Why GA4 and Google Ads Never Agree on Numbers (and What to Do About It)
GA4 and Google Ads report different conversions for the same campaigns. Here are five structural reasons why and how to report the gap honestly to clients.
Updated 7 min read
You pull your Monday morning report. Google Ads says 247 conversions last week. GA4 says 183. Your client asks which one is right.
Neither is wrong. GA4 is a web analytics tool that tells you what users did on your property. Google Ads is an advertising platform that tells you whether ad spend led to business results. They were built by different teams with different measurement goals, and the discrepancies are predictable once you understand five structural differences.
1. Attribution models are fundamentally different
This is the single biggest source of divergence for most accounts.
GA4 defaults to a data-driven attribution model. Credit for a conversion gets distributed across multiple touchpoints in the user's journey. If someone clicked a search ad on Monday, came back through an organic listing on Wednesday, and converted on Thursday after clicking a display ad, all three touchpoints typically get partial credit.
Google Ads reports conversions attributed to the ad interaction. Even with data-driven attribution enabled in Google Ads, the model is scoped to ad interactions only. Organic visits, direct traffic, and email clicks do not exist in Google Ads' attribution universe.
When GA4 says your brand campaign drove 40 conversions and Google Ads says 78, part of the gap is that Google Ads gives the campaign full credit for conversions that GA4 splits across multiple channels.
What to check: Open the GA4 attribution model comparison report. Compare last-click to data-driven for your key conversion events. If that gap roughly matches the GA4-to-Google Ads gap, attribution modeling is your primary driver.
2. Conversion counting works differently
GA4 counts a conversion at the moment the event fires. Google Ads counts it at the time of the ad click and applies a lookback window up to 90 days (30 days is the default). A user who clicks your ad on August 5th and converts on September 1st appears in your August report in Google Ads but your September report in GA4.
This is the "conversions keep changing" problem. Google Ads data pulled on August 31st might show 200 conversions. The same data pulled on September 15th shows 218, as late converters trickle in through the lookback window.
Additionally, GA4 lets you count conversions as "once per session" or "once per event." Google Ads counts every conversion by default. A user who submits a form twice registers as one conversion in GA4 and two in Google Ads.
What to check: Filter your Google Ads report by conversion date instead of click date. Also check the counting method in Google Ads conversion settings.
3. Timezone mismatches shift daily totals
This is embarrassingly simple and catches people constantly.
Your GA4 property has a timezone setting. Your Google Ads account has a timezone setting. If they do not match, every metric grouped by date will disagree.
A GA4 property set to America/New_York (UTC-5) and a Google Ads account set to America/Los_Angeles (UTC-8) define "today" three hours apart. A conversion at 11 PM Eastern on August 31st falls on September 1st in Pacific time. GA4 records it in August. Google Ads records it in September.
Over a month, this bleeds 2 to 3 hours of conversions across the date boundary. For high-volume accounts, that is enough to produce noticeable daily and weekly discrepancies. Monthly totals usually wash out, but daily comparisons will always be off.
What to check: Go to GA4 Admin, then Property Settings, and note the timezone. Then go to Google Ads, then Account Settings, and note the timezone. If they do not match, that is your first fix. Note that you typically cannot change a Google Ads account timezone after creation.
4. Conversion windows do not align
GA4 and Google Ads each maintain separate lookback windows, and they default to different values.
Google Ads typically defaults to a 30-day click-through window and a 1-day view-through window. GA4 uses its own lookback, and the two systems do not share a conversion graph. A user who clicked an ad 45 days ago and converts today might appear in one system but not the other, depending on configuration.
The practical impact depends on your sales cycle. For e-commerce with same-session purchases, the windows rarely matter. For B2B lead generation with a 60-day consideration phase, the gap from window mismatches alone can reach 20 to 30 percent of total conversions.
For more detail on how click-through, view-through, and engaged-view windows work, see our guide to Google Ads attribution windows.
5. Consent Mode creates asymmetric data gaps
When a user declines cookies, GA4 and Google Ads each apply their own behavioral modeling to estimate the missing conversions. The two models produce different results because they estimate from different signal sets.
Ad blockers compound this by targeting GA4 and Google Ads requests at different rates. In EU markets with high consent-denial rates, the gap between GA4's modeled number and Google Ads' measured number can be significant.
| Factor | GA4 behavior | Google Ads behavior |
|---|---|---|
| Attribution model | Data-driven across all channels | Data-driven scoped to ad interactions only |
| Conversion date | Date the event fires | Date of the original ad click |
| Counting method | Configurable: once per session or every event | Default: every conversion occurrence |
| View-through conversions | Not counted by default | Counted with 1-day default window |
| Consent Mode modeling | Models at event level across all sources | Models at campaign level for ad conversions |
Debugging the discrepancy
When a client asks why the numbers do not match, work through this sequence: (1) verify the date range, conversion event name, and timezone match; (2) check whether Google Ads counts "one" or "every" conversion; (3) compare by conversion date rather than click date; (4) compare total GA4 sessions against total Google Ads clicks to gauge the consent and ad-blocker gap. After these checks, a 5 to 15 percent residual is typical and comes from cross-device modeling differences.
When to trust which number
Trust Google Ads for ad spend decisions. It is internally consistent with its bidding algorithms. Trust GA4 for cross-channel comparisons. It is the only system that sees all traffic sources in one place. Trust the trend for directional decisions. If both systems show conversions trending upward, the campaign is growing regardless of the exact count. Trust neither for absolute truth. Web analytics is estimation, not accounting. A 5 to 15 percent gap is normal; a 50 percent gap is a measurement problem worth fixing.
How to report the gap honestly
Present both numbers with context rather than reconciling them into one. A defensible position: "We report conversions from GA4 for cross-channel visibility and ROAS from Google Ads because it matches the bidding data."
Add a "source agreement" metric. When GA4 and Google Ads agree within 10 to 15 percent, tracking is healthy. A widening gap means something changed: broken tracking, shifted attribution settings, or consent rate changes. For context on how different attribution models shape these numbers, understanding each platform's model makes the conversation easier.