Why Google Ads, GA4, and CRM Data Don’t Match

If you’ve ever tried to reconcile your Google Ads, Meta Ads, GA4, and CRM data, you already know the struggle. The numbers never match. So how do you decide where to invest your PPC budget? And more importantly, how do you optimize for actual business impact instead of chasing phantom metrics?

Many marketers assume the fix is better tracking, cleaner UTMs, or a more advanced analytics stack. But the real culprit is often something deeper. Let’s call it the attribution trap.

A whole generation of marketers has been trained to be data-driven. The logic is simple: configure your tools correctly, and they’ll tell you what’s working. Just follow the numbers. But attribution can quickly become misleading. Without the right framework, you end up allocating budget based on incomplete insights, sometimes with damaging consequences.

Let’s pause here. Attribution assigns conversion credit to channels. That’s useful, but it can’t tell you which conversions your ads actually caused. That distinction might sound academic, but it’s the key to fixing your measurement problem. So let’s explore why attribution fails, how to triangulate your existing data, and whether incrementality testing is the right next step.

Why Ads, Analytics, and CRM Numbers Never Align

Before you try to fix anything, understand this: aligning ad networks, GA4, and your CRM is fundamentally impossible. These systems were built for different purposes, use different methodologies, and measure different moments in the customer journey.

Imagine a customer who clicks a Meta ad, gets retargeted on YouTube, then searches for your brand on Google before converting , all within seven days. With default attribution windows, both Meta and Google Ads will report one conversion. GA4 and your CRM will show only one, likely crediting Google Ads paid search. Did Meta invent a duplicate conversion? No. Meta simply has no visibility into Google Ads interactions.

Meanwhile, GA4 and your CRM will almost certainly ignore Meta’s role. Should you follow those “insights” and shift budget from Meta to branded search? Probably not.

The structural differences go deeper. Attribution dates differ: ad platforms assign conversions to the day of the click, while GA4 and CRMs use the conversion date. For long customer journeys, that creates persistent discrepancies. Cross-device behavior adds another layer: a user who clicks a Google ad on mobile, returns on desktop via SEO, and converts will generate a conversion across tools, but Google Ads and your CRM will disagree on the source because the CRM can’t merge mobile and desktop visitors. Privacy restrictions , ad blockers, browser tracking prevention, and cookie consent banners , mean a large share of conversions goes unmeasured. Ad networks may fill gaps with modeled conversions, but your CRM still won’t see the real source.

The latter issues are fixable with better configuration, especially server-side tagging, offline conversion imports, and consistent UTMs. But the structural divergence remains. You cannot expect 100% correlation.

The Single Source of Truth Trap

Once teams accept that numbers differ, the next move is often choosing a single source of truth , GA4 or the CRM. That’s where the attribution trap closes.

Every tool follows an attribution model , first-click, last-click, linear, time decay, or data-driven. And every model is fundamentally limited. Last-click rewards the final touchpoint, typically branded search, and systematically undervalues demand generation. First-click does the opposite, rewarding discovery while ignoring the touchpoints that moved someone to convert. Linear and time-decay feel balanced but are largely arbitrary. Data-driven models sound sophisticated, but they’re black boxes. If they were truly reliable, platforms would offer more transparency.

What happens when you rely on a single source? If you trust your CRM, you’ll be driven by last-click attribution and focus on branded search. A few years later, demand may dry up despite strong CRM numbers. If you rely solely on ad platform data, you’ll report inflated results , think 2x, 3x, or 4x more revenue than finance sees. You’ll increase budgets while finance tells you to stop. GA4 sounds like the grown-up in the room, but it only measures on-site behavior. Awareness campaigns designed for views or ad recall don’t necessarily generate website visits.

Once you realize all tools have blind spots, someone will inevitably suggest incrementality. Did this campaign cause conversions that wouldn’t have happened otherwise?

Incrementality Testing: The Perfect Solution?

Incrementality measures the results generated because of your campaign , conversions that wouldn’t have existed without the ad. Think of two parallel universes: the gap between the world where the ad ran and the world where it didn’t is your incremental impact. Everything else is activity you would’ve captured anyway.

This matters more than it seems. A significant share of reported campaign conversions , especially in retargeting and branded search , comes from people who would’ve converted regardless. They were already in-market and close to a decision. Showing them an ad and claiming credit is what attribution does. Incrementality testing measures how much of that credit is real.

For budget decisions, that distinction is everything. A retargeting campaign reporting strong ROAS through attribution might deliver almost no incremental value. Cut it, and conversions barely move. Keep it, and you’re paying for the illusion of performance.

Common approaches include geo holdout (run campaigns in some regions, go dark in others), audience holdout (exclude a percentage of your target audience from seeing ads), and time-based testing (pause a campaign and measure the impact). Each has trade-offs. Audience holdout relies on ad platform data, so you can only compare campaigns within the same network. Time-based testing risks hurting performance if the campaign was incremental.

Is incrementality right for you? If you’re running smaller budgets , under roughly €1 million per month , it’s often not actionable. Reliable tests require large amounts of data. But you can use shortcuts, especially for branded search. Check auction insights to see if competitors are bidding on your brand. If they are, you probably need branded search campaigns to capture demand you created. If not, you can likely pause them and let SEO handle it.

Triangulation: The Actionable Alternative

So if attribution is flawed and incrementality is mostly for top-tier advertisers, what’s left? Triangulation.

Use the tools you already have while staying aware of their flaws. Educate clients and leadership so they don’t blindly follow a single source of truth. Here’s how it works in practice.

Start with your CRM or CMS. Those systems record actual deals and revenue. Treat every other number as an attempt to explain them. When Google Ads and Meta Ads report $50K in revenue while Shopify shows $35K, Shopify reflects reality. It’s also the only system that can reliably tell you whether a conversion came from a new or existing customer. That lets you measure nCAC (new customer acquisition cost) and anchor budget decisions around customers who otherwise wouldn’t have found you.

Then superimpose your customer journey onto ad platform results. The gap between the two represents the ad platforms’ interpretation of their contribution. Your job is to understand each campaign in context and identify where deduplication is needed. For example, if you run both Demand Gen and Meta retargeting campaigns, there’s almost certainly overlap. Time-based incrementality tests, if available, can help determine which channel performs better.

To improve your triangulation, adjust attribution windows for long customer journeys. Segment campaigns around specific stages and use smaller windows with micro-conversions when configured properly. Track ratios between ad platform conversions and CRM/CMS data over time. If the ratios hold, your measurement framework is stable. If they break, investigate , there may be an incrementality insight hiding there.

Triangulation won’t give you a single clean number. But it will give you a defensible, consistent framework for making decisions. That’s far more valuable than false precision.

Welcome to the Real World

The teams that waste the most time on measurement are the ones trying to force three systems to produce the same number, or searching for the attribution model that finally feels fair. The teams that make the best decisions accept that reality is more complex than a single source of truth. They build the data skills needed to reflect that complexity.

So make sure your decision-making process is as close to reality as possible , and embrace the question marks.