How to Read Your Tracking Mismatch Report

Total conversions as reported by each platform over the same 30-day window. These should be close. A large gap means either your Ads tags are over-counting (view-through, cross-device inflation) or your GA4 setup is under-counting (attribution model differences, consent mode filtering).

The simple subtraction: Ads conversions minus GA4 conversions. Positive = Ads over-reports. Negative = GA4 over-reports. This number represents phantom conversions the algorithm uses to make bid decisions that have no GA4 confirmation.

Average of the absolute attribution drift percentage across all campaigns, pulled from the signal_health_intelligence layer. This normalizes the gap so a campaign with 1,000 conversions and 50 drift is comparable to one with 10 conversions and 5 drift.

When more campaigns have tracking problems than don't, the root cause is typically at the container level — a GTM misconfiguration, consent mode blocking conversions in GA4, or mismatched attribution models between platforms.

Each day the gap widens, the algorithm's bid model drifts further from reality. Yesterday's bids were based on slightly wrong data. Today's bids are based on even worse data. The error compounds.

The mismatch exists but isn't worsening. This may be an acceptable baseline difference between platform attribution models rather than a tracking bug.

Ads reports nearly double what GA4 confirms. The algorithm is aggressively bidding on this campaign because it believes conversion volume is much higher than reality. Every CPC is inflated by the phantom signal.

Your primary revenue-driving action has a significant modeled gap. This means the algorithm's tROAS or tCPA calculations are based partly on estimated conversions. Bid amounts are inflated because the algorithm thinks more purchases happened than can be verified.

Search campaigns rarely generate meaningful view-through volume because users either click or don't. High VT on Search suggests a tag configuration issue — likely cross-campaign attribution where a Display impression gets credited to a Search campaign.

If cross-device volume accounts for most of the Ads-GA4 gap, the mismatch is a known platform difference rather than a tracking bug. Both platforms are measuring correctly — they just attribute to different touchpoints.

Ads is capturing conversions over a longer window than GA4. Common when Ads uses a 30-day click window but GA4 uses last-click attribution that doesn't look back as far.

The mismatch has gotten worse recently. Something changed in the last 7 days — a new tag, a consent mode update, or a GA4 property modification.

The algorithm believes this campaign converts twice as often as it actually does. It bids aggressively to win auctions for an audience that isn't converting at the rate it expects. CPCs are inflated, budget is mis-allocated, and the damage compounds daily.

GA4 records fewer conversions than Ads. This could be consent mode filtering, attribution model differences, or GA4 tag firing issues. Less damaging than over-counting, but still causes the algorithm to under-bid on a campaign that may be performing better than GA4 suggests.

More than half of this action's conversions are estimated. The tag is either not firing on the actual conversion event, or it fires alongside a duplicate tag that the system then models around.

This campaign receives significant budget AND most of its signal is phantom. The algorithm is confidently spending large amounts based on data it cannot verify. This is the highest-damage combination in the audit.

The algorithm sees almost no conversions relative to click volume. It cannot distinguish between good and bad traffic at this signal level. Smart Bidding on this campaign is equivalent to random bidding with extra steps.

This campaign has enough conversions relative to clicks for the algorithm to make statistically reliable bid decisions. Tracking fixes here have immediate positive impact on bidding accuracy.

A sudden drift increase almost always traces to a specific event: a tag was added, a consent banner changed, or a GA4 configuration was modified. Check the exact week against your change log.

A sustained decline in signal health means the root cause hasn't been addressed. Incremental tag fixes aren't working — the problem is at the architecture level (consent mode, attribution model, container structure).

The average drift looks manageable, but the tag fires inconsistently — sometimes correctly, sometimes not. This is often caused by race conditions in tag loading, consent mode that applies inconsistently, or A/B test variants that break tag firing.

Each week, a larger proportion of your conversion data is estimated rather than observed. This gradual erosion can go unnoticed because total conversion volume stays stable — but the quality of each data point is degrading.

These campaigns are actively corrupting the algorithm's bid model. Every day they run with Smart Bidding, they waste budget on phantom signal. Fix these before touching any other campaign.

Clean signal, sufficient volume, minimal drift. The algorithm has reliable data to make good bid decisions. Keep automated bidding active.

Signal quality is too low for automated bidding to work. The algorithm is making decisions on data that is fundamentally unreliable. Manual bidding eliminates the algorithm's ability to compound errors.

This action has all_conversions but zero click conversions. It contributes nothing but noise to the bidding algorithm — it inflates conversion counts without any verifiable user action.

This page receives significant traffic but records zero conversions in GA4. Either the conversion tag isn't installed on this page, or it's firing but not transmitting data due to consent mode or a JavaScript error.

A conversion rate above 20–30% on a non-checkout page usually means the tag fires on page load or an intermediate event rather than the actual conversion. This inflates conversion counts for every visitor who loads the page.

Long-consideration purchases (expensive products, B2B) naturally create multi-day lag. Short-window comparisons between Ads and GA4 will always show drift for these actions. This is not a problem to fix — it's a reality to account for.

The percentage of total budget flowing to campaigns where >30% of conversions are algorithmically estimated. This is budget the algorithm allocates based on data it cannot verify — the purest form of tracking risk.

When more than 30% of your budget is exposed to tracking risk, the entire account's bidding model is compromised. Fixes in one campaign can't compensate for noise in dozens of others.