Last updated: November 05, 2026
Quick Answer: To track brand and non-brand rankings separately, classify every tracked query before you judge performance, then split the report into brand and non-brand views. For a clean baseline, use a 28-day or 30-day window and keep the same rule set across each update. This article on How to Track Brand and Non-Brand Rankings Separately shows how to do that without mixing the story.
In short: separate brand and non-brand rankings by classifying every tracked query before you judge performance, then filter your rank tracker so brand queries and non-brand queries each get their own report. That keeps brand terms from hiding weakness in non-brand visibility, and it keeps non-brand gains from looking weaker than they are. Search Central explains that search performance reporting depends on consistent filtering and query interpretation, which is why the split matters. See Google Search Central for Search Console reporting guidance and Moz on brand vs. non-brand keyword intent. (Source: Google Search Central and Moz)
Brand keywords and non-brand keywords do not mean the same thing in a rank tracker. If you mix them, average position, visibility, and share of voice all blur together. A query like “Acme shoes” behaves differently from “best running shoes,” even if both belong to the same site. The first can sit near the top because the searcher already knows the brand; the second depends on competition, content quality, and search intent. For practical segmentation, Ahrefs and Semrush both recommend separating branded and non-branded terms in keyword analysis. (Source: Ahrefs and Semrush)
What you need before you split brand and non-brand rankings
You need a keyword list, a rank tracker that can filter or tag queries, and a clear rule for what counts as a branded query. Without those three things, you end up guessing after the report is already built.
Start with the full list of keywords you track today. That may be 50 terms or 5,000, but the list has to be complete enough to include the terms that drive reports. Next, choose one classification rule and write it down. Most teams use a simple rule: a query is brand if it contains the company name, product name, app name, or a common misspelling; everything else is non-brand.
The last piece is the tracker itself. Look for one of these features in the interface:
- tags or groups on each keyword
- a filter for “contains” or “does not contain”
- a label field
- custom segments in reporting exports
- saved views for brand and non-brand
If the tool cannot do at least one of those, you can still split the data in a spreadsheet, but the workflow gets slower and the chance of mistakes goes up. A CSV export from a rank tracker is usually enough to start, as long as it includes the keyword, current rank, URL, and date.
The real job here is not software setup. It is deciding what question the report should answer. Brand rankings answer, “How visible is the business to people who already know it?” Non-brand rankings answer, “How well does the site win new demand?” If those two questions stay mixed, the report can look healthy while the growth engine is weak. For internal planning, pair this step with a tracked keyword spreadsheet and your rank tracker export in the SEO tools and reporting hub. SEO tools and reporting hub
How do I classify branded queries?

Branded queries are the searches that point to your name, product, or close variations of either. The rule works best when it is written as a short taxonomy, not as a loose judgment call made case by case.
Use a three-part classification rule:
- Core brand terms: the exact company name, product line, app name, and official shorthand. If your brand is “Northstar,” then “Northstar” and “Northstar shoes” belong here.
- Brand plus modifier terms: queries that add a navigational or purchase word to the brand, such as “Northstar sale,” “Northstar reviews,” or “Northstar returns.”
- Misspellings and close variants: common typos, spacing differences, or plural forms that clearly still mean the brand.
This is where a lot of teams go wrong. They classify only the exact company name and forget the rest. That leaves branded demand leaking into the non-brand bucket. The report then says non-brand improved, when part of the movement was just brand traffic being mislabeled.
A useful cutoff is semantic clarity. If a searcher typing the phrase almost certainly wants your brand, mark it as brand. If the phrase could fit many businesses and does not name you, keep it non-brand. For example, “Nike Air Max” is branded for Nike. “Running shoes” is non-brand for almost every footwear site. “Nike running shoes for flat feet” is still branded, because the brand is the anchor of the query.
Do not over-classify. A query like “blue running shoes” is not branded just because your site sells blue shoes. That mistake can inflate branded visibility and erase the very signals non-brand analysis is supposed to reveal.
