Your Google Ads account says 47 conversions last month. GA4 says 31. Your CRM says 22, and only 9 of those were worth a sales call.
Every marketing director running paid search hits this. The instinct is to assume one platform is lying, pick the lower number because it feels safer, and quietly lose confidence in all of it.
That instinct is wrong, and it is expensive. Google Ads and GA4 are not two attempts to measure the same thing. They are two systems built for different jobs, using different rules, and a gap between them is the normal state. The question is never "which number is right." The question is whether the size of the gap is explainable.
Here is how to tell.
1. The two platforms date conversions differently
This is the single largest cause, and it accounts for most gaps that show up mid-month.
Google Ads reports a conversion on the date of the click that led to it. GA4 reports it on the date the conversion happened.
Someone clicks your ad on March 3, thinks about it for a week, comes back and fills in your form on March 10. Google Ads adds that conversion to March 3. GA4 adds it to March 10.
Compare a single week and the numbers will never agree. Compare a full quarter and most of this difference disappears. If your sales cycle runs longer than a few days, this alone can move a monthly number by 20% or more.
How to check it: widen the date range. If the gap shrinks as the window grows, you are looking at date attribution, not a broken tag.
2. They credit different channels
Google Ads reports conversions it can tie to an ad interaction. That is its entire purpose. It is not trying to give a neutral account of how the conversion happened.
GA4 runs its own attribution across every channel. If a user clicked your ad, then came back later through organic search and converted, GA4's model may hand that conversion to organic while Google Ads still counts it.
Neither is wrong. Google Ads answers "did this ad contribute?" GA4 answers "which channel gets the credit?" Those are different questions and they produce different totals.
3. Your conversion counting setting
In Google Ads, each conversion action is set to count either every conversion or one per click.
Set to "every," one person who submits your contact form three times becomes three conversions. GA4 counts every event too, but the click-based grouping is different, so the inflation lands differently in each platform.
Lead generation accounts should almost always use "one." If nobody has reviewed that setting since the account was built, check it now. It is one dropdown per conversion action and it silently inflates your reported performance.
4. Double counting from two sources of truth
This is the one that does real damage, because it makes a campaign look profitable when it is not.
If you have both a native Google Ads conversion tag firing on your thank-you page and a GA4 key event imported into Google Ads as a conversion, the same lead gets counted twice in Google Ads. Smart Bidding then optimizes toward a number that does not exist.
Auditing accounts, this is the most common serious defect we find. It usually happens because a new agency or contractor set up their own tracking without removing what came before.
How to check it: in Google Ads, open Goals, then Conversions, then Summary. Look for two conversion actions that describe the same event, one with source "Website" and one with source "Google Analytics 4." Check the "Include in conversions" column. If both are on, you are double counting.
Finding duplicates like this is exactly what the Conversion Tracking Audit is built for.
5. Consent mode and modeled conversions
When a visitor declines cookies, neither platform can observe the conversion directly. Both fill the gap with modeling, and they model differently.
Google Ads applies behavioral modeling to recover conversions it can attribute to ad clicks. GA4 applies its own modeling with different thresholds. Two different estimates of the same invisible users will not produce the same total.
If you operate in the EU or the UK, or you run a cookie banner anywhere, expect a structural gap here that no amount of tag fixing will close.
6. Conversion windows don't line up
Google Ads defaults to a 30-day click-through conversion window and lets you set anything from 1 to 90 days. GA4 has its own lookback windows, defaulting to 30 days for acquisition events and 90 days for everything else.
If your Google Ads window is 30 days and a customer converts on day 45, Google Ads never counts it. GA4 might. A long consideration cycle turns this into a permanent gap.
7. Time zone mismatch
Your Google Ads account has a time zone, set at creation and effectively permanent. Your GA4 property has its own, and it can be changed.
If one is set to Pacific and the other to Eastern, every daily number is shifted by three hours of conversions. On a low-volume account, that is enough to make two correct systems look broken.
How to check it: Google Ads account settings, then GA4 property settings. They should match.
8. Auto-tagging and lost click IDs
Google Ads attributes conversions using the GCLID appended to your landing page URL. Break that and attribution collapses.
Three common ways it breaks:
- Auto-tagging is switched off in the Google Ads account
- A redirect between the ad and the landing page strips query parameters
- The form or booking tool lives on a different domain and the parameter is not passed through
Calendly embeds, third-party booking widgets and payment processors are the usual culprits. The click still happened. Google Ads just can no longer prove it led anywhere.
9. Filters and thresholds in GA4
GA4 quietly removes data that Google Ads keeps. Internal traffic filters, developer traffic exclusions, and bot filtering all reduce GA4's count. On top of that, GA4 applies thresholding to reports when Google signals is enabled and the sample is small, withholding rows that could identify individuals.
None of this is visible in the interface unless you go looking for it.
So what gap is acceptable?
There is no published benchmark, and anyone who quotes you one is guessing. What matters is whether the gap is explained and stable.
Work through it in this order:
- Compare a 90-day window, not a week. Most date-attribution noise resolves.
- Confirm time zones match.
- Check for duplicate conversion actions in Google Ads. Fix this first if you find it.
- Confirm your counting setting is "one" for lead generation.
- Check auto-tagging is on and your GCLID survives every redirect and every third-party form.
- Compare both platforms against your CRM, which is the only system that knows what a lead was worth.
If you can name the reason for the gap and it holds steady month to month, your tracking is fine and you can stop worrying about it. If the gap moves unpredictably, or if step 3 turned up duplicates, your bidding is being optimized against fiction.
The number that actually matters
All three platforms can be technically correct and still leave you unable to answer the only question your CEO cares about: how much revenue came from paid search.
Conversions are not revenue. A form fill is not a customer. Until your CRM outcomes flow back into Google Ads as offline conversions, Smart Bidding is optimizing for volume of form submissions, and it will happily buy you a thousand worthless ones.
That is a separate problem from the one in this post, and it is the more expensive of the two.
Want someone to just find the leaks?
The Conversion Tracking Audit is a fixed-fee review of exactly this. We go through your Google Ads conversion actions, your GA4 configuration, your tag manager container and your form and booking paths, then hand you a written list of what is double counted, what is missing, and what is silently broken, in the order worth fixing.
No retainer, no call required first. $599, and you keep the findings whether or not you ever work with us.
Get the Conversion Tracking Audit
Only want the Google Ads account reviewed? The Google Ads Audit covers campaign structure, bidding and settings on its own.
