In 2026, Google Ads and GA4 will almost never report the same conversion total for the same campaign, and a 5 to 15 percent gap between the two is normal rather than a sign that something is broken. The instinct is to assume one platform is wrong and chase the numbers until they match. They will not match, because the two tools were built to answer different questions — and once you understand why, the gap becomes information instead of an alarm.
This guide explains the six structural reasons GA4 and Google Ads diverge — attribution, windows, timing, consent, modeling and deduplication — then reconciles them into one trusted figure you can defend to leadership. To check whether your tracking and import setup is feeding both platforms cleanly, run our free 5-axis Google Ads audit.
Updated 2026-05-12 with current Consent Mode v2 modeling, GA4 attribution settings and Google Ads import behavior observed across US, UK and European accounts.
- A 5 to 15 percent gap is normal — both tools work, they just count differently, so stop forcing them equal. 2. Attribution models and windows are the single biggest cause — match them before you compare. 3. Click date vs event date — Google Ads pulls conversions to the click, GA4 to the event. 4. Consent and modeling fill gaps differently per platform. 5. Pick a source of truth by decision — bidding follows Google Ads, reporting follows GA4, and you document the variance.
Why do GA4 and Google Ads count conversions differently?
The first thing to accept is that both tools are working correctly even when their numbers disagree. They are not two measurements of the same thing; they are two different questions that happen to use the same word, conversion. Google Ads asks how many conversions its ads drove. GA4 asks how many conversions happened on the site or app across every channel. Those questions produce different answers by design.
Different scope — Google Ads only counts conversions it can attribute to an ad interaction. GA4 counts every conversion event regardless of source, then assigns credit through its own channel grouping. A conversion from organic search exists in GA4 and is invisible in Google Ads, so the totals start from different universes before any model is applied.
Different counting rules — Google Ads lets you choose to count every conversion or only one per click, and applies its own deduplication. GA4 counts key events on its own logic. The same purchase can be one conversion in one tool and counted differently in the other. If you want the mechanics of how a single action gets inflated, our double-counting fix guide walks the exact traps.
Different defaults — Out of the box the two platforms use different attribution models, different windows and different timing rules. Nothing is misconfigured; the defaults simply do not agree. Understanding that the gap is structural, not accidental, is the foundation for everything that follows.
How do attribution models and windows diverge?
Attribution is the single largest source of disagreement between the two platforms. The model decides which touchpoint gets credit, and the window decides how long after the click a conversion still counts. When either differs, the totals drift apart even though every conversion is real.
Attribution model — Google Ads has moved to data-driven attribution as its default, which spreads credit across touchpoints. GA4 reporting can use a different model depending on your settings. If one platform gives full credit to the last click and the other splits credit across the path, the same conversions land in different campaigns and different totals. Our data-driven vs last-click explainer shows how each model reassigns the same conversions.
Conversion window — Google Ads defaults to a 30-day click window. GA4 uses its own lookback that you configure separately, and it can be shorter or longer. A conversion that lands 28 days after the click counts in a 30-day window and vanishes in a shorter one. A window mismatch alone can produce a double-digit gap that looks like a tracking fault but is pure configuration.
View-through and cross-device — Google Ads credits engaged-view and cross-device paths that GA4 may attribute elsewhere. These paths inflate Google Ads relative to GA4 for the same period. Before you compare a single number, confirm that the model and the window are aligned on both sides, or you are comparing two different definitions of the word conversion.
How does conversion timing (click date vs event date) differ?
Even with identical models and windows, the two platforms place the same conversion on different days. This timing difference is subtle, it is invisible in a monthly total that has fully matured, and it wrecks any comparison of the live week.
Click-date reporting — Google Ads attributes a conversion back to the date of the click that earned it, not the date the conversion happened. A sale today from a click 20 days ago is credited to that earlier day. This is correct for bidding, because the algorithm needs to learn which click drove the outcome, but it means recent days keep filling in as late conversions arrive.
