On March 11, 2026, Google quietly finished rolling out one of the most genuinely useful features Search Console has added in years. The branded queries filter, first announced back in November 2025, is now available to all eligible sites. And while it does not get the breathless coverage that algorithm updates or AI Overviews receive, I think this feature will change how a lot of SEO professionals do their jobs.
I say that because branded versus non-branded traffic segmentation is something I have been doing manually in Search Console for over a decade, and it has always been a pain. You had to set up regex filters, maintain lists of brand variations and misspellings, and hope you caught everything. Most people did not bother, which meant most people were looking at blended data that hid more than it revealed. Now Google does the segmentation for you, automatically, using an AI-assisted classification system that is frankly better than what most of us were doing by hand.
How to set it up
Let me walk through the setup because it is simpler than you might expect but there are a few things to know.
First, open Google Search Console and navigate to your Performance report under Search Results. You need to be looking at a top-level property — this is the domain-level or site-level property, not a URL path property or subdomain property. If you have been using a URL-prefix property like "https://www.example.com/blog/" this filter will not show up there. You need the full domain property.
Once you are in the Performance report, click the "+ New" button above the graph. This is the same button you have always used to add filters for queries, pages, countries, and devices. But now there is a new option in the dropdown. Select "Query" and you will see a new choice alongside the familiar "Queries containing" and "Queries not containing" options. There should now be a "Branded" or "Non-branded" option available.
Select one, click Apply, and the entire Performance report will refresh to show only the data for that query type. All the usual metrics are there — clicks, impressions, average position, CTR — but now filtered to show only branded or non-branded traffic. You can apply this filter alongside other filters too, so you can look at non-branded queries on mobile in Germany, for example.
If you do not see the branded/non-branded option, your site likely does not meet the minimum volume threshold. Google has not published the exact threshold, but from what I have observed, sites with fewer than roughly 500 monthly impressions across both branded and non-branded queries tend not to have the filter available. There is not much you can do about this except grow your traffic, though I suspect the threshold will decrease over time as Google refines the feature.
How the classification actually works
This is the part that surprised me and that I think most people are not paying enough attention to. Google is not using a simple regex match to decide what counts as a branded query. They are using an AI-assisted classification system that goes beyond just matching your brand name.
The system includes your brand name in all languages, which matters if you operate internationally. It catches common misspellings and typos of your brand. And here is the interesting part — it also classifies queries that do not include your brand name at all but refer to a product or service unique to your site. So if you are the only company selling a product called "TurboWidget Pro," a search for "TurboWidget Pro" would be classified as branded even though it does not contain your company name.
I tested this on a few client sites and the classification seems remarkably good. It caught brand name abbreviations I would have missed in my manual regex filters. It handled non-English transliterations correctly. It even picked up on a product nickname that customers use but that we had never explicitly tracked as a brand term.
The one limitation I have noticed is that very new brands or brands with generic names sometimes get misclassified. If your company is called "Fresh" or "Summit," the system has to distinguish between people searching for your brand and people searching for the generic word, and it does not always get that right. But for most brands with at least somewhat distinctive names, the classification is solid.
Why this segmentation matters so much
If you have ever reported SEO results to a client or an executive, you know the branded traffic problem. You run a TV ad campaign, brand awareness goes up, more people search for your brand name, branded traffic spikes, and suddenly "organic traffic is up 30%!" Everyone celebrates the SEO team's success. Except the SEO team did not do anything — the advertising team drove those branded searches.
The reverse is equally misleading. Your SEO work is quietly driving non-branded traffic growth at a healthy 10% per month, but a competitor launches a product that cannibalizes some of your branded searches, and total organic traffic looks flat. The CMO asks why SEO is not producing results and you spend an hour in a meeting explaining that branded and non-branded traffic are different things while everyone's eyes glaze over.
The branded queries filter solves this reporting problem cleanly. You can now show non-branded organic performance in its own chart, completely separated from branded traffic fluctuations. Non-branded organic traffic is the truest measure of SEO effectiveness because it represents people who found you through search without already knowing your brand. That is the audience you are paying an SEO team to reach.
