PMax brand exclusions: cut brand traffic or let PMax keep it?

PMax and brand search: brand exclusions keep the campaign from taking credit for people who were already searching for you, but they also cut reported conversions, and plenty of accounts switch them on without checking what happens next. Take a shoe store that gets around 40% of its conversions from brand queries. With brand left in, PMax's blended ROAS looks great, but some of those sales would have happened anyway, so the number overstates what the campaign actually adds. Excluding brand gives you a truer picture, but then you need a separate brand campaign so that spend stays controlled and visible. My leaning is to exclude brand from PMax and run it on its own. Has anyone run a holdout on brand traffic and seen how far the numbers moved?

Honestly it depends on how much of your brand demand you already capture organically. If you rank first for your own name, have lots of reviews and get plenty of direct traffic, most brand clicks would convert without an ad, so the brand campaign is mostly insurance. In that case Iris's split makes sense, and PMax's number drops because it was taking credit it didn't earn. But if a competitor bids on your name, or your organic listing sits below a marketplace or directory, the brand ad is doing real work, and cutting it just hands those clicks to someone else.

You can get a rough read without a full holdout. Pull branded queries from Search Console for the last few months and look at clicks and average position. Then compare total conversions, not just PMax ROAS, for a few weeks after the exclusion. If total conversions fall far more than the brand share would predict, the ads were adding something.

Search Console only counts organic clicks, so any brand query where your ad takes the click won't show up there at all. If you compare those organic numbers against the same brand terms in your Search terms report, the gap is roughly what your ads are actually winning.

PMax doesn't give you a clean search terms report for brand queries, so you can't cleanly subtract its share from organic. Put brand in its own Search campaign first and the gap becomes readable, because that report is complete. After that you can decide whether PMax should carry brand at all.

My rule of thumb: if you hold position 1 for your name, nobody is bidding on it, and most brand clicks convert without an ad, keep the brand campaign on a small cap. If something sits above you or a competitor is bidding, leave it at full budget. Either way, judge it on total conversions, not PMax ROAS.

Add one check to that cap rule: watch lost impression share due to rank on the brand campaign. If it climbs after you cap it, those clicks are going to whoever ranks above you, so the saving isn't real.

I'm not sure that metric tells you what you think it does. A budget cap shows up as impression share lost to budget, not to rank, because rank comes down to bids and Quality Score, not how much you're allowed to spend. So after you cap, watch the budget figure. Track rank separately in case a competitor starts bidding on your brand name.

Agreed on the metric. After a cap, what matters is whether total brand conversions dip, since rank loss can climb just because a competitor started bidding, even if you changed nothing. Budget loss tells you the cap is biting, while rank loss tells you someone outbid you.

Rank loss doesn't always mean someone outbid you. Quality Score can slip with your bids unchanged, say when the landing page gets slower or the ad copy drifts away from the brand query. Auction insights will tell you which one you're dealing with. If a new domain shows up in the overlap table, a competitor is bidding. If nobody new appears and you're still losing rank, check ad relevance and page speed before raising the bid.