A Performance Max report shows strong ROAS. The business owner asks how much came from people who already knew the brand. That is a useful question, but the answer requires more than removing the company name from a keyword list.
A brand-traffic audit has two outputs. It describes the evidence available about branded demand, and it defines what decision the business wants to make about that demand. Whether to retain, separate, or restrict brand coverage depends on that decision.
Define what counts as branded intent
Create a working dictionary with the company name, distinctive product names, common misspellings, and relevant brand combinations. Separate your own brand from brands you resell and from competitor names.
Some terms are ambiguous. A product name might also be a generic category phrase. Mark uncertain classifications for review instead of forcing every query into a confident branded or nonbranded bucket.
Keep the dictionary version with the report. If the definition changes next month, an apparent change in branded share may be a classification change rather than a change in customer behavior.
State the campaign's commercial role
Ask whether the campaign is expected to acquire new customers, capture existing demand efficiently, support a promotion, or cover a broad product catalog. Several roles may coexist, but they need different evaluation criteria.
For example, a reseller may intentionally advertise manufacturer brands to reach relevant shoppers. A business measuring new demand for its own brand may want a different separation. A blanket instruction to exclude every brand name would ignore that distinction.
Record who owns the decision and which other campaigns serve branded searches. A Performance Max edit can shift traffic elsewhere in the account, so the audit must look beyond the campaign being edited.
Assemble the evidence you can actually see
Export the available search-term or search-category evidence, campaign results, relevant channel views, and changes over the review period. Record any reporting limits. Do not label a partial query view as a complete census of every impression.
Classify visible terms with the dictionary and inspect a sample manually. Compare spend, attributed conversions, and value for the categories you can support. Keep unknown or unclassified rows visible.
Also record brand Search activity, organic brand demand where available, promotions, and major publicity. These are contextual signals, not a formula that automatically assigns credit to one channel.
Verify the control before proposing an exclusion
Google's Performance Max brand exclusion documentation and brand exclusion setup guide describe controls whose inventory scope and available options need to be checked in the actual campaign. Do not assume one brand setting excludes every possible brand-related exposure across all channels.
Record the brand list, campaign scope, relevant inventory options, and any Shopping exception available in the current interface. Read the saved configuration after applying a change. A proposal that merely says “remove brand” is too vague to audit.
The search themes and negatives guide explains why adding guidance signals is different from establishing exclusions.
Use a control inventory
| Item | Evidence to save | Why it matters |
|---|---|---|
| Brand definition | Versioned term dictionary | Keeps classification consistent |
| Campaign role | Written acquisition or coverage objective | Determines what success means |
| Existing coverage | Other relevant campaigns and settings | Reveals possible traffic shifts |
| Proposed control | Exact list and saved scope | Makes the change reproducible |
| Reporting coverage | Available views and known omissions | Prevents false precision |
| Business outcome | Total orders, new customers, contribution | Checks beyond campaign credit |
This inventory also helps when an agency and an internal team use different meanings of “brand.” Resolve the definition before debating the reported percentage.
Do not equate branded revenue with wasted revenue
An existing customer can still respond to an ad, and a branded query can occur at different stages of a purchase decision. Conversely, an attributed purchase may have occurred without the ad. The query label alone does not resolve that counterfactual.
Use the attribution and incrementality guide to choose an appropriate evidence standard. A reporting audit can identify concentration in branded demand. A causal claim about additional revenue requires a design capable of answering that question.
If a geographic test is feasible, the holdout planning guide outlines issues such as region comparability and spillover. It is not a requirement for every operational change, but it is more informative than treating a before-and-after ROAS change as causal proof.
Plan for redistribution after a change
Consider an illustrative account where branded conversions decline in Performance Max after a restriction, while a separate brand Search campaign records more conversions. The Performance Max report looks weaker, but that alone does not show the business lost sales.
Review total spend, total business outcomes, and relevant account-level shifts. Keep promotions and measurement changes in the record. If total sales also move, investigate whether the timing and test design support an explanation or merely show correlation.
Choose a review window that allows the conversion cycle to mature. Avoid declaring success because a campaign's nonbranded share rose; the business still needs to know whether the new mix serves its economics.
Finish with a bounded conclusion
The audit should state the visible brand concentration, reporting limitations, confirmed controls, and the next decision. “Visible query evidence is heavily branded, and the current objective requires a controlled separation test” is stronger than an unsupported claim that all branded conversions were stolen from organic search.
Preserve the original report and settings. The next reviewer should be able to see what changed and why, then compare the business outcome with the intended role of the campaign.