If your Google Ads account is spending budget but delivering poor results, and you are running Performance Max campaigns, there is a high probability that either your conversion tracking is misconfigured or your PMAX campaign structure is working against you. These two problems are deeply connected: PMAX uses conversion signals to decide where to spend, so if the signals are wrong, the spend goes wrong. This guide walks through both issues and gives you a practical audit checklist to diagnose and fix them.
Why PMAX Underperforms When Tracking Is Broken
Performance Max is a fully automated campaign type. It places ads across Search, Display, YouTube, Gmail, and Maps simultaneously, using Google’s AI to allocate budget toward the placements and audiences most likely to convert. That sounds powerful, but the system is only as good as the conversion data it is optimizing against.
When conversion tracking is misconfigured, three things happen. First, Google optimizes toward false signals, like add-to-cart events being recorded as purchases. Second, PMAX distributes budget based on inflated conversion counts, making it appear certain placements or products are performing when they are not. Third, your reported ROAS looks better than your actual business revenue, hiding the true cost of the problem.
Fixing tracking before touching PMAX structure is the correct sequence. You cannot make good structural decisions with bad data.
Step 1: Audit Your Conversion Setup
Open Google Ads and navigate to Tools and Settings, then Conversions. Look at every active conversion action and ask these questions for each one:
Is This a Real Business Conversion?
The most common tracking mistake is using weak signals as primary conversion actions. Add-to-cart, page views, and session duration are useful for analysis but should never be set as primary conversions in a PMAX or Smart Bidding campaign. Google will optimize aggressively toward whatever you call a conversion, and if that signal does not represent actual revenue, your spend will chase the wrong behavior.
Set only one conversion action as your primary: the purchase, the lead form submission, or the booked appointment. Everything else should be set to secondary or observation-only.
Is the Thank-You Page Event Firing Correctly?
The most reliable conversion setup is a thank-you page tag. After a purchase or form submission, the user lands on a confirmation URL that fires a conversion event. Go to Google Tag Manager and check that your purchase or lead conversion tag is firing on the correct URL and only on that URL. A misconfigured trigger that fires on every page will inflate your conversion count massively.
After confirming the tag setup, use Google Tag Assistant to verify that the conversion tag is firing correctly on a real test transaction. This takes 10 minutes and can reveal double-firing issues that have been distorting your data for months.
Do You Have Server-Side Conversion Tracking Enabled?
Browser-side tracking (via GTM tags alone) loses data due to ad blockers, iOS privacy restrictions, and cookie limits. For any account spending more than a few thousand dollars per month, server-side conversion tracking via Google’s Conversion API is essential. This sends conversion events directly from your server to Google, bypassing browser limitations and dramatically improving match rates.
Check your Google Merchant Center if you are running ecommerce. Make sure your product feed is connected, product data is clean and approved, and that you are using enhanced conversions for web, which matches hashed customer data to Google accounts for more accurate attribution.
For a complete walkthrough of browser-side and server-side setup across both platforms, our guide to conversion tracking for Meta and Google Ads covers implementation and verification for each tracking method.
Audit rule: if your Google Ads conversion count is significantly higher than your Shopify or CRM order count for the same period, you have a double-firing or weak-signal problem. Fix tracking first, then restructure PMAX.
Step 2: Check UTM Consistency
Every URL in your Google Ads campaigns should have consistent UTM parameters appended. These flow through to Google Analytics and your CRM, allowing you to reconcile ad-reported conversions with actual revenue in your backend systems.
Common UTM errors include: auto-tagging conflicts with manual UTMs, UTMs being stripped by redirect chains, and campaigns using inconsistent naming conventions that make reporting impossible to read. Audit 10 to 20 destination URLs in your account and confirm the UTMs are present and consistent. Then cross-reference the sessions those UTMs generate in Analytics against the conversion count Google Ads reports for the same period. A large discrepancy means attribution is broken somewhere in the chain.
Step 3: Audit Your PMAX Campaign Structure
Once tracking is clean, look at how your Performance Max campaigns are structured. Several structural patterns will consistently drag down performance.
