Different totals in Google Ads and GA4 do not establish that either setup is broken. First identify the exact metric, report and conversion action on each side. A GA4 key-event total and a Google Ads conversion total are not automatically the same comparison.
In this guide
Identify the report before comparing the number
Google distinguishes Analytics events, key events and conversions. An event records an interaction; a key event marks an action as important to the business. A conversion can be created from an Analytics event for advertising measurement. Shared Google Ads conversions appear in Analytics' Advertising reports, rather than standard Analytics reports. Google's explanation of conversions and key events
This matters because the blanket explanation “Google Ads and GA4 always count differently” is too broad. Google supports shared conversion data across the two products. Availability and settings can depend on the property and conversion type. Check the actual conversion source and reporting view instead of relying on an old screenshot or terminology. Google's shared conversion setup documentation
Build a like-for-like comparison
Copy the following worksheet into your working notes. Record what is visible before changing anything. If a setting or source cannot be verified, mark it unknown.
| Check | Record for both reports |
|---|---|
| Report and metric | The exact report, selected column and conversion action or event |
| Business action | The real-world milestone the metric represents |
| Collection source | The named event source and linked account/property |
| Time basis | Date range, time zone and whether the view uses interaction time or conversion time |
| Credit and counting | Relevant attribution, eligible channels, conversion window and counting rules |
| Filters and freshness | Report filters, action selection, export time and processing status |
Google's discrepancy guidance identifies timing, counting and attribution among the reasons reports can differ. It recommends considering the “Conversions (by conv. time)” columns for time-aligned comparisons and allowing for processing delays. A conversion window and an attribution report's lookback window are distinct settings. Google Ads discrepancy troubleshooting
As an illustration, an ad interaction on one date and a submitted inquiry on a later date can fall into different reporting periods. Write both timestamps into the worksheet. This example explains the comparison; it does not establish the cause of a discrepancy in your account.
Separate event collection from campaign reporting
An event can be present in one report without appearing in the campaign column you selected. In Google Ads, primary actions are generally used in the Conversions column and bidding when the relevant standard goal is used. Secondary actions usually appear in All conversions. Custom goals are an important exception: their actions can be used for bidding even when marked secondary. Primary and secondary conversion actions
Do not switch an action to primary just to make two totals look closer. First establish the campaign's intended outcome, the action's definition and the effect of the proposed change.
Write a conclusion the evidence can support
A useful comparison ends with a specific explanation or a specific next check. “Tracking is wrong” is not enough to guide a repair.
- Different definitions: the reports measure different milestones or populations. Correct the comparison before diagnosing delivery.
- Timing or settings need reconciliation: record the exact difference and reassess using comparable views.
- A delivery problem is observed: preserve the relevant test or record evidence and identify the next component to inspect.
- The evidence is incomplete: list the inaccessible report, setting or stage. An unknown is not a pass or a confirmed fault.
Two matching totals are not proof that every underlying event is correct. Two different totals are not proof that a tag should be replaced. Keep the proposed change tied to the observed issue.
Write an illustrative comparison record
Consider a fictional journey in which a person interacts with an ad on October 1 and submits an accepted inquiry on October 3. A report organized by ad-interaction date and another organized by conversion date can place that same outcome in different daily rows. This example illustrates timing; it is not a claim that this explains your account.
Before counting the difference, name the conversion action and report on both sides. “Lead” in a dashboard might refer to one selected action, all actions in a goal, an Analytics key event or a business record. The same display label does not establish the same population. Record the actual selection beside the exported number.
| Observation | Next useful check |
|---|---|
| Different daily placement | Inspect interaction-time versus conversion-time reporting and time zones |
| Different selected milestones | Align the agreed business action before comparing counts |
| One source cannot be verified | Obtain the permitted configuration or mark the comparison incomplete |
| Comparable views still differ | Trace collection and delivery with a bounded evidence review |
A good comparison may end before any technical change is needed. If the reports answer different questions, explain which one supports the decision. If the question is accepted inquiries received by the business, inspect that business record as well. An advertising report and a sales ledger have different purposes.
Preserve a useful before and after record
Record the export time, filters and relevant settings with the comparison. Otherwise a later reviewer may unknowingly compare a fresh report with an earlier, less complete export. Keep the same definitions when reassessing the result, and document any justified change rather than silently rewriting the baseline.
If several people maintain the measurement setup, agree who owns the next check. A campaign manager can explain goal selection; a developer may need to inspect a form's success signal; an integration owner may hold delivery records. Sending everyone the same unexplained screenshot often creates parallel changes without a shared diagnosis.
For an implementation handoff, state what a proposed fix must demonstrate. An accepted action arriving once at the intended destination is a clearer acceptance condition than “make GA4 equal Ads.” Our guide to a useful diagnostic report provides a finding template for that decision.
Keep the investigation focused
If your report combines submissions, qualified leads and customers, start with clear lead-stage definitions. If a sale is recorded in the CRM but its downstream path is unclear, follow the CRM-to-ad-platform evidence worksheet.
Our US$195 conversion tracking diagnostic reviews one agreed existing journey, one selected ad platform/account, one GA4 property where relevant and up to three agreed events. You receive written evidence and prioritized next steps. Book a fit call to confirm scope; implementation is separate.