Why Your Google Brand Search ROAS Is the Most Misleading Number in Your Account

The campaign that looks like your best performer is probably not doing what you think it is.

Open any Google Ads account and look at the brand search campaign. The one where you bid on your client’s own brand name. The one where someone types “Harbor Coffee” or “Modern Realty Group” and your ad appears at the top.

The ROAS is probably exceptional. Maybe 6x. Maybe 10x. Maybe higher.

It is also almost certainly the most misleading number in the entire account.

Here is why.

The person who typed your client’s brand name into Google already knew the brand. Something made them aware of it before they opened a browser and searched for it by name. Maybe it was a Meta ad they saw last week. Maybe it was a TikTok video. Maybe it was a friend’s recommendation, a podcast, a display ad, a Reddit thread. Maybe it was an awareness campaign you have been running for three months that has no measurable ROAS because it was not designed to convert — it was designed to build the recognition that eventually produced this Google search.

Google brand search did not create this customer’s intent. It captured it. Your campaign is standing at the exit of a journey that other campaigns built, collecting the conversion credit, and reporting a ROAS that makes it look like the star of the show.

This matters enormously for budget decisions.

If you evaluate your marketing mix by ROAS and use it to determine where to allocate budget, brand search will consistently look like your best-performing channel. The logical next step is to put more money into it. But brand search has a ceiling — you can only capture as much branded search volume as exists. If you increase brand search budget, you will not create more people searching for your client’s brand. You will pay more for the same intent that already exists, and CPC will rise as a result.

Meanwhile the campaigns that are actually generating new customer awareness and consideration — the Meta prospecting campaigns, the TikTok content, the YouTube pre-roll — will continue to look weak on ROAS because last-click attribution gives them no credit for the conversions they initiated. Budget pressure on those campaigns will reduce new audience acquisition. Fewer new people will discover the brand. Branded search volume will start declining. And your brand search ROAS will also start declining, for reasons that are invisible if you are only looking at the channel in isolation.

This is the death spiral that ROAS-based allocation creates when applied uniformly across campaign types. And it plays out in accounts everywhere, gradually, over quarters, before anyone identifies the cause.

**The right way to evaluate brand search is not ROAS. It is impression share.**

Is the campaign capturing 90%+ of branded search queries? If yes, the defense is working and the budget is appropriate. If impression share is declining, competitors may be bidding on the brand’s terms — a strategic problem that needs to be addressed but is a very different problem from a ROAS optimization question.

Brand search budget should be sized to maintain strong impression share at reasonable CPC, nothing more. It is a defensive cost of doing business, not a growth lever. Treating it as a growth lever because it produces great ROAS numbers will eventually starve the campaigns that are actually building the brand.

The campaigns that deserve budget investment are the ones that generate new demand — not the ones that harvest demand that other campaigns already created.

Knowing the difference, and building an evaluation framework that measures each campaign against the right standard, is one of the most practically valuable things an agency can do for its clients’ long-term performance.

—
Kaivo is an AI-native advertising intelligence platform built for digital agencies.