What Attribution Model Does Google Analytics Use?

The short answer is that Google Analytics 4 uses a Data-Driven Attribution (DDA) model by default. This is a significant shift from the old Universal Analytics, which defaulted to the "Last Non-Direct Click" model, and it changes how your marketing channels get credit for conversions. We’ll show you exactly what this means, where to check the settings, and how to compare different attribution models to get a complete picture of your performance.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Google Analytics 4's Default: The Data-Driven Attribution Model

Unlike older models that used fixed rules to assign credit, the Data-Driven Attribution model in GA4 uses machine learning to analyze the entire customer journey. It looks at all the touchpoints a user has with your business - paid ads, social media posts, organic search results, email campaigns - and determines how much credit each step should get for leading to a conversion. It considers factors like time between interactions, the device used, and the type of ad creative.

Let's make this real with an example. Imagine a customer does a Google search and clicks on one of your blog posts (Organic Search). A week later, they see a retargeting ad on Facebook and click it to browse your products (Paid Social). Finally, two days later, they type your website's URL directly into their browser, an email reminder pops up and they close it, then they complete their purchase through the direct website visit (Direct).

• Universal Analytics, using its default "Last Non-Direct Click" model, would have given 100% of the credit to "Paid Social". Why? Because the very last touchpoint a user had before buying, "direct," doesn't get counted, so what UA sees is Paid Social at the end of the line. The initial blog post gets no credit.
• GA4's Data-Driven model, however, looks at this whole path. Its algorithm might determine that the Organic Search touchpoint was critical for awareness and that Paid Social played a key role in rekindling interest. It might assign something like 45% credit to Paid Social, 40% to Organic Search, and the remaining 15% to other factors it observed.

This approach gives you a much more nuanced and realistic view of what’s actually working. It helps you recognize the value of your top-of-funnel marketing efforts that warm leads up, instead of only crediting the final touchpoint that a user remembers.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

A Quick Tour of All Attribution Models in Google Analytics 4

While Data-Driven is the default, GA4 still allows you to use or compare results using other rule-based models. Understanding these is helpful for running different "what-if" scenarios and for communicating with stakeholders who may be used to the old ways of measurement. You don't have to choose only one, and later we'll show you how to compare models next to each other in one screen.

These models are all "cross-channel," meaning they look at and distribute credit across all your marketing channels (Email, Social, Paid, Organic, etc.).

Last Click

This is the simplest - and often, the most misleading - model. It gives 100% of the credit for the conversion to the very last channel the person clicked on. In our example above, the 'Direct' channel would get all the credit, and both the social ad and organic search would get nothing. Its simplicity is its biggest flaw, as it completely ignores everything a customer did before that final click.

First Click

First Click is the exact opposite of Last Click. It gives 100% of the credit to the very first channel that brought a user to your site. This model is useful for understanding which channels are best at creating initial awareness and introducing people to your brand. So, for our example up top, the 'Organic Search' channel that got folks on our radar from viewing that blog post would get total credit.

Linear

You can think of this as the "everyone gets a trophy" model. It splits credit equally among all the touchpoints in a conversion path. If a user had four interactions before converting (e.g., Organic Search → Facebook Ad → Email link → Direct purchase), each one would receive 25% of the credit. It’s fair, but it assumes every touchpoint is equally valuable, which is rarely a realistic business scenario.

Position-Based

This model is a hybrid that's a bit more sophisticated than linear. It assigns 40% of the credit to the first touchpoint, 40% to the last touchpoint, and divides the remaining 20% evenly among any interactions in the middle. It acknowledges the critical roles of both introducing the brand (first click) and closing the deal (last click), and doesn't throw a ton of credit around. The assumption here, of course, is that first and last touches deserve most of the credit, but there are great counterarguments to be made when multi-touch campaigns go against that thought pattern.

Graphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Build Your First Agent

Time Decay

The Time Decay model assigns more credit to the touchpoints that happen closer in time to the conversion. The click from last Friday gets more credit than the click from three weeks ago. This model is particularly useful for businesses with longer consideration cycles, where recent interactions are often more influential in the final decision-making process.