Mixing All Products in One Campaign
Running all products in a single PMAX campaign prevents you from controlling budget allocation across product categories. If you have high-margin items and low-margin items in the same campaign, Google will spend budget on whatever drives the most conversions, which is often the cheapest product, not the most profitable one.
Segment your PMAX campaigns by product category, price band, or margin tier. A luxury goods account might have one PMAX campaign for items over 500 USD and a separate one for accessories under 100 USD. This gives you separate budgets, separate asset groups, and separate ROAS targets for each tier.
For catalog-based businesses where inventory items are unique or one-of-a-kind, our guide to Performance Max for one-of-a-kind catalogs covers the specific feed structure and asset group approach needed when no two products are identical.
Leaving Sold-Out or Out-of-Stock SKUs Active
PMAX learns from historical performance data at the product level. If out-of-stock items are still active in your campaign, you are wasting budget sending traffic to dead-end pages and polluting the campaign’s learning data with no-conversion signals. Exclude out-of-stock products from your feed or create a suppression rule in Merchant Center to prevent them from being served.
No Audience Signals
PMAX can run without audience signals, but it performs significantly better when you provide them. Add your customer match lists, website visitors, and high-intent custom intent audiences as signals in your asset groups. These are hints to Google about who is most likely to convert, which accelerates the learning phase and improves targeting precision from day one.
No Parallel Search Campaigns for High-Intent Queries
PMAX absorbs search traffic and by default takes priority over standard search campaigns for overlapping queries. But PMAX search placements often lack the granular keyword control you need for your highest-intent branded and competitor terms.
Run dedicated standard search campaigns for branded keywords and high-converting exact match terms in parallel with PMAX. Use brand exclusions in PMAX settings so it does not cannibalize your branded search traffic, and let PMAX focus on discovery and new audience acquisition while your search campaigns capture confirmed demand.
Step 4: Add Negative Keywords (Yes, in PMAX)
Performance Max has limited negative keyword support compared to standard search campaigns, but you can apply account-level negative keyword lists that affect all campaign types including PMAX. If you are seeing irrelevant search term placements, create an account-level negative list in the Shared Library and add your exclusions there.
Common negative keyword categories for PMAX: competitor brand terms (if you do not want to bid on them), unrelated product categories, informational modifier terms (“free”, “DIY”, “how to”), and geographic exclusions if needed.
The Full Audit Checklist
Here is the complete checklist to work through, in order:
- Verify every conversion action: primary action is purchase or lead only; weak signals set to secondary
- Confirm thank-you page tag fires once, on the correct URL, using Tag Assistant
- Enable server-side Conversion API if not already active
- Check UTM consistency across 10 to 20 destination URLs
- Cross-reference Google Ads conversions vs. backend order count for the same 30-day period
- Confirm Merchant Center feed is approved with no product disapprovals
- Segment PMAX by product category or price band if mixing all products in one campaign
- Exclude out-of-stock and discontinued SKUs from all active campaigns
- Add customer match and website visitor audience signals to PMAX asset groups
- Create dedicated branded search campaigns and exclude brand from PMAX
- Build account-level negative keyword list for irrelevant traffic
Timeline and What to Expect After Fixing
The audit itself takes 48 to 72 hours to work through properly, including time to place test orders and verify tracking. Once fixes are implemented, expect a 2 to 4 week stabilization period while PMAX re-learns on cleaner data. Reported ROAS may temporarily drop as false conversion inflation disappears, but actual revenue per dollar spent should improve as the algorithm starts optimizing toward real buyers.
Accounts that have had weak conversion signals running for months will need longer to recover. Be patient with the first 30 days post-fix, watch the trend direction rather than day-to-day fluctuations, and compare performance to the same period from the prior year rather than the immediately preceding (corrupted) months.
Is Your Google Ads Account Optimizing Toward the Wrong Signal?
We audit and rebuild Google Ads accounts for ecommerce and service businesses. If your PMAX is spending but not converting, let us find out why.
