GA4 and Google Ads will almost always show different conversion counts. The mismatch is expected. The real question is whether the difference is structural and acceptable, or whether it signals a tracking problem that needs fixing.
Why do GA4 and Google Ads conversions not match?
GA4 and Google Ads use fundamentally different reporting logic. GA4 reports all conversions from all traffic sources, using data-driven attribution across the full user journey. Google Ads reports only the conversions it can attribute to its own clicks, using attribution rules that differ from GA4. They are answering different questions about the same conversion event.
Which platform should you trust for Google Ads optimization?
Use Google Ads data for optimization. Smart Bidding trains on Google Ads conversion data, so the algorithm will make decisions based on Google Ads reporting. Use GA4 for cross-channel analysis and understanding how Google Ads sits within your overall marketing mix, not for optimizing Google Ads bids.
What mismatch is normal and expected?
Differences in attribution model, lookback windows, and cross-device handling create structural gaps. A 10-30% difference is often normal. GA4 will almost always report more conversions than Google Ads because it credits assisted conversions from other channels. If the gap is consistent month-to-month, it is likely structural, not a tracking bug.
What mismatch indicates a real tracking problem?
Red flags: GA4 showing zero conversions while Google Ads shows activity (or vice versa). A sudden spike or drop in one platform without a corresponding change in the other. Revenue totals mismatching between GA4 and Google Ads ecommerce tracking when they should align. These indicate tag firing, conversion definition, or data import errors that need investigation.
How do attribution windows affect the comparison?
Google Ads and GA4 both have configurable lookback windows, but they default to different settings. Google Ads uses 30-day click and 30-day view attribution windows by default. GA4 settings vary by configuration. If these windows differ, GA4 and Google Ads will attribute conversions to different date ranges, making the comparison meaningless.
How should ecommerce accounts diagnose this differently from lead-gen?
Ecommerce should compare purchase event totals and revenue values. Lead-gen should compare form submission counts. Ecommerce: if GA4 purchase event revenue matches Google Ads revenue (roughly), tracking is likely consistent. Lead-gen: if GA4 form submission count is within 15-20% of Google Ads form submission conversion count, the setup is probably sound.
| Item | GA4 | Google Ads | Why It Matters |
|---|---|---|---|
| Default attribution | Data-driven (cross-channel) | Data-driven (paid only) | Google Ads takes more credit for assisted conversions |
| Lookback window | Configurable (default 30 days) | Configurable (default 30 days) | If set differently, numbers diverge more |
| Cross-device tracking | User-level stitching where possible | Google signed-in data | Both have gaps; neither is complete |
| Conversion definition | Events you configure in GA4 | Conversion actions you define | Mismatched events = different totals |
| Data freshness | Processing delay of hours | Usually same day | Short-term comparisons can mislead |
| What is counted | All traffic sources | Google Ads clicks only | GA4 total will always be higher |
| Revenue reporting | All sources combined | Google Ads attributed only | Never compare revenue totals directly |
Step 1: Check conversion definition consistency
- –Are the GA4 events and Google Ads conversion actions measuring the same user action?
- –Do the event names match between GA4 and the Google Ads linked import?
Step 2: Check lookback windows
- –Are Google Ads and GA4 using the same lookback window?
- –Is the comparison date range accounting for the lookback window correctly?
Step 3: Check for double-counting
- –Is GA4 data being imported into Google Ads AND a native tag firing?
- –Is the same purchase event firing twice under different names?
Step 4: Check the numbers that should agree
- –Compare conversion volume, not revenue, first (revenue attribution amplifies differences)
- –Use a date range at least 60 days wide to smooth daily noise