End-to-end marketing analytics: a step-by-step tutorial from UTM to dashboard

There is a quote that marketers have been repeating for over a century. It is attributed to the American retailer John Wanamaker: “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.” This was said at the end of the 19th century, when ads were published in newspapers and tracking tools simply didn’t exist. But here is the strange part: in 2026, despite having Google Ads, Meta, TikTok, pixels, cookies, and dozens of analytics platforms, most marketers are still living with the exact same problem. Except now, they are looking at three different dashboards, each showing its own ROAS. And they never add up. This is exactly where end-to-end marketing analytics comes into play.

What is End-to-End Analytics and How Does It Differ from GA4?

End-to-end analytics is a system that tracks the entire customer journey: from the very first ad impression to the actual payment in the CRM or at the cash register. Not just to the click, not just to the form submission, but all the way to the money in the bank. It sounds logical. Yet, most companies stop much earlier.

From First Click to Payment: The Full Customer Journey

A typical buyer’s journey in e-commerce or B2B looks something like this:

Without end-to-end analytics, each of these stages lives in its own silo, completely cut off from the others. GA4 tracks website conversions but has no idea how many of them actually turn into closed deals. Your CRM tracks the deals but doesn’t know where the customers came from. Meanwhile, Meta Ads is utterly convinced that every single sale is its own personal achievement.

End-to-end marketing analytics solves this by pulling all data into a single source of truth, tracing every single touchpoint from the very first interaction to the last.

Why GA4 + Ad Dashboards Do Not Equal True Analytics

GA4 is an excellent tool for analyzing on-site user behavior. However, it comes with several fundamental limitations that are rarely talked about openly:

The situation with ad platforms is even trickier. Each platform (Google, Meta, TikTok) counts conversions based on its own rules and within its own attribution window. If a customer sees an ad on Meta on Tuesday but buys via Google on Friday, both platforms will claim that conversion as their own. As a result, the combined ROAS from all your dashboards can end up being several times higher than your company’s actual revenue.

This is exactly why marketing analytics in Ukraine (and globally) is increasingly being built around a centralized data warehouse where all sources converge, rather than around GA4 alone. It is the only place where you can see the truth.

In short: GA4 shows what is happening on your website; end-to-end analytics shows what is happening to your business. The difference is fundamental.

3 Conversion Attribution Models: Which One to Choose

Conversion attribution is the set of rules a system uses to decide which channel gets credit for a sale. It might sound like a technical detail, but it actually dictates exactly where your budget will go next month.

Last Click: Simple but Flawed

The Last Click model awards 100% of the credit to the very last channel the customer interacted with before converting. This is the default setting in most systems.

Picture this scenario: a user sees a banner on Meta $\rightarrow$ googles the brand name $\rightarrow$ clicks through from Google and buys. Last Click will attribute all the credit to Google. Meta gets zero. Next month, the paid social specialist gets a budget cut, even though it was their ad that triggered the entire chain reaction.

When Last Click is acceptable: For simple products with a short decision-making cycle where the customer buys on the first touchpoint.

Linear and Time Decay

These are more sophisticated models that distribute credit across multiple touchpoints:

None of them are perfect; they are all simplifications of reality. However, in most scenarios, any of them is fairer than Last Click.

Data-Driven Attribution (Google)

Data-Driven Attribution is a machine-learning-based model that Google launched across Google Ads and GA4. It analyzes the actual conversion paths in your account and distributes credit based on which channels statistically increased the probability of a purchase.

It sounds perfect, and it genuinely works better than rule-based models. But there is a catch: it operates entirely within the Google ecosystem, completely ignoring offline touchpoints, phone calls, and CRM data. This means it is insufficient for comprehensive end-to-end analytics; it serves merely as one of many incoming signals.

The takeaway on attribution: For most businesses, the optimal strategy is to leverage Data-Driven attribution within Google, while simultaneously building your own attribution model inside a centralized data warehouse that pulls together CRM data, offline interactions, and all marketing channels at once.

How to Aggregate Data Across All Marketing Channels

Before you can build dashboards and calculate ROI, your data must be collected correctly. This is the most tedious yet critical part of the entire system. If you make a mistake here, all subsequent reports might look beautiful, but they will be feeding you lies.

