Brand searches are the most stable traffic most sites have. People who type your company name are looking for you, and you usually rank first for them. That stability is useful for the business, but it is a problem for update analysis. When branded clicks make up a large share of organic traffic, they can cushion the site-wide total and hide a sharp drop in the non-branded searches that core updates actually reshuffle.
For years, separating the two meant building regex filters by hand and maintaining lists of misspellings. In November 2025, Google added a branded queries filter to Search Console's Performance report, which classifies queries as branded or non-branded for you. For eligible sites, that makes it much easier to see an update's true effect.
This guide explains how the filter works, its eligibility limits, a step-by-step workflow for update analysis, and how to combine it with your own regex where the automatic classification falls short. For the strategic difference between the two query types, see our guide to branded versus non-branded keywords.
Key Takeaways
- Google announced the branded queries filter on 20 November 2025 and rolled it out gradually; in March 2026 Search Engine Land reported it was available to all eligible sites.
- Branded queries include your brand name, variations, misspellings and brand-related products or services; Google classifies them with an internal AI-assisted system.
- At launch, Google said the filter was only for top-level properties with sufficient query and impression volume.
- Core updates mostly show up in non-branded queries, so analyse branded and non-branded performance separately.
- Keep a regex fallback for properties that are not eligible, and sanity-check the automatic classification on your top queries.
How the branded queries filter works
The filter sits in the Search results Performance report. Search Engine Land's coverage of the launch reported Google's definition: a branded query includes your brand name, and misspellings and the names of brand-related products or services also count. Its later article on the expansion described how classification is done:
“Google uses an internal AI-assisted system to determine whether queries are branded.”
— Search Engine Land, Google Search Console branded queries filter expands
Google also added an Insights report card that, in its words as reported by Search Engine Land, shows the breakdown of total clicks for branded versus non-branded traffic. Both are reporting features. They change what you can see, not how you rank.
| Fact | Detail | Source |
|---|---|---|
| Announced | 20 November 2025, rolling out gradually over the following weeks | Search Engine Land, 20 November 2025 |
| Broader availability | Reported as available to all eligible sites | Search Engine Land, 11 March 2026 |
| Classification | Internal AI-assisted system; brand names, variations, misspellings, brand products and services | Search Engine Land |
| Eligibility | Top-level properties only (not URL-path or subdomain properties); sufficient query and impression volume | Search Engine Land, 20 November 2025 |
| Search types | Web, Image, Video and News | Search Engine Land, 11 March 2026 |
| Effect on ranking | None; it is a reporting filter | Reporting feature only |
“This is only available for top level properties.”
— Google, as quoted by Search Engine Land, Google Search Console adds branded queries filter
Why branded data distorts update analysis
Imagine a site where half of organic clicks come from brand searches. A core update cuts non-branded clicks by a third, while brand clicks are unchanged. The site-wide total falls by about a sixth, which might look like normal noise in a weekly report. The non-branded line tells the real story. The reverse also happens: a TV campaign or press coverage can lift branded clicks during an update and make a non-brand loss look like a gain.
Brand searches also behave differently from the queries updates usually affect. You generally rank first for your own name regardless of what a core update does to your informational pages. Separating them is the cleanest way to see whether Google's reassessment of your content, as opposed to demand for your brand, moved your traffic.
Step-by-step: isolating update impact
- Confirm the update window. Note the start and end dates from the Search Status Dashboard, and wait at least a week after completion, as Google recommends for core updates.
- Choose equal periods. Compare a period after completion with an equal period before the start, avoiding holidays and campaign launches. Our guide on measuring update recovery covers period selection.
- Apply the non-branded filter. In the Performance report, add the branded queries filter and select non-branded. Record clicks, impressions, CTR and average position for both periods.
- Repeat for branded. Switch to branded and record the same metrics. If brand demand changed, check marketing activity and Google Trends.
- Break non-branded down further. Look at pages, countries and devices within the non-branded view. Our guide to query-level analysis after an update shows how to find the losing query groups.
- Check for SERP feature effects. If non-branded positions held but CTR fell, read about rankings versus traffic impact before concluding you were demoted.
- Annotate and report. Mark the update and your findings on the chart with custom annotations, and report branded and non-branded results separately.
| Non-branded result | Branded result | Interpretation |
|---|---|---|
| Down | Stable | Likely update impact on non-brand visibility |
| Stable | Down | Brand demand fell; look at marketing, PR or seasonality |
| Down | Down | Possible site-wide issue: check technical problems, tracking and manual actions |
| Up | Stable | Update likely helped non-brand visibility |
| Stable | Up | Brand campaign or coverage effect, not an update |
When the filter is not available
If your property is a URL-prefix or subdomain property, or lacks the volume Google requires, build your own split. Use regex in Search Console to include or exclude queries that contain your brand name, common misspellings and product names. A pattern such as (acme|acmee|akme) with the "doesn't match regex" option gives a rough non-branded view. It will not catch every variation, but it is consistent over time, which is what matters for before-and-after comparisons.
Even when the filter is available, spot-check the classification on your top 50 queries. Product names that are also generic words, and brands that share names with other things, are the most likely places for disagreements between your list and Google's.
Using the split beyond updates
The same split helps you measure brand health. Rising branded clicks often reflect marketing, PR and word of mouth, which you can explore in our brand awareness guide. Non-branded clicks reflect SEO reach. Reporting both, rather than a single organic total, gives stakeholders a far clearer picture. Our Performance report guide covers the other filters worth combining with it.
Common mistakes
- Reporting only site-wide totals during an update. Brand traffic can hide or exaggerate the effect.
- Comparing different filters across periods. Switching from regex to the built-in filter mid-analysis breaks the comparison.
- Assuming the classification is perfect. Spot-check your most important queries.
- Ignoring brand demand changes. A branded drop is usually a marketing or demand signal, not a ranking problem.
- Expecting the filter on every property. At launch it was limited to top-level properties with enough volume.
- Thinking the filter affects rankings. It only changes how you view data.
Related Guides
- Checking Crawl Stats and Indexing After an Algorithm Update
- Cleaning Up User-Generated Spam After a Spam Update
- Recovering When a Hacked Site Gets Caught in a Spam Update
Frequently Asked Questions
When did Search Console add the branded queries filter?
Google announced it on 20 November 2025 and rolled it out gradually. In March 2026, Search Engine Land reported it was available to all eligible sites.
Why don't I see the branded queries filter?
At launch, Google said it was only available for top-level properties with sufficient query and impression volume. URL-path and subdomain properties were not eligible.
What counts as a branded query?
Queries containing your brand name, variations or misspellings, and names of brand-related products or services, as classified by Google's AI-assisted system.
Does the branded filter change my rankings?
No. It is a reporting filter in Search Console and has no effect on how Google ranks your pages.
Why focus on non-branded queries after an update?
You usually rank first for your own brand whatever an update does, so non-branded queries show the effect of Google's reassessment of your content more clearly.
Conclusion
The branded queries filter removes one of the biggest sources of confusion in update analysis. Split branded from non-branded, compare clean periods, dig into non-branded pages and queries, and report the two lines separately. If you would like help reading what an update did to your non-brand visibility, our SEO services team can run the analysis with you, or you can request a quote.
References
- Google Search Console adds branded queries filter — Search Engine Land
- Google Search Console branded queries filter expands — Search Engine Land
- Introducing the branded queries filter in Search Console — Google Search Central Blog
- Google Search's core updates and your website — Google Search Central
- Debugging drops in Google Search traffic — Google Search Central
