Google Ads
Work the causes in order rather than jumping straight to feed or creative changes: first confirm Google Ads can actually see your Shopify purchases, then check the conversion goal, Merchant Center feed health, landing pages, pricing, budget realism, and audience signals. Performance Max is a fully automated, black-box bidding system — it optimizes only against the conversion signal it's given, so a tracking gap looks identical to “the campaign doesn't work,” and most PMax accounts that look broken are actually just blind.
PMax bids across Search, Display, YouTube, Discover, Gmail and Maps inventory in a single campaign, deciding in real time where to spend based on one thing: the conversion signal reported back to it. If your Purchase conversion action isn't firing reliably, is firing with a $0 or incorrect value, or is being double-counted, PMax isn't underperforming — it's optimizing against a number that doesn't reflect what's actually happening on your store.
Check this before touching anything else. In Google Ads, open Goals → Conversions and look at your Purchase action's recent status. A “no recent conversions” flag, or a conversion count that doesn't come close to matching real Shopify orders for the same date range, points at a tracking problem, not a demand problem. Full detail on how this breaks on Shopify specifically is in why Google Ads isn't tracking Shopify purchases and Google Ads conversion tracking for Shopify.
Attribix compares Google Ads' reported conversions against actual Shopify orders side by side, which is the fastest way to catch a tracking gap without manually cross-referencing two dashboards for the same week.
Once tracking checks out, confirm PMax is actually targeting purchases. A campaign can technically be “converting” while producing no sales if its primary conversion action is set to something like Add to Cart or a micro-conversion instead of Purchase, or if the account's conversion goal is shared across campaigns in a way that pulls optimization toward a different action than the one you think it's chasing. Check the campaign-level goal settings and confirm Purchase is the conversion action actually driving bids — not just present somewhere in the account.
PMax pulls its product listings straight from your Merchant Center feed. A disapproved product simply doesn't run — it's excluded from the auction entirely, not shown with a warning label. If a meaningful share of your catalog is disapproved, the campaign is spending against a shrunken version of your store without telling you that's what happened. Open the Diagnostics tab in Merchant Center and check your item-level issues and disapproval rate before assuming the campaign itself is the problem. The common causes and how to read them are covered in Google Merchant Center product disapprovals.
A click that lands on a 404, a redirected collection page, an out-of-stock variant with no clear alternative, or a checkout with unexpected friction converts at a fraction of what it should regardless of how well-targeted the click was. Spot check landing pages for the products actually driving impressions — not just your homepage or bestsellers — since PMax will show whichever products in the feed match demand, and those aren't always the ones your team checks most often.
Merchant Center's price competitiveness report shows how your listed prices compare with other retailers advertising similar products. A campaign can generate plenty of impressions and clicks on a fairly matched, well-tracked, well-built product page and still convert poorly if the price is meaningfully above what shoppers are seeing elsewhere in the same results — Shopping and PMax surface price directly, so this comparison happens before the customer ever reaches your site.
Two opposite budget problems both produce “no sales.” A Target ROAS set well above what your account or category has historically supported restricts PMax to only the small slice of auctions it's confident will hit that target — which for a new campaign with little history can mean almost no spend gets through at all. On the other end, a budget too small to generate enough conversion volume gives the algorithm too little signal to optimize with, stretching out the learning period indefinitely rather than letting it exit.
If a Maximize Conversion Value strategy with a Target ROAS isn't producing sales, try removing the target temporarily to see whether spend and conversions pick up — that isolates whether the target itself is the constraint.
Audience signals — customer lists, custom segments, interest categories — don't restrict who PMax can show ads to, but they give the algorithm a starting point during the early phase when it has little else to go on. A campaign launched with no audience signals at all typically takes longer to find its footing than one seeded with a customer list or a relevant in-market or custom segment, simply because there's less for the system to anchor its early guesses to.
Google generally describes a one to two week learning window where PMax is exploring inventory and performance can be erratic. That's a legitimate reason to avoid major structural changes in the first week. It isn't a reason to wave off zero purchases for a month while spend keeps going out. If tracking is confirmed correct, the feed is clean, landing pages work, price is competitive, and budget is reasonable, and you're still at zero sales after a full two weeks, that's a real problem worth escalating — not something to keep waiting out.
We'll check tracking, feed health, and account setup against your actual Shopify orders before recommending a single change.