Proper UTM Tagging Setup

UTM tags are parameters appended to a URL that allow GA4 and other systems to identify exactly where a user came from.

A standard UTM structure looks like this (example):

[https://iwis.io/?utm_source=google&utm_medium=cpc&utm_campaign=brand_search&utm_content=ad1]

The main problem most companies face is the lack of a unified standard. One marketer writes utm_source=Google, another uses utm_source=google, and a third goes with utm_source=Google_Search. In GA4, these are treated as three completely different sources. As a result, your data becomes fragmented, and the reports never match up.

Connecting Ad Platforms (Google Ads, Meta, TikTok)

Each platform has its own API (Application Programming Interface) through which you can extract data on spend, impressions, clicks, and conversions. There are two main approaches to doing this:

Comparison of approaches:

Integrating with CRM and Sales Systems

This is the most critical and most frequently skipped step. Without CRM data, you can see your leads, but you have no idea how many of them actually turned into paying customers or how much revenue they generated.

This division allows each tier to see exactly what is needed for decision-making without losing details where they are critical.

Real Channel ROI: How to Calculate It Right

The ROI of marketing channels is what you get after comparing channel costs against the actual revenue generated by the customers that specific channel brought in. It is not what the ad dashboard shows you.

The formula is as follows:

However, “revenue” here refers to actual payments recorded in the CRM. And this is exactly where most companies run into unpleasant surprises.

Three Typical Mistakes When Calculating ROI:

• Counting leads instead of money. A channel might generate a high volume of low-quality leads. It may look effective based on CPL (Cost Per Lead) but turn out to be unprofitable in terms of actual revenue.

• Ignoring team costs. Channel expenses must include not only the ad budget but also salaries or agency fees for campaign management.

• Comparing different attribution windows. Google Ads counts conversions within 30 days, while Meta tracks them within 7 days after a click or 1 day after a view. If you compare them head-to-head, the numbers will be inaccurate.

The right approach: All conversions are pulled from the CRM based on the deal closure date, mapped to the source via Client ID or first-touch UTM tags, and only then compared to the ad dashboard spend for the same period.

This is exactly what end-to-end analytics delivers: instead of three optimistic reports from three different dashboards, you get one honest report from a single source of truth.

Industry Case Study: How an E-commerce Business Cut Ad Spend Without Losing Sales

Let’s look at a generalized scenario that reflects a typical e-commerce situation after implementing end-to-end analytics.

The Baseline: An online clothing store with a monthly ad budget of around $15,000. Active channels: Google Search, Meta, and email newsletters. GA4 showed stable conversion rates, and all three channels reported a positive ROAS. Yet, profits were stagnant, and the owner couldn’t understand why.

What End-to-End Analytics Revealed: After connecting the CRM to a single data warehouse and setting up first-touch attribution, the picture changed dramatically:

• Google Brand Search (ads targeting brand-name queries) showed an 8x ROAS in the dashboard. However, the analysis revealed that 74% of these users would have purchased anyway—they were already in the database or came through email campaigns.

• Meta generated the most leads, but the lead-to-payment conversion rate was twice as low as Google Performance Max.

• The Email channel was misattributed as “direct” in GA4 due to missing UTM tags in some emails, effectively disappearing from reports.

The Changes Made: The Brand Search budget was partially reallocated to Performance Max and new prospecting audiences in Meta. Email campaigns received proper UTM tagging. For Meta, the strategy shifted: instead of conversion campaigns, the focus turned to awareness, backed by remarketing through Google.

The Result After 3 Months: The ad budget was reduced by 22% by cutting underperforming campaigns. The volume of actual payments remained steady, while CAC (Customer Acquisition Cost) dropped by 18%.

Key Insight: The problem wasn’t that the channels were performing poorly. The problem was that without end-to-end analytics, it was impossible to see which ones were actually driving results.

End-to-End Analytics Readiness Checklist

Before building dashboards and connecting data warehouses, it’s crucial to check the basics. Here is the minimum set of prerequisites without which end-to-end analytics simply won’t work: