Performance Max vs Standard Shopping is the first argument I have with almost every new e-commerce client. Somebody read that PMax replaced Shopping, or that Shopping is dead, or the opposite. Neither is true in 2026. Both campaign types are alive, they can run side by side on the same products, and since late 2024 Google lets them compete on equal terms. So the useful question is which one goes first for your store, and how you prove when it’s time to add or switch to the other.
Here’s the rule I use, what changed on Google’s side, and the test plan I run before moving budget.
What each campaign actually gives you
Standard Shopping shows product ads from your Merchant Center feed on Search, the Shopping tab and Images. You choose how products are split into product groups, you set the bids (or pick a Smart Bidding strategy), you read every search term and add negatives. When a product burns money, you can see it and stop it the same afternoon.
Performance Max takes the same feed plus whatever assets you give it and serves across all Google inventory from one campaign: Search, Shopping, YouTube, Display, Discover, Gmail and Maps (About Performance Max campaigns, as of 2026-10-24). It bids with Smart Bidding toward conversions or conversion value. Remarketing comes built in.
PMax is a lot less of a black box than it was at launch. You get search terms, a channel performance report, campaign-level negative keywords and brand exclusions, and Google is rolling out channel prioritization across campaign types (About channel prioritization, as of 2026-10-24). What you still don’t get is a bid per product or a promise about where the money goes.
| Standard Shopping | Performance Max | |
|---|---|---|
| Inventory | Shopping placements on Search, Shopping tab, Images | All Google channels from one campaign |
| Bidding | Manual CPC or Smart Bidding | Smart Bidding only |
| Control over products | Product groups, bid per group, campaign priority | Listing groups per asset group |
| Search terms and negatives | Full | Search terms report, campaign negatives |
| Remarketing | Separate campaign | Built in |
| Needs from you | Attention and a clean feed | Conversion data, budget, assets, a clean feed |
What changed: PMax no longer wins by default
Until late 2024, PMax got automatic priority over a Standard Shopping campaign selling the same products in the same account. Running both was mostly pointless, because Shopping only got what PMax left over.
Google changed that in October 2024. In its own words, PMax “will no longer be automatically prioritized when the campaigns are in the same account targeting the same products. Instead, Ad Rank will determine which campaign serves an ad” (New Performance Max features, October 23, 2024). Google’s page on how PMax interacts with other campaigns confirms that the campaign with the highest Ad Rank serves when several could take the same impression (as of 2026-10-24).
One exception still holds: a Search campaign with an exact match keyword for the query wins over PMax (About Performance Max campaigns, as of 2026-10-24). Brand terms belong in a Search campaign anyway.
In practice, running Shopping and PMax on the same products is now a legitimate setup, and you can let the auction decide product by product. It also means overlap is something you design on purpose. If both campaigns hold the same product, you can no longer assume which one pays for the click.
The decision rule I use
I look at three things before choosing. None of them is the size of the brand or how “advanced” the account feels.
1. Conversion volume. PMax learns from your purchase data. If the store records a purchase only every few days, Smart Bidding has very little to work with, and PMax spreads a small budget across YouTube and Display while it guesses. I want the purchase action reporting sales most days, from one clean source, before I give an algorithm that much room. If tracking is in doubt, I fix that first with the Shopify Google Ads conversion tracking audit.
2. Feed size and concentration. A catalog of 40 products where five do most of the revenue is easy to manage by hand in Shopping, and you’ll learn more from the search terms than PMax will tell you. A catalog of 8,000 SKUs with a long tail is a different job. Nobody bids on that product by product, and PMax (or Shopping with a target ROAS) is usually the better fit.
3. Margin spread. If every product carries a similar margin, a single ROAS target works fine and PMax can optimize toward it. If margins range from thin to fat, I want separate control over each band: custom labels in the feed, then either separate Shopping campaigns or separate PMax asset groups per label. With wide margin spread and low volume, Standard Shopping wins easily, because PMax can’t learn the difference with so little data.
How that plays out:
- New store, low volume, small catalog. Standard Shopping first, one campaign, little segmentation. Learn which products and queries sell.
- Steady sales, Shopping profitable. Test PMax against Shopping with a real experiment (below). I often start PMax with only the feed and no extra assets. That way the test measures the campaign type, not whether we had good video.
- Large catalog, good data, similar margins. PMax can go first, with a Shopping campaign kept for products that need tight control.
- Mature account. Usually both. PMax for reach, Shopping for products where I want a fixed bid or a clean read of search terms.
Whichever you start with, the feed decides more than the campaign type does. Bad titles, missing GTINs and wrong prices hurt both, and a suspended Merchant Center account switches off both at once.
The 2026 twist: AI Max for Shopping
There’s a third option now. AI Max for Shopping is a beta set of features you switch on inside a Standard Shopping campaign. It can rewrite product titles to match the query and, with Final URL expansion, send traffic to category or collection pages it finds on your site (About AI Max for Shopping campaigns, as of 2026-10-24).
The details matter. Text customization is limited to English feeds for now, Final URL expansion requires a target ROAS bid strategy, and Google warns that tracking templates without a proper {lpurl} tag can send people to broken landing pages. So it’s no shortcut around PMax. It’s a way to get some of the AI reach while keeping Shopping’s campaign structure. For an English-language store already running Shopping on target ROAS, I test it before I test a full move to PMax.
How to test PMax against Standard Shopping
Don’t test by launching PMax next to Shopping and comparing the two campaign rows. The campaigns overlap, attribution moves between them, and you’ll end up reading noise. Use the Experiments page.
Google has a built-in experiment for exactly this: Standard Shopping vs Performance Max (as of 2026-10-24). You choose an existing Shopping campaign, split traffic between control and treatment, and either pick an existing PMax campaign or have Google build a new one. Note that Google’s uplift experiments don’t support PMax with a Merchant Center feed (Experiments FAQs, as of 2026-10-24), so for a store this “versus” test is the one you want.
My plan:
- Freeze tracking. One primary purchase action with real values. Don’t change conversion settings during the test.
- Match the scope. Same products in both arms, and make sure no other campaign targets them. Google recommends the same target ROAS in PMax as in the Shopping campaign you’re testing against.
- Pick the split. A 50/50 traffic split gives the quickest answer. If the Shopping campaign carries most of the revenue and you’re nervous, go lower on the treatment side and accept a longer test.
- Budget for the treatment. Google’s FAQ says the bigger PMax’s spend is relative to the account, the better your chance of a significant result.
- Run it long enough. Google recommends at least 4 to 6 weeks and 1 to 2 conversion cycles, and it discards the first 7 days of data as ramp-up. Avoid changing budgets or targets in the first two weeks.
- Judge on profit. Compare conversion value, ROAS and, if you can, margin per arm. Check the PMax channel report so you know whether new revenue came from Shopping placements or from Display and YouTube.
- Decide what “apply” means. If you apply a winning PMax arm, Google makes PMax take 100% of traffic and pauses the Shopping campaign automatically (About Performance Max experiments, as of 2026-10-24). If you want to keep both, end the experiment and set it up by hand.
Don’t run this test in November. Black Friday traffic will tell you about Black Friday, not about the campaign type.
Conclusion
Start with Standard Shopping when data is thin, the catalog is small or margins vary a lot. Bring in PMax once purchases flow every day and tracking is clean, and prove it with a proper experiment before moving budget. Ad Rank now decides between the two, so running both is a valid setup. If you want someone to set up the test or read the results with you, that’s part of my Google Ads for e-commerce work. Schedule a call and bring your current campaign structure.