Heads Up: Cross-channel vs. 'Ads-preferred' Models

This is a small but important detail. When selecting models, you'll see a second category for some of them labeled “Google Ads-preferred.” For example, you’ll see "Last click" and then "Ads-preferred last click."

Here's what that means: The "Ads-preferred" version gives 100% of the credit to a Google Ads click if one appears anywhere in the path. This ignores every other channel. For instance, if a customer clicked on an Organic social media post, then on a LinkedIn Paid Post, and followed up to make their purchase with a search ad, the only channel to get credit is Google Ads.

For a fair and holistic view of channel performance, you should almost always stick to the cross-channel models. These models evaluate all your channels on equal footing, which is the whole point of multichannel attribution.

How to Find and Change Attribution Settings in GA4

Want to check which model your GA4 property is using or perhaps change it? It’s buried in the settings, but easy to update once you know where to look. Let's make it quick.

• 1. Go to Admin: Log in to your Google Analytics 4 account and click the 'Admin' gear icon in the bottom-left corner of the screen.
• 2. Find Attribution Settings: In the 'Property' column (the middle one), scroll down to the 'Data Display' section and click on 'Attribution Settings'.
• 3. Change Your Model: You’ll see a dropdown menu under “Reporting attribution model.” You can change the default model for anyone on your team from here whenever you want. We'll also mention that Data-Driven is not available for all accounts yet and for those smaller users, Last Click - Cross-Channel will likely be the default setup. Don't worry, you can always go back and review things using Data Driven with a few clicks in the "Attribution" workspace.

On this same page, you can also adjust the "Lookback window." This determines how far back in time Google attributes credit to your channels for conversions and other acquisition signals. It should match your common 'purchase cycle' or however long you feel information about the customer is relevant.

AI Agents for Marketing

Crash Course

Graphed

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Download the PDF

Checking Different Scenarios with the Model Comparison Tool

In our opinion, one of the most powerful features in all of Google Analytics lives in this tiny workspace nobody uses. What this Attribution section and particularly the Model Comparison Tool allows you to do is not just look at your default, set-in-stone Attribution model, but it’s an opportunity to reexamine things through a totally different lens any time you want with past and current data.

This tool is invaluable for showing your team or clients how the perception of channel performance changes based on the attribution model selected.

• 1. Go to Advertising: Open your GA4 property and select 'Advertising' from the reports list or in the main right-hand navigation menu.
• 2. Find Model Comparison: Within the Advertising section, click 'Attribution' in the sub-menu if it's available, and then proceed with a click 'Model comparison' button.
• 3. Compare and Analyze: In this report, you can add up to three different attribution models and compare them side-by-side. For example, set one column to Data Driven to see what Google says is your true baseline. Then to see who starts the conversation, set your middle column to 'first-click cross-channel' so no data point can be thrown away if it includes an Ad click. Lastly, set that third column to Time-Decay to account for the timeliness of each click coming in.

The results and new baseline data view you create by manipulating those models side-by-side tell incredible stories in and of themselves about your marketing's performance that you never would have seen through one model.

Final Thoughts

In sum, Google Analytics 4 moves the conversation forward by making the intelligent, AI-driven Data-Driven Attribution model the default. This is a huge step up from last-click measurement, empowering you to see which channels are truly contributing value across the entire customer journey, not just at the end. Get familiar with the other models and especially with the comparison tool to truly elevate your analysis and back your strategic decisions with solid data.

Understanding attribution in GA4 is a great start, but the real power comes when you combine that data with the rest of your marketing and sales stack - like performance from Paid Social channels, sales cycle, LTV data, costs, and information held in your CRM. Piecing all of that together manually every week is tedious and slow. To address this, we built Graphed to connect all your data sources in one place. You can ask for a dashboard in plain English, like "Compare my Facebook Ads cost against Shopify revenue and tie to new contacts in Klaviyo by campaign" in 30 seconds to see the full, real-time picture of what is truly working without needing 3 people or a spreadsheet to support the analysis!