ConsoleKit
Connect GSC
Back to blogConsoleKit / Blog / How-to

How-to

Brand vs non-brand: how much of your SEO growth is really demand for your name?

How to split branded and non-branded searches in Google Search Console, for free, and why non-brand is the number that tells you whether SEO is working.

When clicks from Google go up, someone will want to credit the SEO work. Before they do, ask a simple question: how many of those clicks came from people who were already looking for you?

Someone who types your company name into Google has mostly made up their mind. They'd probably have found you whatever your rankings looked like. Their searches go up when you run a TV campaign, get some press, open a new shop or have a busy season. That's good news, but it's not evidence that your SEO is working.

So the first thing to do with any Search Console report is split it in two:

  • Brand: searches that include your name, its misspellings, or products only you sell. For Nike, that's nike, nkie, and air max.
  • Non-brand: everything else. running shoes, best trainers for flat feet, how to clean white shoes.

Non-brand is where you compete with everyone else for people who haven't chosen yet. It's the better measure of whether your SEO work is paying off.

Option 1: Search Console's branded queries filter

In November 2025 Google added a branded queries filter to the Performance report. Where you have it, it's the quickest way to get the split:

  1. Open Performance > Search results.
  2. Add a filter and choose Branded or Non-branded.
  3. Compare your date ranges as usual.

Google doesn't use a list of keywords for this. An AI-assisted system decides which searches are about your brand, including other languages, typos and searches for products that only you sell. That's clever, but it comes with limits:

  • You can't see or change Google's definition. If it counts a search you think is generic, or misses a sub-brand, you can't correct it.
  • Not every property gets it. It's only for top-level properties (not URL-prefix folders or subdomains), and only once a site has enough searches and impressions.
  • Google says some searches will be misclassified. It's a model making judgement calls, not a rule you set.

If the filter is missing from your property, or you want to control exactly what counts as brand, use a pattern instead.

Option 2: a regex filter you control

Search Console lets you filter queries with a regular expression (regex), a short pattern that matches text. You decide what counts as brand, and you can reuse the same pattern every time.

  1. Open Performance > Search results.
  2. Select Add filter > Query, then change the drop-down to Custom (regex).
  3. Choose Matches regex and enter your pattern. For Nike it might be:
(?i)nike|nkie|niek|air ?max|swoosh
  1. Note the clicks and impressions. That's your brand traffic.
  2. Edit the filter, switch it to Doesn't match regex, and keep the same pattern. That's non-brand.

A few things worth knowing about the pattern:

  • | means "or", so each name or misspelling sits between bars.
  • (?i) at the start makes it ignore capitals. Google matches case-sensitively by default.
  • air ?max matches both airmax and air max. The ? makes the space optional.
  • The pattern matches anywhere in the search, so nike also catches nike running shoes. Watch for short brand names that sit inside ordinary words: a brand called "Ace" would match place and space. Use \bace\b to match it as a whole word only.
  • Search Console uses RE2 syntax, so some features from other tools (like lookaheads) don't work. Patterns can be up to 4,096 characters.

To find misspellings, look through your Queries table for anything close to your name. People search the way they type.

Where the free way runs out

Both options work, and if you check your split once a quarter they may be all you need. If you track it every week, you'll hit some walls.

Brand plus non-brand comes to less than your total. Google hides rare searches to protect people's privacy, and it keeps them out of any total you filter by query. So the two halves won't add up to your unfiltered clicks. Google says so in its own troubleshooting guide. The gap is real traffic you can't assign to either side, and on some sites it's large.

You rebuild it every time. Search Console has no saved filters. Bookmarking the report's URL keeps the pattern, but every check still means two filtered views and copying the numbers somewhere.

You can't see the two lines together over time. To see whether non-brand is growing while brand is flat, you'll be exporting both to a spreadsheet and charting them yourself.

The tables stop at 1,000 rows. That's fine for totals, but if you want to know which non-brand searches grew, long-tail searches can fall off the bottom of the table.

How to read the split

Once you have both numbers, some patterns are worth recognising:

  • Brand up, non-brand flat. Demand for your name grew, probably from marketing, PR or the season. Your SEO didn't get worse, but it didn't drive the growth either.
  • Brand flat, non-brand up. This is what SEO progress usually looks like: more people finding you before they've decided who to buy from.
  • Brand down, non-brand flat. Look outside SEO first: did a campaign end, or did a competitor start bidding on your name?
  • Both down at the same time. Something affected the whole site. Check for technical changes or a Google update before blaming content.

Timing isn't proof, though. Two things happening in the same week doesn't mean one caused the other.

Doing it in ConsoleKit

ConsoleKit turns the split into two saved segments that update every day:

  1. Open Segments and select Define brand and non-brand traffic.
  2. In Brand terms, type your names and misspellings, separated by commas, for example nike, nkie, air max. You can add up to 12.
  3. Select Create required segments. You'll get Brand and Non-brand.

From then on, Overview's Compare segments puts both on one chart, the Queries table gets Brand and Non-brand quick views, and the weekly brief tells you when either one moves. Like Search Console, query segments only count the searches Google reports, so they still add up to less than your total.

Sources

  1. Google Search Central blog: Introducing the branded queries filter in Search Console
  2. Google Search Console Help: Troubleshooting data discrepancies
  3. Google Search Central blog: A deep dive into Search Console performance data filtering and limits
  4. RE2: Syntax reference

Get new articles by email

Research and practical guides on Search Console, about twice a month. Unsubscribe any time.

We only use your email for this newsletter. Privacy policy

Keep goingRead more articles