Why GA4 and Google Ads Conversions Don't Match (and Which Number to Trust)

Why don't GA4 and Google Ads conversions match?

GA4 and Google Ads measure the same actions using different rules. Google Ads attributes a conversion to the date of the ad click and only credits its own traffic. GA4 records the key event on the day it happened, across every channel. Different date logic, attribution scope, and counting methods mean the totals will never be identical, even when both tags are working perfectly. A gap of roughly 10-30% is normal. A sudden gap of 40% or more, or one side near zero, is a tracking problem worth investigating immediately.

Every few weeks in the accounts we manage, a store owner spots the discrepancy and rings to ask whether their Google Ads reporting is inflated, or whether GA4 is losing purchases. The answer is almost always: neither. The two systems were built for different jobs, and comparing them as if they should agree is like comparing your gross sales to your accountant's net revenue figure and asking why they don't match. They are not supposed to. This post explains the six structural reasons for the difference, covers the specific cases where it does signal a real problem (enhanced conversions double-counting purchases, Google Ads revenue running higher than Shopify), and tells you which number belongs in each decision.

Six Structural Reasons the Numbers Differ

1. Click date versus event date

Google Ads, by default, attributes a conversion to the date of the click that preceded it, not the date the purchase happened. If someone clicks your Shopping ad on 28 October and buys on 3 November, Google Ads logs that sale in October. GA4 logs it in November, on the day the purchase event fired. Compare the two over a calendar month and the boundary effect alone can produce a visible gap, particularly around month-end or at peak periods like Black Friday where the click-to-purchase window stretches out.

You can reduce this effect in Google Ads by switching to Conversions (by conversion time) in your columns, rather than the default click-date view. Google's own attribution documentation explains both column types. This does not change how Smart Bidding reads data, only how you inspect it in the interface.

2. Attribution scope: Google Ads only counts its own clicks

Google Ads only claims credit for conversions it can link to a Google Ads click (or, in some cases, a view-through interaction). GA4 sees every channel: organic search, direct, email, Meta, comparison shopping sites, and everything else. A customer who clicked a Google Shopping ad four weeks ago, received an email yesterday, and purchased this morning might appear in Google Ads (because the click is within the conversion window) and in GA4 attributed to email. Same revenue, different source in each report.

If you filter GA4's acquisition reports to google / cpc only, the gap narrows considerably. It does not disappear, because attribution models still diverge, but you are at least comparing the same slice of traffic.

3. Counting rules

In Google Ads, each conversion action is set to count Every or One conversion per ad interaction. Purchases are typically set to Every, meaning two orders from the same click count twice. GA4 key events, depending on how they are configured, may count once per session or once per event. If the counting rules differ, the totals will differ even on identical orders.

Check this in Google Ads under Tools > Conversions > the purchase action > Count. If it says Every and your GA4 purchase event is session-scoped, that alone can explain part of the gap for repeat buyers.

4. Conversion windows

Google Ads click-through conversion windows default to 30 days for most campaign types, but they can be set as long as 90 days, and many accounts have never had this reviewed. GA4 session-scoped key events typically look back 30 days from the conversion. A 90-day Google Ads window claims sales that happened weeks after the last click, sales that GA4 and Shopify both attribute to the channel the customer came back through later. This is one of the most common reasons Google Ads revenue runs materially higher than Shopify. We covered this in detail in our post on Google Ads conversion windows and inflated ROAS.

5. Modelled data and Consent Mode

Since the rollout of Consent Mode v2, both platforms fill in gaps for users who decline cookies, but they do it differently. Google Ads uses click IDs, enhanced conversions, and its own modelling to recover attributed sales. GA4 models at the property level using a different methodology. The two models do not produce the same output, and neither is wrong: they are estimates from different vantage points. Accounts with a high proportion of UK traffic, where ICO-compliant consent banners produce meaningful opt-out rates, can see the modelling diverge by more than accounts with predominantly opted-in audiences. This gap has widened since the second half of 2024 as Consent Mode adoption became standard.

6. Processing lag and time zones

GA4 can take 24-48 hours to fully process data; imported key events in Google Ads can lag further still. If your Google Ads account and GA4 property are in different time zones (a common oversight on accounts set up quickly), day boundaries shift, and totals across any period that includes that boundary will not reconcile.

Enhanced Conversions Double-Counting Purchases

This is the one case where a large discrepancy is entirely a tracking problem, not a structural difference. It happens when an account has two active primary conversion actions recording the same purchase: typically a native Google Ads purchase tag (with enhanced conversions enabled) and a GA4-imported purchase event, both set to primary. Every order is counted twice in the Google Ads conversion column.

In the accounts we manage, we audit for this during onboarding because it is more common than most agencies admit. The symptom is a conversion volume that is suspiciously high relative to Shopify orders, often double. The fix is to designate one source as primary. Our preference is the native Google Ads purchase tag with enhanced conversions enabled, because it has lower latency and feeds Smart Bidding more accurately than an imported GA4 event. The GA4 purchase event stays active for analysis, but set to secondary in Google Ads so it does not count towards the bidding signal. See the Google Ads conversion tracking setup guide for the full configuration.

