Why I Still Run Search Ads When Everyone Says PMax Replaced Them
Run standard search when you have high-intent, high-value keywords where you need to control the match type, the ad copy, and the negative list. Run PMax for everything else. That's the honest answer to search campaigns vs Performance Max, and anyone telling you to pick one for your whole account hasn't managed a real budget.
The short answer
Keep search campaigns alive for any keyword group generating more than 15% of your conversions. I set that threshold on a DTC brand I scaled from $100K to $3M a year. Below 15%, PMax's automation does fine because the downside of a bad placement is small. Above it, I want to see every search term report, every negative keyword, every piece of ad copy. PMax hides that data behind "search category insights," which is a nice way of saying you get a sample, not a ledger. When real money is on the line, I don't manage off a sample.
What changes the timing
A few things push you toward keeping search campaigns running longer, or pull you toward handing more over to PMax sooner.
- Brand term volume. If branded search is 30%+ of your account, run it in standard search. PMax will happily spend your budget bidding on your own brand name and call it a win. I've seen this inflate a "ROAS of 9" down to a real 4 once you strip out brand.
- Average order value. On anything over $500 AOV, the cost of a wrong click is high enough that you want match type control. PMax's broad-match-everything approach is built for volume, not precision.
- Sales cycle length. B2B with a 60-90 day cycle needs search campaigns because you need exact-match terms tied to specific funnel stages. PMax optimizes to its last-click conversion signal, which punishes anything that doesn't convert fast.
- Feed quality for ecommerce. PMax leans hard on your product feed. If your feed is thin or your titles are generic, PMax underperforms standard search by a wide margin because it has nothing good to match against.
Signs you are overdue for a search campaigns vs Performance Max review
- Your search terms report (what little PMax gives you) shows irrelevant queries eating spend
- CPA on your top 10 keywords has crept up 20%+ over two months with no explanation
- You can't tell which specific keyword drove a $50K deal, only that it came from "Performance Max"
- Branded search spend is rising faster than branded search volume
- Your agency says "just trust the algorithm" when you ask a direct question about placement
What happens if you wait too long
You lose visibility first, then you lose money, then you lose the ability to diagnose why. I watched this happen on a $2.2B infrastructure project I handled marketing for. We had a small paid search line, maybe $10K a month, for recruiting specialized engineers. Someone suggested folding it into a broader PMax-style automated campaign to "save management time." Within six weeks, cost per qualified applicant doubled because the algorithm was optimizing for cheap clicks from unrelated job seekers, not the five-person talent pool we actually needed. We caught it because I pulled weekly reports. A team checking monthly wouldn't have caught it until the quarter was shot.
That's the real cost of waiting: not the wasted spend itself, but the lag between the problem starting and someone noticing. PMax campaigns don't send up a flare. They just quietly reallocate budget toward whatever converts easiest, which is often not what you actually want to sell.
The most common mistake
People treat search campaigns vs Performance Max as an either/or decision made once, at the account level. It's not. It's a decision made weekly, at the keyword-group level. The mistake is migrating 100% of spend to PMax because Google's rep recommended it, then never revisiting that choice once volume or margins shift. I've audited accounts where the PMax migration happened 18 months ago and nobody has looked at whether it's still the right split since. Run the math quarterly: pull your top 20 keywords by revenue, check whether they're still getting dedicated search budget, and check whether PMax is cannibalizing your branded terms. That's a 30-minute task. Most people never do it because PMax feels like it's "handling things." It is handling things. Just not always the things you'd choose if you were watching closely.
Practical takeaway: Don't ask whether PMax replaced search campaigns. Ask which 10-20% of your keywords are worth the extra control, keep those in standard search, and let PMax run the long tail. Review the split every quarter, not every year.