Everything can look good inside an advertising platform: campaigns generate conversions, ROAS increases, and cost per lead stays within the target KPI. Yet the business’s financial performance may barely change. The reason is that an advertising platform answers one question, while the business often needs an answer to a completely different one.
Google Ads or Meta Ads show which conversions they are able to attribute to their campaigns. The business, however, needs to understand how much money remains after advertising costs, cost of goods sold, discounts, returns, logistics, and other variable expenses. That is why evaluating marketing solely by the number of conversions or ROAS is not enough.
Why advertising platforms do not show the full picture
Let us consider a typical customer journey. A person sees the brand on Instagram for the first time. A few days later, they search for the company name on Google, click on a paid ad, but do not make a purchase. Another week later, they return through organic search and place an order.
Which channel generated the sale? Meta may count its own interaction. Google Ads may count its interaction. Google Analytics may distribute conversion value across several touchpoints depending on the selected attribution model. This is why the total number of “sales” reported across advertising platforms can sometimes exceed the actual number of orders recorded in the CRM.
This is not necessarily an error. Different systems evaluate the same customer journey using different models. Google defines attribution as the process of assigning credit for conversions to different advertising interactions along the user journey. GA4 also offers data‑driven attribution, which uses data from a specific property to assess the contribution of different touchpoints. Therefore, the answer to the question “Which channel generated the sale?” does not always come down to a single click.
ROAS does not necessarily mean profitability
ROAS is a useful metric. However, it only compares advertising revenue with advertising spend. For example:
- advertising spend – UAH 100,000;
- revenue attributed to advertising – UAH 500,000;
- ROAS – 500%.
At first glance, the result looks strong. But suppose the average gross margin on the products is 25%. In that case, out of UAH 500,000 in revenue, approximately UAH 125,000 in gross profit remains to cover other expenses. After subtracting UAH 100,000 in advertising costs, only UAH 25,000 remains – before accounting for logistics, payroll, returns, banking fees, and operating expenses.
This means that two advertising channels with the same ROAS can have completely different unit economics if they sell different product categories, operate with different margins, or acquire customers with different repeat‑purchase value.
Google Ads itself allows advertisers to assign different conversion values and optimize campaigns not simply for conversion volume, but for their actual business value. Google’s documentation explicitly mentions revenue and profit margins as examples of conversion value signals that can be used for bidding optimization.
How to compare marketing channels correctly
Problems arise when all channels are evaluated using the same metric without considering context. SEO, branded search, Performance Max, Meta campaigns focused on acquiring new audiences, and email marketing operate at different stages of the customer journey.
For example, branded Google Ads campaigns may show an exceptionally high ROAS because users already know the company and are actively searching for the brand. This does not mean that the campaign generated all of that demand on its own. Similarly, Meta Ads may appear weaker under a last‑click attribution model while still being one of the primary sources of initial brand discovery.
That is why the analysis should consider at least the following chain: spend → new customers → revenue → margin → repeat purchases. Only then does it become clear whether a channel simply generates low‑cost conversions or actually creates economically valuable customers.
Why attribution changes the outcome
The last‑click model is simple: the conversion is attributed to the final channel before the purchase. However, a modern customer journey often consists of multiple interactions. A user may see an ad, read an article, return through Google, receive an email, and only then make a purchase.
In GA4, data‑driven attribution analyzes both converting and non‑converting paths and attempts to estimate the contribution of individual advertising touchpoints to the probability of a target action. This does not mean that any single attribution model provides an absolute truth. Its purpose is to help the business interpret a complex customer journey more accurately.
That is why deciding to turn off a channel simply because it “generates almost no sales under the last‑click model” may be premature. Before making such a decision, it is worth reviewing assisted interactions along the conversion path, branded demand, new‑user acquisition, purchase cycle length, and cohort behavior.
What data should be sent to advertising platforms
An algorithm optimizes toward the signal it receives. If Google Ads treats all leads as having the same value, the system does not know that one lead resulted in a UAH 200,000 deal while another turned out to be unqualified.
The same principle applies to ecommerce. If the advertising platform receives only the purchase event but not the correct transaction value, it will optimize toward the number of orders rather than their actual value. This is why tracking quality directly affects the quality of automated optimization.
Google recommends using conversion values to evaluate ROI and to bid based on the value of conversions. Enhanced Conversions can further supplement standard conversion tracking by using hashed first‑party data to improve conversion measurement accuracy.
How to build a marketing analytics system
The purpose of analytics is not to create yet another attractive dashboard. It should help the business make decisions. For example:
- can the Google Ads budget be increased by 30%;
- does Meta actually acquire new customers;
- which campaign generates the most profit rather than the most revenue;
- which product categories should be promoted more aggressively;
- how many months does it take to recover CAC;
- can the business afford a higher customer acquisition cost because of repeat purchases.
To answer these questions, advertising data must be combined with information about actual orders or deals.
In practice, the system may look like this: Google Ads / Meta Ads → GA4 → CRM or ecommerce platform → financial data → dashboard. This gives marketers visibility not only into CPC and CPA, while the business no longer sees only total revenue. Both sides gain a shared frame of reference.
The channel that generates the most sales is not always the most valuable
Consider two channels:
- The first generates 300 orders per month with low margins and almost no repeat purchases.
- The second generates only 180 orders, but it attracts new customers who return more often and purchase higher‑margin products.
If you look only at the number of orders, the first channel appears to perform better. If you take contribution margin and LTV into account, the conclusion may change. This is where marketing stops being a collection of advertising platforms and becomes part of the business’s financial model.
How to tell whether your budget is actually being “wasted”
Not every campaign with a low ROAS is automatically a bad campaign. Likewise, a high ROAS does not guarantee that the budget is being used efficiently.
Potential warning signs include situations where advertising campaigns optimize toward micro‑conversions rather than sales, duplicate conversions are recorded in the system, CRM data and advertising platform data differ significantly, campaigns assign the same value to leads of very different quality, or budgets are scaled without taking profit margins into account.
That is why a marketing audit should not begin with the question, “Which campaign should we turn off?” It should begin with: “Do we have enough accurate data to evaluate whether this campaign is actually profitable?”. If the answer is no, measurement should be fixed first. Only after that should the budget be reallocated.
From advertising metrics to business economics
Effective marketing in 2026 is not only about generating low‑cost clicks and conversions. A competitive advantage emerges when a business can connect advertising spend with actual financial outcomes.
At that point, scaling decisions are no longer based on the fact that “ROAS looks good in the ad platform,” but on a clear understanding of:
- how much it costs to acquire a new customer;
- how much margin that customer generates;
- whether the customer makes repeat purchases;
- how long it takes to recover the acquisition cost;
- which channel creates the most value for the business.
This kind of system makes it possible to distinguish marketing channels that genuinely support growth from those that merely produce attractive numbers in reports.
Not sure which marketing channels are actually driving business results? Message us on WhatsApp – we’ll help you analyze the data, evaluate channel performance, and understand where your marketing budget can work more effectively.
