Attribution basics: which channel actually drove the sale?

If you are running multi-channel campaigns, you have probably stared at your dashboard and asked: how do I know which marketing channel is working? It is the most challenging question in digital marketing.

In a simple world, a user would click an ad and purchase a product immediately. But in the real world, the buyer's journey is messy. A prospect might find your B2B SaaS platform through an organic search click on Monday, read a case study, click a LinkedIn retargeting ad on Wednesday, receive an email newsletter on Friday, and finally type your URL directly into their browser to buy on Sunday.

If you look at your sales data, which channel gets the credit for that purchase? Google Search, LinkedIn Ads, Email, or Direct?

The answer depends entirely on your attribution model. Understanding how these models work is the difference between scaling a profitable channel and accidentally shutting down the very campaigns that feed your funnel.

The flaws of last-click attribution

For decades, the default attribution model on the web was last-click (or last-interaction). Under this model, the very last touchpoint before the conversion receives 100% of the credit.

In the example above, Direct traffic gets all the credit. Google Search, LinkedIn, and Email get zero.

This creates a dangerous blind spot for founders:

• Undervaluing awareness: Last-click makes it look like your top-of-funnel marketing (like organic SEO or top-of-funnel social) is generating no return.
• Overvaluing closing channels: It makes your bottom-of-funnel channels (like branded search ads or email campaigns) look incredibly profitable.
• Bad budget decisions: If you look only at last-click metrics, you might decide to stop spending money on LinkedIn retargeting because it shows zero conversions. But when you turn off LinkedIn, your direct conversions drop, because the ads were nurturing the buyers who eventually typed your URL.

Last-click measures who touched the ball last, not who built the play.

The marketing funnel: awareness vs. closing channels

To understand which channel is working, you must assign roles to your marketing channels. They cannot all be evaluated on the same timeline.

I divide channels into three categories:

1. Awareness channels (The openers)

These channels introduce your brand to new prospects. They drive the first visit.

• Primary channels: Organic search (non-branded SEO), top-of-funnel paid social ads, cold outreach.
• Metric to watch: First-user acquisition.

2. Nurturing channels (The assists)

These channels keep your brand top-of-mind while the prospect evaluates the purchase.

• Primary channels: Retargeting ads, email newsletters, case study pages.
• Metric to watch: Multi-session engagement and return visits.

3. Closing channels (The closers)

These channels capture the prospect when they are ready to buy.

• Primary channels: Branded search ads, direct traffic, affiliate links.
• Metric to watch: Session conversion rate.

Understanding GA4's data-driven attribution

To move away from the binary limitations of first-click and last-click, Google Analytics 4 defaults to Data-Driven Attribution (DDA).

DDA uses machine learning algorithms to evaluate the paths of both converting and non-converting users. The algorithm analyzes how the presence or absence of a specific touchpoint (such as a retargeting ad) impacts the likelihood of a conversion.

For example, if the model observes that user paths containing a LinkedIn ad click convert 2x more often than paths without it, the algorithm will allocate fractional credit (e.g., 0.4 conversions) to the LinkedIn campaign, even if the user eventually bought via direct traffic.

This provides a more balanced, realistic view of how your channels interact. However, DDA has a catch: it requires consistent conversion volume to build accurate models. If your site only receives ten conversions a month, GA4's algorithm does not have enough data to calculate probability accurately, and the model will revert to a layout that looks very similar to last-click.

How to find your true winners

To get a clear picture of which channel is driving sales, do not rely on a single acquisition table. Use GA4's attribution tools to compare models:

1. Navigate to Advertising in the left menu of GA4.
2. Select Model Comparison under Attribution.
3. Compare the Last click model to the First click model.
4. Look for the channels with the largest discrepancies.

If your Paid Social channel shows 10 conversions under Last Click but 50 conversions under First Click, you know that Paid Social is a powerful awareness driver. It is initiating the customer relationship, even if other channels are closing the deal. If you cut that paid social budget, your closing channel numbers will eventually dry up.

Conversely, if a channel shows high Last Click conversions but near-zero First Click conversions, it is a pure closer. It is not introducing new users; it is simply capturing the demand generated by other sources.

Choosing a source of truth

Marketing attribution is an approximation, not a perfect science. There is no single "correct" model.

The key is consistency. Choose one attribution model (usually GA4's blended Data-Driven model) to make your macro budgeting decisions, and stick to it across months. Comparing a last-click report from January to a data-driven report from February will show changes that are purely structural, not real.

If you need help configuring your attribution lookback windows or building a custom channel grouping that accurately reflects your business model, a focused analytics setup and mapping project can clarify your dashboard. Stop evaluating your marketing channels in isolation; look at the entire journey to identify where your real buyers are won.