For a stricter version of the rule, many teams use a spreadsheet lookup against a maintained brand term list. That list can include exact names, product families, and known typos, then everything else gets flagged as non-brand unless a reviewer changes it. Even a basic 2-column sheet is better than relying on memory. If the brand set is large or the taxonomy is unclear, it can help to consult an SEO professional before finalizing the rule.
Set up brand and non-brand filters in your rank tracker
Set up two saved views: one that includes brand queries and one that excludes them. That gives you a clean split you can reuse every week without rebuilding the logic.
The exact labels depend on the tool, but the process is the same. Open your tracked keywords, assign each keyword a brand or non-brand tag, and save those tags as filters. If the platform supports rules, you can often automate the split with “contains” logic on the keyword field.
A practical setup looks like this:
- Brand view: keywords tagged brand, or keywords containing brand terms
- Non-brand view: keywords tagged non-brand, or keywords that do not contain brand terms
- All keywords view: the full set, kept for reference only
If the tracker supports regular expressions, one brand rule can catch several variants at once. That helps when the brand has a short root with many endings or a known typo pattern. For example, a rule might include the brand name, product line names, and a list of common misspellings. Use the exact syntax the tool supports; regex formats differ enough that copying a pattern from one platform to another can break the filter.
Keep the same filter logic across locations, devices, and search engines. If your desktop report uses one rule and your mobile report uses another, the comparison stops meaning much. The split should be identical whether you are looking at Google US on desktop, Google UK on mobile, or another tracked search engine. For a related workflow, use the SEO audits and reporting guide alongside your tracker setup. SEO audits and reporting guide
When the setup works, you should be able to switch from brand to non-brand and see the same keyword list shrink in the expected way. A branded query like “Acme pricing” should appear only in the brand view. A generic query like “best accounting software” should appear only in the non-brand view. If a keyword shows up in both, the filter logic is wrong.
For larger accounts, this is the point where a spreadsheet audit pays off. Export the keyword list once, sort by brand tag, and spot-check the edge cases. Twenty minutes here can save a month of bad reporting.
How to keep query classification from breaking your report

Because the split depends on one written rule, keep one owner and one maintained rule for classification, or your report will drift within a month. The weak point is not the rank tracker; it is human inconsistency.
The most common error is changing the rule after the report already exists. Someone decides “brand + product category” should count as brand in one month, then not in the next. That makes the trend line useless because the segment itself changed. If you need a rule change, version it. Write the new rule date, then recalculate the historical report if the tool allows it.
The second error is letting paid search language creep into organic ranking segments. A branded PPC campaign can make branded demand look stronger than organic ranking actually is. Keep ranking segmentation tied to organic query classification, not to campaign naming or ad group naming.
The third error is assuming all branded searches are helpful. They are not. Brand rankings usually have better positions because they are easier to win, but those positions do not prove the site is expanding reach. A report that only celebrates brand gains can hide a drop in non-brand presence that matters more for acquisition.
This is also where query intent matters in a practical way. Brand terms are often navigational or transactional. Non-brand terms can be informational, commercial, or transactional. If you want the report to reflect opportunity, keep that distinction out of the brand split itself. Do not use “intent” as a substitute for “brand.” A query can be commercial and still be branded.
A clean system is usually simple enough to audit in 10 minutes. Each keyword should have one tag, one definition, and one reason for being there. If you cannot explain why a query is in the brand bucket, it probably does not belong there.
What separate brand and non-brand tracking changes in visibility analysis
Separate tracking changes the meaning of your visibility metrics, not just the look of the dashboard. Brand-heavy visibility can make a site look strong even when non-brand rankings are flat, and non-brand weakness can sit hidden behind a strong branded average.
This matters because branded rankings usually sit near the top of the results page. That pulls average position upward and can inflate visibility. A site with strong brand demand may show stable or improving visibility even when it loses ground on the generic terms that bring new users. Without separation, the report answers the wrong question.
