A monthly report says organic search sessions are up eighteen percent. It is a good number and it means almost nothing on its own, because two completely different events produce it. Either more people are finding you who did not know you existed, or a podcast interview sent a wave of people to search your company name.
Both arrive as organic traffic. Only one of them is search work.
Why does the branded split matter more than total organic traffic?
Because branded and non-branded queries measure different departments. Branded search volume is a lagging indicator of everything else the business does — advertising, PR, word of mouth, an existing customer base. Non-branded volume is the closest thing you have to a direct measure of whether the site is winning searches from people who have never heard of you.
Blend them and the larger one moves the total. For most established businesses that is branded, which means a report of combined organic traffic is mostly a report on brand awareness wearing an SEO label. The failure mode runs both ways: a brand campaign can float a flat SEO programme, and a quiet quarter for the brand can bury a genuinely successful one.
Branded search is demand you already created being harvested. Non-branded search is demand being captured for the first time. A single line on a chart cannot tell you which one you are paying for.
What actually counts as a branded query?
Anything containing your company name or a term only your customers would use. That is wider than the name itself, and the edges are where most splits go wrong.
The set that belongs on the branded side:
- The name, including spaced, joined and hyphenated forms.
- Common misspellings. Look at your own query list rather than guessing which ones people type.
- Name plus a modifier — reviews, pricing, login, careers, the name plus a city.
- Proprietary product names that nobody else uses.
- Named people from the business, if they attract search volume of their own.
The awkward case is a company whose name is a normal phrase. If you are called Northside Plumbing, "northside plumbing" is genuinely ambiguous: some of those searchers want you and some want a plumber on the north side. There is no clean answer. Pick a rule, write it down next to the report, and keep it stable — a consistent imperfect definition tells you more over twelve months than a definition that gets refined every quarter.
How do you build the split in Search Console?
With a regex query filter on the Performance report, applied twice. Search Console lets you filter queries by a custom regular expression, and it also lets you invert that filter, which is the whole mechanism.
- Write one pattern that matches every branded form — the name, the misspellings, the joined variants — separated by pipes.
- Filter queries matching it. That total is branded.
- Switch the same filter to the "doesn't match" option. That total is non-branded.
- Export both. Repeat the export monthly, because Search Console does not keep data forever.
Google Search Console is the only free source that has the query behind the click, which is why the split has to start there rather than in your analytics package. Analytics tools see the landing page and the session; they do not see what was typed.
Treating the two filtered totals as if they add up to your reported organic clicks. They do not. Search Console withholds queries made by very few people, so the query-level view is always smaller than the site-level total. Compare branded to branded month over month, not branded to the headline figure.
What do the two numbers actually tell you?
Different things, and they should trigger different conversations.
| Branded | Non-branded | |
|---|---|---|
| What it measures | Demand you already created | Demand you captured new |
| What moves it | Ads, PR, product, reputation | Rankings, coverage, content |
| Typical conversion rate | High — intent is settled | Lower — often still researching |
| A sudden fall means | Check brand activity, not the site | Check rankings and the index |
| Whose result it is | Marketing, broadly | Search work, specifically |
The most useful reading is the ratio over time. Non-branded rising as a share of the whole is the signal that search is building something rather than servicing what already exists.
Where the split changes a decision
Three places, immediately.
Judging a campaign. A programme that lifts non-branded clicks while branded stays flat is working, even if total traffic looks unremarkable. Without the split, that programme looks like it did nothing.
Setting conversion expectations. Branded traffic converts better because those visitors already decided. Holding non-branded pages to the branded conversion rate produces a permanent sense of failure and, eventually, the wrong pages get cut.
Diagnosing a drop. When organic falls, the split tells you which building to walk into. Branded down and non-branded steady is not a ranking problem, and a week spent auditing the site would find nothing.
How often should you look at it?
Monthly for the trend, and immediately whenever the total moves enough to prompt a question. The split is worth almost nothing as a one-off audit and a great deal as a standing pair of lines, because its value is entirely in the comparison to previous months.
Set up the two exports this week and give them a year to accumulate. The first month tells you your current mix. The twelfth tells you whether search is growing the business or following it.
Reporting organic traffic as one number?
We will rebuild your search reporting so you can see which half of the traffic your marketing bought and which half your rankings earned.
Get in Touch →Does this apply to your site?
Reading about it is one thing. Point the scan at your own site and see whether this applies to you, and what it is worth fixing.
Free and unlimited. No account, no card, and you get every finding rather than a teaser.