your top-level property.
- [ ] If not, build and document a regex classification; test it on exported queries.
- [ ] Decide how to treat product names, comparisons, competitors and support queries.
- [ ] Report non-brand as the SEO visibility measure; report brand separately.
- [ ] Note that classification can mislabel some queries.
Separating branded from non-branded search turns one ambiguous traffic line into two clear signals: whether people are looking for you, and whether people who are not looking for you are finding you anyway.