Why GA4 Should Not Be Your Source Of Truth

Treating any single analytics platform as the truth is how marketing teams end up arguing about numbers instead of making decisions. Every platform measures a different thing, and none of them measures revenue.

Why the numbers never agree

Your ad platform reports more conversions than your analytics. Your analytics reports more than your CRM. Your CRM reports more than your accounting system.

None of them are broken. They are counting differently.

Ad platforms count conversions they believe they influenced, including view-through, and they attribute to themselves. Analytics counts sessions and events in the browser, and loses everything blocked, consented away or arriving on a device it cannot connect to the first one. The CRM counts records a human created. Accounting counts money that arrived.

Each is measuring a real thing. The mistake is expecting them to match.

What analytics is actually good at

Behaviour on your own property. Which pages people land on, where they go, where they stop, what proportion complete an action, how that changes when you change something.

That is genuinely valuable and it is what the platform is built for. Use it for that.

What it is not good at

Revenue truth. It knows about the events it was told about. It does not know which deal closed, which one was refunded, or which customer stayed for three years.

Cross-device journeys. Someone sees an ad on a phone, researches on a laptop and calls from the office. Analytics sees fragments.

Offline conversion. For any business where the sale happens on a phone call or in person, the most important step is invisible unless you deliberately feed it back.

Fair channel credit. Attribution models are assumptions with a user interface. They are useful for comparison over time, not for settling which channel deserves the money.

The hierarchy that works

Money is the truth. What actually landed in the bank, from which customer, is the only number that cannot be argued with. It is also the slowest and the least granular.

The CRM is the operational truth. Every lead, its source, and what happened to it. This is where the answer to "what does a customer cost" actually lives, and it is where most businesses have the biggest gap.

Analytics is the behavioural truth. What people do on your site.

Ad platforms are the optimisation signal. Directionally useful, structurally biased toward themselves, and necessary because the algorithm needs the feedback.

Use each for what it is reliable at, and stop trying to reconcile them.

The one thing worth fixing first

Get the source of every lead into your CRM, and get the outcome of every lead back out.

Almost every business with an attribution problem actually has this problem. Without it, you can see what enquiries cost and nothing about what customers cost, which means every budget decision is a guess.

It is not a technical project. It is a form field and a habit.

Feed the outcome back

Once the CRM knows which leads became customers, send that back to the ad platforms as conversion data.

This is the single highest-return analytics work available to most accounts. The algorithm optimises toward whatever you tell it success is. Tell it "form submitted" and it will find you form submissions. Tell it "became a customer worth this much" and it goes looking for a different person.

Then hold one number

Pick the number the business is actually run on. For most, that is cost per customer, and after that, return on spend.

Report it from the same source, the same way, every month. A number measured consistently and imperfectly is more useful than a perfect number measured differently each time.