How Multi-Touch Attribution Works in Google Analytics 4

What GA4's attribution reports show, the two models it still offers, and why Google Analytics sees only part of what drives your sales.

Google Analytics is the first place most marketers look to see which channels drive conversions. It's free, it sits on nearly every website, and its attribution reports promise to show how search, social, email and other channels work together on the way to a sale.

Those reports are useful, as long as you know what they can and can't see. This guide covers how attribution works in Google Analytics 4 today and where it fits in a wider measurement plan.

From Universal Analytics to GA4

Multi-touch attribution assigns credit for a conversion to several of the touchpoints that came before it, rather than giving it all to one. Customers rarely convert on their first visit. They come from a search, return through a social ad, open an email and buy days later.

Universal Analytics showed this through its Multi-Channel Funnels reports, including assisted conversions: how often a channel appeared earlier in a path, not just at the end. Google Analytics 4, launched in October 2020, replaced Universal Analytics entirely in 2023, when Universal Analytics stopped processing data.

What GA4's attribution reports show

GA4 keeps its attribution tools in the Advertising section. The name is misleading: the reports cover every channel, including organic search, email and direct traffic, not only paid ads.

Two reports matter most:

  • Model comparison. Shows conversions and revenue by channel under two attribution models side by side, so you can see which channels gain or lose credit when you switch from last click to data-driven.
  • Conversion paths. Shows the sequences of channels that led to conversions, and how credit is distributed across early, middle and late touchpoints.

You choose the reporting attribution model and the lookback window in the property settings. In 2024, Google renamed conversions in GA4 to "key events," so you'll see both terms in reports and documentation.

The two models GA4 still offers

In 2023 Google removed the first-click, linear, time-decay and position-based models from GA4. Two remain:

  • Data-driven attribution, the default, uses Google's machine learning to estimate how each touchpoint changes the probability of converting, based on your property's own paths. It's more nuanced than a fixed rule, but you can't see or adjust how it weighs channels.
  • Last click gives all the credit to the last channel the visitor came through before converting.

Comparing the two is still useful. A channel that gains a lot of credit under data-driven attribution is probably doing work earlier in the journey that last click ignores.

What Google Analytics can't see

GA4 measures your website and app. It builds paths from the visits it can record, and that sets its limits.

  • Views without clicks. A social ad someone scrolls past, a connected TV commercial, a podcast ad or a billboard leaves no visit, so it gets no credit. The sale usually goes to whatever the person clicked last, often search or direct.
  • Lost tracking. Consent choices, ad blockers, Safari and Firefox tracking protections and Apple's App Tracking Transparency all reduce what GA4 can link together. Cross-device journeys break when users aren't signed in.
  • Offline activity. Unique URLs, QR codes, promo codes and call tracking can connect some offline campaigns to site visits, but they capture only the people who use them.
  • Correlation, not causation. Attribution divides credit among touchpoints that were present before a sale. It doesn't tell you whether the sale would have happened anyway. Brand search and retargeting often look strong in GA4 because they reach people who were already going to buy.
  • Its own position. Google Analytics is made by the company that sells search and YouTube ads. That doesn't make its numbers wrong, but it is one more reason to check them against an independent view.

We cover these gaps in more detail in GA4 is only showing you part of the story and why GA4 can't measure CTV.

Get the basics right first

Attribution reports are only as good as the traffic data underneath them. Three habits make the biggest difference:

  • Tag every campaign consistently. UTM parameters decide which channel a visit is assigned to. Inconsistent source and medium names split one channel into several or push paid traffic into "direct" or "unassigned."
  • Link your ad accounts. Connecting Google Ads and other supported platforms brings cost and campaign detail into GA4, so you can compare credit with spend.
  • Define key events carefully. Count the actions that matter to the business, such as purchases, qualified leads and sign-ups, not every page view or button click. Too many key events blur the paths you're trying to read.

How to use GA4 attribution well

GA4 is a good tool for the questions it was built for:

  1. Understand on-site behavior. Which channels bring visitors who convert, and how long the path to conversion takes.
  2. Compare models, not just numbers. The difference between last click and data-driven credit is a clue about which channels assist rather than close.
  3. Optimize within digital channels. Campaign-level and landing-page decisions inside channels GA4 can see clearly.
  4. Don't set the overall budget from it. Deciding how much to spend on TV versus search, or whether a channel is adding sales at all, needs a method that sees every channel, including the ones nobody clicks.

That method is incrementality measurement, usually through marketing mix modeling, which works from aggregate spend and sales data and needs no cookies or user tracking. Used together, GA4 shows how people move through your site, and incrementality measurement shows what each channel actually adds to sales.

Frequently asked questions

Which attribution models does GA4 offer?

Two. Data-driven attribution, the default, assigns credit using Google's model of how each touchpoint changes the chance of converting. Last click gives all the credit to the final channel before the conversion. Google removed the first-click, linear, time-decay and position-based models in 2023.

Where are the attribution reports in GA4?

They sit in the Advertising section, even though they cover all channels, not just paid ads. The model comparison report shows how credit changes between models, and the conversion paths report shows the sequences of channels that led to conversions.

What happened to Universal Analytics and its multi-channel funnel reports?

Universal Analytics stopped processing data in 2023 and was replaced by GA4. Its multi-channel funnels and assisted-conversion reports were replaced by GA4's attribution reports, which serve a similar purpose.

Can Google Analytics track offline marketing?

Only indirectly. You can connect some offline activity to the site with unique URLs, QR codes, promo codes or call tracking, but GA4 can't see exposure to TV, radio, print or billboards. Those channels are better measured with marketing mix modeling.

Is GA4 attribution accurate?

It is a reasonable view of the paths it can see, but it misses ad views without clicks, many cross-device journeys and visitors who decline tracking. It also shows which channels were present before a conversion, not which ones caused it.

Free Download

The 2027 Attribution Playbook.

A practical field guide for proving impact when the cookie is gone, the click is weaker, and finance needs one answer. Includes examples from retail, CPG, financial services, and subscription brands, plus the frameworks behind a finance-ready measurement rollout.

  • → Why spike-based TV attribution broke — and what replaced it
  • → The halo effect on Amazon, Walmart, Target, and Chewy
  • → How to align Finance and Marketing on one source of truth
  • → A practical framework for validating incremental lift
Download the playbook →
2027
Attribution Playbook

Proving Impact in a Cookie-less, Click-less World

A field guide for marketing leaders who need to defend every dollar.