Event-date reporting — GA4 records the conversion on the date the event actually fired. The same sale lands today, not 20 days ago. So for any window that is still open, Google Ads and GA4 are literally describing different days, and a day-by-day comparison will never line up.
The practical rule — Compare a closed, mature period, not the current week. Wait until late conversions have landed on both sides before you read the totals. If you compare a live seven-day window you will see a large gap that closes on its own as Google Ads backfills the click dates. Judging the platforms on fresh data is the most common self-inflicted reconciliation error.
How do consent and modeling affect each platform?
Privacy changes mean a growing share of conversions are not directly observed but modeled, and each platform models the gap on its own logic. This is a legitimate part of the difference, not missing data, and treating modeled conversions as errors leads people to under-report real performance.
Consent Mode v2 — When a user declines tracking, the platforms receive consent signals rather than full event data. Google Ads uses these signals to model the conversions it could not directly measure. GA4 applies its own behavioral modeling. The two fill the same gap with different methods, so the modeled portion rarely matches exactly.
Where the gap comes from — In a market with high consent-banner rejection, the modeled share can be substantial, and that is precisely where Google Ads and GA4 diverge most. A region with strict privacy enforcement will show a wider gap than a region with high opt-in rates, on the same campaign, purely because more of the total is modeled.
Why you keep modeling on — Turning modeling off does not make the numbers truer; it makes them lower and biased toward consenting users. If you are seeing far fewer conversions than expected after a consent change, the cause may be a tracking break rather than modeling — our zero-conversions fix guide separates a real measurement failure from an expected modeled gap.
How do you reconcile and pick a source of truth?
Reconciliation does not mean forcing the two numbers to be equal. It means aligning what you can, explaining what you cannot, and assigning each platform to the decision it should own. The mistake is using one number for every purpose; the fix is choosing the right number for each job.
Align the comparables — Use the identical date range, the same single conversion definition, and matching attribution models and windows. Most apparent gaps shrink the moment you stop comparing all GA4 key events against one Google Ads action over mismatched periods. Setting up the import correctly is half the battle — our GA4 setup and conversion import guide covers the configuration that keeps both sides comparable.
Source of truth for bidding — Smart Bidding optimizes toward the exact conversion signal in Google Ads, including its modeled and cross-device conversions. For any bidding or ad-optimization decision, Google Ads is authoritative. Feeding a GA4 number into a bidding decision starves the algorithm of the signal it was trained on.
Source of truth for reporting — For cross-channel performance, blended ROI and full-funnel analysis, GA4 is the better lens because it sees every channel, not just paid. Business reporting to leadership should anchor to GA4 while clearly noting that ad-platform totals will differ. Assign the decision first, then the number follows.
The GA4-vs-Google-Ads reconciliation table
Work this table top to bottom — it pairs each symptom you see with the structural cause and the reconciliation step, ordered from the cheapest alignment to the deepest.
Forcing the two platforms to report the identical number is the wrong goal and usually means breaking something to hit it — disabling modeling, narrowing a window, or removing a legitimate channel. A 5 to 15 percent residual gap is healthy and expected because the tools answer different questions. Align the comparables, explain the rest, and document the accepted variance. The moment you chase a perfect match you start trading accurate measurement for a tidy spreadsheet.
How do you document an acceptable variance range?
The final step turns a one-off reconciliation into a repeatable policy. If every report restarts the argument about which platform is right, you have not reconciled anything. Write the variance down once and let it govern future months.
Set your baseline — Compare three to six closed months with aligned date ranges, definitions and windows, and record the typical gap as a percentage band. Most healthy accounts land in a 5 to 15 percent band. That band, not zero, becomes your definition of normal for this account.
Name the authority per report — State plainly that bidding and ad-optimization numbers come from Google Ads and cross-channel reporting comes from GA4. A one-line note on each dashboard prevents the monthly debate and stops anyone from quietly reconciling to the wrong source.
Flag drift, not difference — Once the band is set, you only investigate a month that drifts outside it. A move from a 10 percent gap to a 30 percent gap is a signal worth chasing — usually a tag, a window, or an import problem. A move from 10 to 12 percent is noise. To pressure-test that your tracking is firing correctly on the destination URL before you blame attribution, validate the page first with our free 5-axis audit and confirm the tags resolve with the GA4 URL tester.