Use the split to compare three things:
- Brand visibility: how often the brand appears when people already know it
- Non-brand visibility: how much of the open market the site wins
- Movement between them: whether branded growth is covering up generic decline
If your tracker shows share of voice, treat the brand and non-brand shares as different signals. Brand share of voice can grow because more people search the brand directly, because rankings improved, or because competitors weakened. Non-brand share of voice usually reflects content strength, link competition, and how well pages match search intent. Those are different levers, so they should not be read as one number. Google’s documentation on Search Console performance metrics is a useful reference for how impressions, clicks, and queries are counted. (Source: Google Search Console Help)
A useful habit is to report the two segments side by side in the same date range. A 28-day or 30-day window is often enough for weekly monitoring. Longer windows, such as 90 days, are better when rankings move slowly or when seasonal demand changes the mix. The exact window matters less than using the same one every time.
The split also makes opportunity analysis sharper. If brand rankings are stable but non-brand rankings fall, the problem is not awareness; it is discovery. If non-brand rankings rise while brand rankings stay flat, the site may be expanding beyond existing demand. Those are different stories, and they should not be forced into one line chart.
For a broader framework on reading the metrics once they are split, see the hub on rank tracking data interpretation and diagnostics. For related examples, the SEO metrics glossary and the keyword research guide can help connect the report to search demand. SEO metrics glossary Keyword research guide
Reporting brand and non-brand rankings without mixing the story
Report brand and non-brand rankings in separate rows, separate charts, or separate tabs. One combined chart usually invites the wrong conclusion.
A practical reporting layout is:
- top row: brand keywords
- middle row: non-brand keywords
- bottom row: combined keywords for reference only
Keep the same fields in each row: keyword count, current rank, visibility, and share of voice if the tool provides it. The point is not to create more numbers. The point is to stop one segment from masking the other.
When you write the summary, name the segment before you name the change. Say “brand rankings held steady, but non-brand visibility slipped” instead of “visibility changed.” That small difference forces the reader to see where the movement happened.
Do not let a combined average sit at the top of the page as the main headline. A mixed average can be useful for finance-style summaries, but it is a weak diagnostic tool. If brand is 80% of the tracked set, the combined average will mostly describe brand demand, not market reach. For this reason, keep the combined view as context rather than the headline.
If the site is small and almost every search is branded, the split may look thin. That is normal. In that case, the non-brand set may be too small to carry much meaning, and the report should say so plainly. This method is not for sites whose keyword universe is almost entirely branded, because the non-brand sample may be too narrow to support confident conclusions.
The cleanest report is the one a stakeholder can scan in 30 seconds and still see the difference between awareness and discovery. That is the real value of the split.
Key takeaways
- Separate the list before you read the numbers, or the metrics will blur together.
- A branded query should match your name, product, or a clear variant of either.
- Save two reusable views in the rank tracker: brand and non-brand.
- Keep the classification rule stable, or trend data loses meaning.
- Read brand visibility as awareness and non-brand visibility as new-demand capture.
- Combine the segments only as a reference view, not as the main story.
Why separate brand and non-brand rankings?
Because they answer different business questions. Brand rankings show how well the site captures existing demand for its name; non-brand rankings show how well it competes for broader search demand. If they are mixed, strong brand performance can hide weak acquisition, and weak brand demand can make solid non-brand gains look smaller than they are.
How do I classify branded queries?
Classify a query as branded if it contains the company name, product name, official shorthand, or a clear misspelling or variant of those terms. A written rule and a maintained brand term list are usually the safest practical approach. If a search phrase could fit many businesses and does not name yours, keep it non-brand.
Should navigational searches count as brand?
Yes, most navigational searches should count as brand if they point to your site or brand name. A query like “Acme login” or “Acme support” is still branded, even though the user has a specific destination in mind. The only time to exclude it is when your reporting rule intentionally separates support or account queries from core brand terms.
Drafted with AI; not yet reviewed by a person.