Why Google Ads Revenue Runs Higher Than Shopify

When a store owner sees Google Ads claiming, for example, £28,000 in revenue over a month and Shopify reporting £19,000 in total orders, the most common explanations are:

• A long conversion window. A 90-day click-through window claims sales from customers who last clicked a Google ad months ago and returned through a completely different channel. The revenue is real; the attribution is generous. Reduce the window to 30 days (or 7 days for purchases with a short consideration period) to bring the figures closer to reality.
• Double-counting from two primary purchase tags, as described above.
• Cross-device attribution. A customer clicks on mobile, does not convert, then purchases on desktop. Google Ads stitches these together using signed-in Google account data. Shopify records one order, from the desktop session, with no memory of the mobile click. Both are accurate; they are reporting on different things.
• Cancelled or refunded orders. Google Ads does not automatically deduct order cancellations unless your tracking implementation sends a negative conversion or you use the Google Ads order ID deduplication feature. Shopify removes cancelled orders from its revenue total. Over a period with meaningful return rates, this adds up.

Which Number Should You Trust for Which Decision?

The rule we follow in every account we manage is straightforward: use Google Ads data for Google Ads decisions; use GA4 for channel-mix and site behaviour decisions.

Smart Bidding (target ROAS, target CPA, maximise conversion value) reads from the Google Ads conversion column, not from GA4. If you lower your tROAS target because GA4 is showing fewer conversions than Google Ads, you are adjusting the strategy based on a number the strategy does not use. You may starve campaigns that are performing correctly.

GA4 is the right tool for understanding what happens after the click: bounce rates, multi-step checkout drop-off, which product categories convert for non-Google traffic, and whether a channel mix shift is happening. It is also the right cross-check for sense-checking Google Ads figures: if Google Ads claims 200 purchases and Shopify shows 60 orders, something is clearly wrong. But a 15-20% difference in the same direction, consistently, is not a problem.

The practical diagnostic check: pull both reports for the same 30-day period, filter GA4 to google / cpc only, switch Google Ads to conversion-time columns, and compare. If the gap is below 25% in the same direction each month, your tracking is behaving normally. If it flips direction month to month, or the gap has grown since a specific date, start investigating from that date: a tag was deployed, a setting was changed, or a Shopify app was updated. For stores on Shopify, the August 2026 removal of thank-you page additional scripts is a known trigger point; see our post on fixing conversion tracking on Shopify if that date matches your gap.

A Quick Pre-Audit Checklist

Before changing any tROAS target or budget because the two dashboards disagree, run through these five checks:

1. Are there two primary purchase conversion actions in Google Ads? (Ads account: Tools > Conversions. Look for duplicate purchase events both marked primary.)
2. What is the click-through conversion window on the purchase action? (30 days is appropriate for most e-commerce; 90 days inflates reported revenue.)
3. Do the Google Ads account and GA4 property share the same time zone?
4. Are cancelled or refunded orders being deducted? (Check whether order IDs are passed and whether a cancellation signal is sent.)
5. Has anything changed in the tag setup in the last 60-90 days? (GTM container versions, Shopify app changes, a new developer touching the thank-you page.)

If you pass all five and the gap is still larger than 30%, a full conversion tracking audit is the right next step before touching any campaign settings.

Frequently Asked Questions

Why don't GA4 and Google Ads conversions match?

GA4 and Google Ads measure the same actions using different rules. Google Ads attributes conversions to the date of the ad click and only credits its own traffic. GA4 records the key event on the day it happened across all channels. Different date logic, attribution scope, and counting methods mean the totals will never be identical, even when both tags are working perfectly.

Can enhanced conversions cause double-counting of purchases?

Yes, if your Google Ads account is set to import GA4 purchase events AND you have a separate Google Ads purchase tag firing on the same thank-you page, both will record the same order. The fix is to use one source only: either the native Google Ads tag with enhanced conversions enabled, or GA4 import, not both marked as primary conversions.

Why is Google Ads revenue higher than Shopify?

The most common causes are a long click-through conversion window (often set to 90 days) claiming sales that email, organic, or direct traffic actually drove; double-counting from two active purchase tags; or cross-device attribution where Google stitches a mobile click to a desktop purchase that Shopify records independently. Check your conversion window setting first.

Which number should I use for Google Ads Smart Bidding?

Use the Google Ads conversion column for Smart Bidding decisions. Target ROAS and target CPA bid strategies read from Google Ads conversion data, not GA4. Use GA4 to understand channel mix, site behaviour, and whether the conversion journey makes sense, but do not adjust tROAS because the two dashboards show different totals.

What is a normal gap between GA4 and Google Ads conversions?

A difference of roughly 10-30% is typical when both are set up correctly, because of the structural reasons above. A gap that appears suddenly, exceeds 40%, or shows one side near zero while the other is normal almost always points to a tracking problem worth investigating immediately.