GA4 Attribution Models Explained

Key Takeaway

Data-driven attribution distributes credit across the full conversion path using machine learning, but it requires sufficient conversion volume to be reliable. Properties with fewer than 300 conversions per month should consider last-click as a more stable baseline.
Intermediate

GA4 attribution is not one switch. It combines a reporting attribution model with a key event lookback window, and the two settings answer different questions. If you do not review both, channel reporting can look inconsistent even when the implementation itself is sound.

7 or 30 days

Available acquisition key event lookback window options

GA4 Admin > Attribution Settings

30 to 60, or 90 days

Available lookback windows for other key events

GA4 Admin > Attribution Settings

Data-driven
Paid and organic last click
How it works
Uses property data to distribute credit across eligible touchpoints on the path to a key event.
Gives all credit to the last eligible non-direct touchpoint before the key event.
When to use
Useful when you want GA4's default, path-based view of contribution across channels.
Useful when you need a simpler last-touch reporting frame or a closer comparison with some ad-platform views.
Direct traffic treatment
Direct is excluded unless the full path is direct.
Direct is excluded unless the full path is direct.
Channel bias
Can give partial credit to assist channels that appear earlier in the path.
Concentrates credit on the final eligible touchpoint, which can reduce visibility for assist channels.

Which attribution models GA4 still offers

GA4 no longer offers the older first-click, linear, time-decay, or position-based models in standard attribution reporting. The models you typically compare today are data-driven and paid and organic last click. In some Google Ads contexts, Google paid channels last click is also relevant.

That matters because many older attribution articles still describe models that are no longer available in GA4. If your reporting playbook was written from theUniversal Analytics era, review it before using it as a governance standard.

The two lookback windows in GA4

  • Acquisition key event lookback window: Applies tofirst_openandfirst_visit. The available options are 7 days or 30 days.
  • Other key event lookback window: Applies to all other key events. The available options are 30 to 60, or 90 days.

Google's documentation notes that the selected key event lookback window also applies to session attribution. That is why session-based channel reporting can shift when teams change attribution settings, and why the difference betweenUser acquisition and Traffic acquisitionreports can suddenly look bigger.

How the lookback window changes the story

Imagine a user first visits via paid social, returns later through email, and converts weeks after the first session. If the relevant touchpoint falls outside the configured lookback window, it cannot receive credit in the attribution model, no matter how persuasive that first click was in reality.

This is why longer buying cycles often need longer review windows. It is not because GA4 becomes more truthful at 90 days by default. It is because the property is allowing more eligible touchpoints into the analysis.

Need a faster review of attribution settings, lookback windows, and campaign-governance issues before you trust the channel reports?

Where teams misread attribution reports

One common mistake is comparing user-scoped or session-scoped traffic dimensions to event-scoped conversion reports as if they obey the same attribution logic. Google's documentation distinguishes these scopes. User-scoped and session-scoped dimensions use paid and organic last click, while event-scoped traffic dimensions reflect the property's selected reporting attribution model.

Another mistake is forcing GA4 to match an ad platform exactly. GA4 is useful for channel QA and cross-channel analysis, but it is still a separate measurement system with separate rules, identity constraints, and consent inputs — the same dynamic that makesGA4 and Google Ads numbers divergeby design. Consent gaps also feed intoconversion modelling, which adjusts attributed conversions when raw signals are missing.

Choosing settings that fit the business

  • Short sales cycles often work with shorter windows because fewer materially relevant touches happen far from conversion.
  • Longer B2B or considered-purchase journeys usually need a wider review window so earlier sessions remain eligible for credit.
  • If your team reports heavily on assists and channel contribution, data-driven often gives a broader view than last click.
  • If your team needs a simpler operational comparison, last click can be easier to explain, but it narrows the story — especially forpaid social channels like Metathat often appear earlier in the path than the converting touchpoint.

Validating your GA4 attribution configuration

Use this plan to confirm your attribution model and lookback windows are correctly set for your business type.

Validate

  • Go to Admin > Attribution Settings and note the current reporting attribution model and both lookback window values
  • Review whether the team understands which reports are event-scoped versus user-scoped or session-scoped
  • Review whether the current lookback window fits the observed buying cycle, not just a default setting
  • In GA4 Advertising > Attribution > Model Comparison, compare Data-Driven vs Last Click to see how credit distribution changes between models
  • If GA4 conversions are imported into Google Ads, check whether the Google Ads conversion window is consistent with the GA4 lookback setting

Fix

  • Set the reporting attribution model deliberately rather than leaving it undocumented
  • Adjust the key event lookback window when the current window is clearly shorter than the real buying cycle
  • Align Google Ads conversion window settings with the GA4 lookback period to ensure Smart Bidding uses consistent attribution
  • Use the Model Comparison tool to document the before/after impact of any lookback window change before communicating revised channel performance to stakeholders

Watch for

  • Upper-funnel channels persistently showing little credit despite strong evidence they initiate demand, which may indicate an overly narrow reporting frame
  • A sudden shift in channel credit after an admin change, which may be a settings change rather than a market change
  • A change in conversion totals after updating the lookback window, this is expected and reflects different touchpoints now qualifying for credit

Conversion windows checklist

  • Current lookback window setting reviewed in Admin > Attribution Settings
  • Reporting attribution model is documented and understood by the team using the reports
  • Window is set based on the real buying cycle for the business, not just a default
  • Google Ads conversion window settings reviewed after changing GA4 lookback period
  • Model Comparison tool used to compare how attribution changes between settings

Audit your attribution model and conversion windows

GA4 Audits helps surface attribution settings, channel-governance issues, and tagging problems that make conversion reports harder to trust.

Audit findings should be reviewed by a qualified analyst before they are used for major reporting, media, or implementation decisions. Review your findings

GA4 Audits Team

GA4 Audits Team

Analytics Engineering

Specialising in GA4 architecture, consent mode implementation, and multi-layer audit frameworks.

Share