Why Is My Performance Max Campaign Spending But Not Getting Any Conversions?

The most common cause is that Performance Max is optimizing toward the wrong signal, not that it is broken. If your conversion tracking is misconfigured, your conversion goal is set to something low-intent (like page views instead of purchases or leads), your budget is too small for your target cost per action, or your asset groups have weak Ad Strength and no useful audience signals, the system’s machine learning has nothing solid to learn from. It will keep spending because the campaign is technically “delivering,” it just cannot tell what a good outcome looks like, so it buys cheap, low-quality inventory instead.

Performance Max is the most automated campaign type Google Ads offers. That is exactly why it fails quietly. There is no keyword report to squint at and no obvious “disapproved” banner. It just spends and underperforms, and most people’s first instinct is to blame the algorithm. In almost every account I have audited, the real cause is one of five things below.

1. Conversion tracking is sending bad or incomplete data

Performance Max leans on Smart Bidding harder than any other campaign type, and Smart Bidding is only as good as the conversion data feeding it. Check these specifically:

• Go to Goals > Conversions and confirm the conversion actions your campaign is optimizing for are actually marked “Primary,” not “Secondary” or excluded from bidding.
• Check for duplicate conversions. A common issue is a purchase firing once through Google Ads’ own tag and again through a Google Analytics 4 import, which inflates conversion counts and confuses attribution, or the opposite: a tag that only fires intermittently, which starves the model of data.
• Look at conversion lag and whether conversions are being counted at all in the last 7 to 14 days, not just historically. A campaign can look “set up correctly” while the tag has silently stopped firing after a site update.

If you are not confident your conversion setup is clean, that is the first thing to fix before touching bids, budgets, or creative. No amount of asset group optimization will fix a campaign that is optimizing toward a broken signal.

2. The campaign is optimizing for the wrong conversion goal

Performance Max campaigns inherit their optimization goal from your account-level conversion goal settings unless you have set a campaign-specific goal. If your account’s primary conversion actions include something low-value like “contact page view” or “add to cart” alongside your actual purchase or lead-form conversion, the campaign can end up chasing the easy, cheap action instead of the one that matters to your business. Go into the campaign’s settings and confirm which conversion goals it is actually using, and remove anything that is not a genuine business outcome.

3. Your budget is too small for your target CPA

A frequently missed rule of thumb: if your daily budget cannot cover roughly 10 to 15 conversions’ worth of spend per week at your expected cost per conversion, Performance Max often cannot gather enough signal to optimize properly. When the budget is tight relative to CPA, the algorithm tends to fall back on cheaper, lower-intent inventory such as Display and YouTube placements to keep spending within budget, rather than the higher-intent Search and Shopping inventory that actually converts. This shows up in the account as spend concentrated in low-converting channels inside the asset group’s channel-level reporting.

4. Weak Ad Strength and missing audience signals

Ad Strength (rated Incomplete, Poor, Average, Good, or Excellent in the asset group) is a genuine performance signal, not just a nag from Google. Campaigns with “Excellent” Ad Strength see meaningfully more conversions on average than campaigns stuck at “Average” or below, because the system has more creative combinations to test and match to different audiences. To move the needle:

• Use all available headline (15) and description (5) slots with genuinely different messaging angles, not minor rewordings.
• Upload video in horizontal, square, and vertical orientations instead of letting Google auto-generate one from your images.
• Add at least 6 sitelinks with real, distinct value propositions.

Separately, your audience signals (customer lists, website visitors, custom segments based on search terms and competitor sites) are not hard targeting in Performance Max, but they give the algorithm a starting point during the learning phase. A campaign with no audience signals at all is starting from nothing, which slows down how quickly it finds converting traffic.

5. You judged it too early or changed too much too soon

Google’s own guidance is to allow 2 to 3 weeks after launch, or after any significant change to budget, bids, or assets, before drawing conclusions. Every time you edit an asset group, adjust the target CPA or ROAS, or change the budget meaningfully, you reset part of the learning process. If you have been making adjustments every few days because performance looked slow, that constant tweaking is very likely a bigger problem than whatever prompted the first change.

What to actually do, in order

1. Audit conversion tracking first. Confirm the right actions are primary, check for duplicates or gaps, and verify conversions are still recording in the last two weeks.
2. Confirm the campaign’s conversion goal settings match your real business outcome, not a proxy metric.
3. Compare your daily budget to your expected cost per conversion and raise it if it is structurally too low to gather signal.
4. Improve Ad Strength to Good or Excellent and add real audience signals.
5. Stop making changes and give it a clean 2 to 3 week window to actually optimize.

If you have already worked through all five and the campaign is still burning budget with nothing to show for it, that usually points to a structural issue, like Performance Max cannibalizing a Brand Search campaign or competing against your own Shopping feed, which is a different and more account-specific problem worth getting a second set of eyes on.

Performance Max rarely fails because the algorithm is bad. It fails because it was never given clean data, a clear goal, or enough budget to learn from in the first place.