GA4 attribution explained (and what changed from Universal Analytics)
Attribution is simply how GA4 decides which marketing channel gets the credit when someone converts. Get it wrong and you defund the channels that are actually working. Here is how GA4 does it, what changed from Universal Analytics, and how to read it without being misled.
What attribution actually means
Most buyers touch several channels before they convert: an Instagram ad, a Google search, a newsletter, then a direct visit. Attribution is the rule GA4 uses to split, or assign, the credit for that sale across those touches. The model you choose changes which channel looks like the hero, which is why it matters for budget.
The big change from Universal Analytics
Universal Analytics defaulted to last non-direct click, giving all the credit to the final non-direct channel. GA4 defaults to a data-driven model that uses Google’s machine learning to distribute credit across the journey based on what actually influenced conversions. GA4 is also event-based rather than session-based, so it counts interactions differently, which is one reason your GA4 numbers never match your old UA ones.
The models you can choose
- Data-driven (default): credit shared across touchpoints based on modelled impact.
- Last click: all credit to the final channel (paid and organic, ignoring direct where possible).
- Paid and organic last click, and Google Paid Channels last click for Ads-focused reporting.
First-click, linear, time-decay and position-based models were retired from GA4’s reports, so the practical choice now is data-driven versus a last-click view.
Is your attribution data even reliable?
If channels are landing in Direct or paid is showing as organic, your attribution is lying to you. Run a free 10-second check of your live tags.
Check my tracking freeWhere people get misled
Two traps catch most people. First, Direct is a dumping ground: when the source signal is missing (a stripped click ID, an app-to-browser hop, an untagged email), GA4 files the visit as Direct, which makes it look bigger than it is. Second, the conversion credit in reports can differ from Google Ads, because Ads uses its own modelling and lookback windows. Neither is a bug; you just need to know which report answers which question.
How to use it well
Pick one model as your source of truth for budget decisions and stick to it, usually data-driven. Make sure every channel is tagged consistently so credit is not lost to Direct. Check your lookback window suits your sales cycle. And when a channel looks weak, confirm it is genuinely underperforming rather than simply losing its attribution signal before you cut its budget.
Get your attribution audited and fixed
We check how your conversions are really being credited, tell you what’s wrong in plain English, and give a fixed quote. The audit comes off any build.
See how it works & pricingRelated guides
Why paid shows as organic in GA4 What is server-side tracking? Fix duplicate conversions in GA4 PricingDatahound are analytics and ad-tracking specialists. We fix broken GA4, Google Ads, Meta and server-side tracking. See pricing · All guides · Back to datahound.agency
GA4 attribution: common questions
What attribution model does GA4 use by default?
GA4 defaults to the data-driven attribution model, which uses Google’s machine learning to share conversion credit across the touchpoints that actually influenced the conversion, rather than giving it all to the last click.
Why don’t GA4 and Google Ads conversions match?
They measure differently. Google Ads uses its own attribution and lookback windows and can count conversions by click date, while GA4 reports by its own model. Some difference is normal; a large gap usually points to a tracking or tagging problem.
Can I still use last-click attribution in GA4?
Yes. GA4 offers last-click style models (paid and organic last click, and Google paid channels last click). First-click, linear, time-decay and position-based models were removed from GA4 reporting.