Why Google Ads and your CRM report different conversions

Google Ads reports 40 conversions this month. The CRM shows 12 leads from paid search. Neither number is a bug, and neither one is lying. They disagree because they're counting two different things, in two different ways, on two different clocks.

This gap shows up on nearly every B2B SaaS team that runs paid search, and it produces a familiar argument: marketing quotes the Google number, sales quotes the CRM number, and finance trusts neither. The fix is to understand why the two systems can't agree, then connect them so one explains the other.

They define "conversion" differently

Start with the word itself. In Google Ads, a conversion is whatever action you told it to count: a form submission, a demo request, a button click, sometimes a page view. In the CRM, a conversion is usually a record that a human touched: a lead that got qualified, an opportunity that got created.

So Google counts the form fill. The CRM counts the form fill that survived qualification, wasn't a competitor or a job seeker, and became a real lead. The 28-record gap between 40 and 12 is mostly junk that Google correctly counted as a form submission and the CRM correctly discarded.

They use different attribution models

Google Ads credits conversions using its own model, applied only to interactions Google can see. By default that leans toward last-click within Google's own ecosystem. The CRM applies whatever model it's set to, often first-touch or last-touch across every channel.

When a prospect clicks a Google ad, then later arrives through an organic search and fills the form, Google may still claim the conversion under its lookback logic while the CRM credits organic as the source on the record. Same person, same form, two different owners. Why models split credit differently is the whole reason two honest systems land on two different answers.

They use different windows

Google's conversion window and the CRM's lookback window are rarely the same length. If Google counts any conversion within 30 days of a click and the CRM stamps the source at the moment the record was created, the two systems are measuring different slices of time around the same event. Attribution windows that don't match will never reconcile, because they're eligible to credit different touchpoints.

Google counts clicks and estimates; the CRM counts records

Three mechanics push Google's number above the CRM's:

Cross-device. Google can connect a click on someone's phone to a conversion on their laptop when they're signed in to Google. The CRM sees one form submission and has no idea the earlier mobile click was the same person.

View-through and modeled conversions. Google credits some conversions to ads that were seen but not clicked, and fills gaps left by consent and cookie loss with modeled, estimated conversions. Those estimates are real numbers in the Google report and have no matching row anywhere in the CRM.

De-duplication. Google can count more than one conversion from a single click if you're tracking multiple conversion actions. The CRM collapses everything to one lead record per person.

They run on different clocks

Google credits a conversion to the date of the click. The CRM often stamps activity on the date the deal was created or closed. In a business with a 90-day sales cycle, an ad clicked in March and a deal created in June land in different months entirely. Compare the two reports for a single month and they'll disagree purely on timing, even when every underlying event matches.

So which number should you trust?

Neither, on its own. They answer different questions.

Google Ads is closest to the click. It's the right tool for optimizing bids, keywords, and creative, because it sees the full ad-interaction picture including views and cross-device clicks that never reach the CRM.

The CRM is closest to the money. It knows which of those clicks became a qualified lead, an opportunity, and eventually revenue. For deciding whether paid search actually pays back, the CRM's connection to pipeline and revenue is what counts.

The mistake is treating either as the single source of truth. Google will always report more because it counts earlier, wider, and with estimates. The CRM will always report less because it only counts records that survived. Both are right about their own question.

Close the gap by connecting the two

The reconciliation isn't a spreadsheet that forces the numbers to match. It's a connection that ties each ad click to the CRM record it eventually became.

That means passing the click identifier and UTM data into the CRM at form submission, then using identity resolution to connect the anonymous ad click to the known lead once they convert. Do that, and a Google conversion stops being a number you argue about and becomes a specific person you can follow all the way to closed-won. The two systems still report different totals, but now you can trace exactly why, click by click.

This is a piece of the larger cross-channel attribution problem: every platform counts credit in its own favor, and the only fix is measuring from one connected dataset instead of several self-interested ones.

Cascayd ties each ad click to the CRM record and revenue it became, so the two reports stop being an argument and turn into one traceable story. Try Cascayd for free.