PMax or Shopping? Mostly a control question.
The performance difference between them is smaller than the reporting difference. What you're really choosing is how much you're willing to not be able to see.
Our bias, declared
We run both, and we start most accounts on Standard Shopping even though PMax usually wins eventually. The reason is diagnostic, not ideological: Standard Shopping teaches you which products and queries actually convert, and PMax will not. Once you know that, PMax is a better vehicle for scale. You're handing automation a decision you've already validated, not one you're guessing at.
Side by side
| Factor | Performance Max | Standard Shopping |
|---|---|---|
| Inventory reach | Search, Shopping, YouTube, Display, Discover, Gmail, Maps | Shopping surfaces only |
| Search term visibility | Category-level themes, heavily redacted | Full search terms report |
| Bid control | Target ROAS only, no manual override | Manual, target ROAS, or target impression share |
| Negative keywords | Account-level list, limited campaign control | Full campaign and ad-group negatives |
| Product-level bidding | Listing groups within asset groups, coarse | Down to individual SKU |
| Performance at scale | Usually higher volume at target ROAS | Caps out earlier |
| Speed to results | 2–6 weeks of learning, often volatile | Faster to stabilize |
| Brand traffic separation | Absorbs brand searches unless excluded | Cleanly separable |
| Creative requirements | Full asset set, images, video, headlines | Feed only |
| Incrementality risk | High, easily repackages existing demand | Lower and more legible |
Choose performance max when
You already know which products carry the margin
PMax is a scale tool, not a discovery tool. If your custom labels are accurate and you know which SKUs are worth buying traffic for, automation has something real to optimize toward and will usually beat manual bidding on volume at the same target.
You have real creative assets
PMax reaches YouTube and Display, and it will fill those placements with whatever you give it, including auto-generated combinations you didn't approve. With a proper asset set it earns its extra inventory. Without one it spends your budget on placements representing you badly.
Your catalog is large and long-tail
Beyond a few thousand SKUs, manual product-level management stops being realistic. PMax handles breadth far better than a team can, and the loss of granular control matters less when no human was exercising it anyway.
You're past the diagnostic stage
Once you have a validated picture of which queries and products convert, the search term opacity costs you less. You're no longer using the campaign to learn.
Choose standard shopping when
You're still learning what converts
The search terms report is the single most valuable artifact in a new Shopping account. PMax won't give it to you at the resolution you need, so starting there means paying for education you never receive.
Brand and non-brand need separating
PMax will happily spend on people searching your brand name and report it as conversion volume. Unless you exclude brand explicitly, your reported ROAS includes demand you already had. That's the most common reason PMax looks miraculous in month one.
A minority of SKUs carry the margin
When twenty products drive most of the profit, SKU-level bidding is worth more than incremental reach. PMax's listing groups are too coarse to express that, and margin-negative products end up funded by the winners.
You need to diagnose a problem
When something breaks, like a feed error, a disapproval or a competitor shift, Standard Shopping shows you where. PMax gives you a number that moved and very little explanation, which turns troubleshooting into guesswork.
What the migration actually costs you
Moving from Standard Shopping to PMax is usually framed as free: same budget, better automation. The real cost is a two-to-six week learning period during which performance is volatile and often worse, and you can't roll back cleanly because the historical campaign has stopped accumulating signal.
The second cost is diagnostic. Once brand traffic and non-brand traffic are in the same automated campaign, you lose the ability to answer 'is this incremental?' without running a geo holdout. That test costs real money and most advertisers never run it, which is why so many PMax accounts report strong ROAS alongside flat total revenue.
Our practical recommendation: run Standard Shopping on your margin-carrying SKUs and PMax on the long tail, with brand excluded from PMax at the account level. You keep the diagnostic surface where it matters and give automation the breadth it's really good at. It's more work to maintain than either alone, which is the honest tradeoff.
Related questions
Standard Shopping has been predicted dead since PMax launched and it still exists, so we wouldn't plan around a forced migration. What is true is that new capability lands in PMax first and Standard Shopping isn't receiving meaningful investment. Build your account so a migration is possible, with an accurate feed, clean custom labels and real creative assets, instead of betting on either outcome.
Other comparisons
Services referenced
Get the audit before you decide anything.
Thirty minutes, free, no obligation. We'll tell you honestly whether you need an agency at all, including when the answer is that you don't.