Google Ads reports one conversion total, Meta Ads another, and GA4 shows even fewer enquiries. This does not always mean the provider made a mistake or analytics is broken. The systems may count different actions, assign them to different dates and distribute credit among sources differently.
The business owner’s task is to understand the reasons for discrepancies and distinguish reporting characteristics from lost data. There is no need to force the totals to match: first define what you are comparing.
Start by defining the outcome
Establish what the metric represents in each report:
- an ad click or a visit to the website;
- a GA4 session;
- a click on a contact link;
- a confirmed enquiry submission;
- a qualified lead;
- a sale or payment received.
A click is not a session, and an enquiry is not a customer. A Meta instant form can also collect a contact without a website visit. If the form outcome is not sent to GA4 separately, the absence of a corresponding website event is expected.
For a service website, agree beforehand on events from enquiry through qualification. The comparison then starts with the same business meaning.
Align dates and time zones
In standard Google Ads reporting, a conversion is generally assigned to the date of the advertising interaction, while GA4 assigns it to the date of the event itself. Google Ads provides specific columns for comparisons based on the actual action time. These differences are described in Google’s conversion discrepancy guidance and data reconciliation instructions.
Hypothetical example: someone clicks an ad on 30 September and submits an enquiry on 2 October. In a Google Ads report based on click date, the outcome may appear in September; in GA4, based on event date, it appears in October. Comparing October totals alone produces a difference without losing the enquiry.
Also check the time zones of advertising accounts, the GA4 property and the CRM export. An enquiry around midnight can fall on different calendar days.
Check attribution and the reporting window
Attribution determines how a system distributes credit for an outcome across interactions. The attribution window limits which preceding interactions can take part in that calculation.
Record the model, window and interaction types included in each report. Check separately whether the selected advertising metric includes outcomes after a view without a click. Do not assume one universal Meta window for every account: use the settings of the specific campaign and export. Meta’s official SDK provides separate parameters for attribution windows and the time used to report an action. Meta reporting parameters.
Choosing the right analysis scope matters in GA4 too. The source of a user’s first acquisition, the session source and the source in an attribution report answer different questions. Changing the attribution model does not change all these dimensions in the same way. GA4 attribution settings.
Do not add sales that Google and Meta attribute to their advertising together as unique customers. The same deal can appear in both platforms’ assessments. Counting actual deals requires a separate record with duplicates removed.
Align counting rules
Check whether all repeated actions are counted or just one after an interaction. In Google Ads, the “One” and “Every” settings, as well as “Conversions” and “All conversions” columns, produce different datasets.
Check the goal set: does it include a phone link click, a form submission and the qualification of that same enquiry simultaneously? Their sum represents stages, not new customers.
Align monetary metrics too. Contract value, the amount paid and revenue adjusted for refunds are not interchangeable. The currency and what the amount includes must be clear throughout.
When to investigate a technical error
A sudden change in the discrepancy after a website update, widespread loss of source information or repeated counting of one action warrants investigation.
Check:
- whether the correct tag is installed on every landing page;
- whether the event fires after an actual successful action;
- whether two independent configurations send the same event;
- whether parameters survive navigation and redirects;
- whether the property or advertising destination identifier is correct;
- whether consent settings and data collection conditions have changed;
- whether CRM outcome transmission has errors.
Google separately explains how page loading, blocked measurement and lost parameters can cause differences between clicks and sessions. Not every advertising click has to become a measured session. Clicks and sessions discrepancy guidance.
A monthly reconciliation process
- Choose one action, such as a confirmed enquiry.
- Record the period, time zone and date basis of the report.
- Retain the model, windows, counting rules and goal set.
- Allow data processing to finish and account for the delay before a deal closes.
- Trace several control enquiries through the entire chain.
- Separate explained differences from errors and assign owners to fixes.
- Compare subsequent periods using the same method.
To assess the economics, use the differences between ROAS, CPL and customer acquisition cost. Advertising accounts support campaign management, GA4 analyses behaviour and sources within its measurement capabilities, and CRM and payment data verify actual enquiries and sales.
When arranging advertising setup and management, agree on the reconciliation method and metric definitions. Discuss reporting scope using our commercial proposal.