Five things to do with this data right now
I have been playing with the filter since it became available on my client sites and there are several analyses that I think are immediately valuable.
Establish your branded ratio baseline
Look at the last three months of data and calculate what percentage of your total Search Console clicks come from branded versus non-branded queries. Write that number down. For most sites, branded traffic represents 20-40% of total organic traffic, but the range varies wildly by industry and brand maturity. SaaS companies with strong brand recognition might see 50% or more branded traffic. New e-commerce sites might see under 10%.
This ratio is a strategic indicator. A high branded ratio means you have strong brand awareness but may be underinvesting in non-branded SEO. A low branded ratio means your SEO is working to drive discovery traffic but your brand building may need attention. Neither is inherently good or bad — it depends on your business model and growth stage.
Recalculate your true SEO ROI
If you have been reporting total organic traffic as an SEO metric, your ROI calculations are probably wrong. Non-branded traffic is the number that SEO actually influences. Go back to your last quarterly report and recalculate the traffic, conversions, and revenue attributable to non-branded organic queries. I have seen cases where SEO ROI looked mediocre on blended data but stellar when you isolated non-branded performance, and the opposite is also true.
Set up separate tracking dashboards
Create two views in whatever reporting tool you use — one for branded and one for non-branded. Track them independently over time. Set alerts for significant changes in either metric. A sudden drop in branded traffic might indicate a brand perception issue or a shift in competitor advertising. A sudden drop in non-branded traffic might indicate an algorithm update or a technical SEO problem. These are different problems with different solutions, and blending them into one number obscures both.
Use the split for forecasting
This is where the filter gets really powerful. Branded traffic is relatively predictable — it correlates with advertising spend, PR activity, and seasonal brand awareness patterns. Non-branded traffic is harder to predict but more responsive to SEO work. By separating the two, you can build more accurate forecasting models. You can forecast branded traffic based on your marketing calendar and forecast non-branded traffic based on your SEO roadmap, then combine them for total organic traffic projections that are grounded in different causal drivers.
I have been testing this approach with two clients and the forecast accuracy has improved noticeably compared to models built on blended data. When you are projecting organic traffic for a board presentation, that improved accuracy matters.
Audit your keyword cannibalization
The branded filter makes it easy to spot a specific kind of cannibalization that was hard to detect before. Look at your branded queries and check which pages are ranking for them. Are brand searches landing on your homepage? Your about page? Or are they somehow landing on random blog posts? If branded queries are sending traffic to the wrong pages, you have a cannibalization issue that is probably hurting conversion rates. This was always theoretically possible to analyze in Search Console but the manual regex approach was error-prone and time-consuming enough that most people skipped it.
A few things to watch out for
The classification is not perfect and probably never will be. Google's system is making probabilistic judgments about whether a query is branded or not, and it will get some of them wrong. Do not treat the branded/non-branded split as gospel truth — treat it as a very good approximation that is better than what you had before.
Also be aware that the data is retroactive. Google applied the classification to your historical data, so you can see branded/non-branded splits going back as far as your Search Console data extends (up to 16 months). This is great for establishing baselines but be cautious about interpreting historical patterns. The classification is being applied retroactively with the current model — Google has not said whether the model's accuracy was consistent for older queries.
One more thing. The branded filter data is available in the Search Console API, which means you can build automated reports and dashboards that pull branded and non-branded data separately. If you use Looker Studio or a similar reporting tool connected to the Search Console API, update your data source configuration to take advantage of this.
What this means for the industry
I think the branded queries filter is a bigger deal than the SEO industry is giving it credit for. It solves a measurement problem that has plagued organic search reporting for as long as I have been in this field. It makes SEO ROI calculations more honest. It gives SEO professionals better ammunition for demonstrating their value. And it makes it harder for anyone — agencies, in-house teams, or executives — to confuse brand-driven traffic with SEO-driven traffic.
We have been asking Google for better Search Console tools for years. This is one of the rare cases where they delivered something genuinely useful, and I am not going to pretend otherwise just because it is fashionable to complain about Google. The branded queries filter is good. Use it. Set up your dashboards this week. The data is already there waiting for you.