Sources
Official sources consulted for this guide:
-
support.google.com — attribution in GA4
-
support.google.com — about conversion attribution
-
support.google.com — GA4 channel groups
-
support.google.com — about conversion tracking
FAQ
Why don't GA4 and Google Ads conversions match?
They never match exactly because they were built to answer different questions, and at least six structural differences pull the numbers apart. Google Ads counts ad-driven conversions on the click date with its own attribution window; GA4 counts events on the event date across every channel with its own model. Add deduplication rules, distinct conversion windows, Consent Mode modeling that differs per platform, and import lag, and a gap is the default state, not a bug. A 5 to 15 percent variance between the two is normal and expected. The job is not to force them equal but to understand and document the gap.
Which is more accurate, GA4 or Google Ads?
Neither is more accurate in absolute terms; each is more correct for a specific job. For Smart Bidding and ad optimization, Google Ads conversions are the right source because the algorithm bids on that exact signal, including its modeled and cross-device conversions. For full-funnel and cross-channel reporting, GA4 is the better lens because it sees organic, direct, email and social alongside paid. Asking which is right in general is the wrong question. Pick the platform that owns the decision: bidding follows Google Ads, business reporting follows GA4, and you reconcile the two on purpose.
How big a discrepancy between GA4 and Google Ads is normal?
For most accounts a 5 to 15 percent gap is normal and not worth chasing. Under 5 percent is excellent alignment; 15 to 25 percent is worth investigating but often explainable by attribution and timing; above 25 percent usually points to a real problem like a missing tag, double counting, or a broken import. Set your own baseline by comparing the same date range, the same conversion definition, and the same attribution window on both sides. Once you know your account's typical gap, you can flag any month that drifts outside it instead of panicking at every difference.
Does the attribution window cause GA4 and Google Ads to differ?
Yes, the attribution window is one of the largest single causes of mismatch. Google Ads defaults to a 30-day click window for conversions, while GA4 reporting uses its own lookback that you can configure separately. If one platform credits a conversion that happened 28 days after the click and the other has already closed its window, the totals diverge for that period. Click-date versus event-date reporting compounds the effect. Always confirm both windows match before you compare numbers, because a window mismatch alone can create a double-digit gap that looks like a tracking fault.
Should I import GA4 conversions into Google Ads?
Import them only with a clear plan to avoid double counting, because the most common mistake is running a native Google Ads tag and a GA4 import for the same action at once. If both fire, one conversion is counted twice and Smart Bidding over-credits that path. Pick one primary source for each conversion action: either the native Google Ads tag or the GA4 import, not both. If you import GA4 key events, mark the duplicate native action as secondary so it reports but does not feed bidding. Our import and double-counting guides walk the exact setup so the signal stays clean.
Why does Google Ads show more conversions than GA4?
Google Ads usually shows more conversions than GA4 for three reasons. First, modeling: Google Ads fills consent and cross-device gaps with modeled conversions that GA4 reporting may not surface the same way. Second, attribution scope: Google Ads credits view-through and cross-device paths that GA4's channel grouping can assign elsewhere. Third, timing: click-date reporting pulls conversions back to the day of the click, inflating recent ad periods relative to GA4's event-date view. The reverse can also happen when GA4 captures conversions from channels Google Ads never sees. Match windows and definitions before concluding either platform is wrong.
How do I reconcile GA4 and Google Ads conversions for a report?
Reconcile by aligning four things, then documenting the residual gap. First, use the identical date range on both platforms. Second, compare the same conversion definition, not all GA4 key events against one Google Ads action. Third, set matching attribution models and windows where the tools allow it. Fourth, account for click-date versus event-date by checking a closed, mature period rather than the live week. After alignment, a single-digit residual gap is expected and should be stated in the report as your accepted variance, with bidding decisions anchored to Google Ads and channel reporting anchored to GA4.