What is the best attribution model for ecommerce?
Usually there is no single best model. In GA4, data-driven is the sensible default and last click the check. But every model only splits credit among touches GA4 recorded, so if most revenue comes from one-touch journeys, the choice barely changes the answer.
By Joris van Huët, Founder & CEOUpdated 6 min read
Usually there is no single best model. If you use GA4, start with data-driven, which is already its default, and keep last click as a check. But every model only splits credit among the touches GA4 recorded. If most of your revenue comes from one-touch journeys, the model you pick barely changes the answer.
What one store's data shows
The usual answer is to pick a multi-touch or data-driven model and move on. One store's export shows how little of the money that choice can reach.
One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.
| What the export shows | Share of revenue | Source cell |
|---|---|---|
| Journeys with 1 touch (0.5 days to buy) | 79.5% | Journeys sheet, 1 touch row |
| Journeys with two or three touches (12.5 days to buy) | 12.2% | Journeys sheet, two to three touches row |
| Journeys with four to nine touches (16.9 days to buy) | 5.4% | Journeys sheet, four to nine touches row |
| Journeys with ten or more touches (16.0 days to buy) | 3.0% | Journeys sheet, ten or more touches row |
| Direct, in last click, first click and touched views | 57.7% | Channels sheet, Direct row |
An attribution model decides how to split credit across the touches in one journey. A journey with one touch has nothing to split. In the Journeys sheet, those journeys hold 79.5% of revenue, and every model hands that money to the same place.
The model debate is about the rest. Journeys with two or more touches hold 20.6% of revenue in the Journeys sheet (12.2% + 5.4% + 3.0%), about a fifth.
Those longer journeys are also where the variety lives: 3,656 of the export's 3,670 distinct path sequences. Distinct paths are not purchases, so this says nothing about how many people bought. It says the tail is long and thin.
Direct holds 57.7% of revenue in this store whether the Channels sheet credits the last click, the first click or every channel touched. GA4's own models credit Direct only when a path is direct from start to finish. So for this one store, choosing a model is an argument about a fifth of revenue.
What the export cannot show is what any touch caused. A model splits credit for sales that happened, using only the touches GA4 recorded. Even data-driven, which compares paths with and without a touch, cannot weigh one GA4 never saw.
Why does the usual answer mislead?
The usual answer runs like this: last click is unfair to the channels that start journeys, so use data-driven. That is half right.
GA4 now offers three models in its Attribution reports: data-driven, paid and organic last click, and Google paid channels last click. First click, linear, time decay and position-based went in November 2023. So if you report in GA4, "which model is best" is mostly a choice between two families.
Data-driven is smarter than a rule. GA4 says it uses your account's data and a counterfactual approach to judge which touchpoints most likely drive key events. But it only weighs touchpoints GA4 recorded, and Direct still gets nothing unless a whole path is Direct.
Then the tools disagree by design. Google Ads uses data-driven by default for most conversion actions, and the model you choose there changes how automated bidding optimises. Yet GA4 uses last click for every Google Ads conversion based on a GA4 key event. Same sale, two models, two answers, both from Google.
So the honest answer to "which model is best" is "best for which job". For reporting to your team, pick one model, write it down and stop switching. For Google Ads bidding, Google itself recommends testing a new non-last click model before you trust it. For moving budget between channels, a credit-splitting model is the wrong tool. That job needs evidence of cause, such as a holdout.
What can an attribution model not tell you?
It cannot tell you what a channel caused. Last click and data-driven both share out credit for sales that already happened. To learn what a channel adds, run a holdout: pause it in some regions, keep it running in others, and compare total sales.
It cannot see what never reached GA4. Most ads seen and not clicked, a podcast, a shop shelf, a friend at dinner: none of them leaves a touchpoint. Those journeys begin where GA4's records begin, often at Direct.
It cannot tell you whether a channel pays. Credit is not profit. You still need your margin and your spend to know whether a channel earns its keep.
What to do this week
- Write down which model each tool uses. In GA4, go to Admin, click Events under Data display, then Attribution settings, and note the reporting attribution model. Pass: every report your team shares names its model. Fail: people compare numbers made under different models and argue about the gap.
- Compare two models side by side. In GA4, open Advertising, go to Attribution > Attribution models and compare data-driven with paid and organic last click for purchases. Pass: each channel's revenue moves little, so the model is not your problem. Fail: a channel swings hard, so test it before you move budget.
- Size what any model can touch. Open Attribution paths under Attribution, filter the path length to one touchpoint and compare that purchase revenue with the total. Pass: multi-touch paths carry a real share, so comparing models is worth your time. Fail: single-touch paths carry most of it, so fix tagging and run a holdout instead.
Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework
Sources, 1 October 2026: Get started with attribution (Google); Default channel group (Google); Key event attribution models report (Google); About attribution models (Google); Key events attribution paths report (Google).
Related answers
Frequently asked questions
Is data-driven attribution better than last click?
Often, for reporting. Data-driven uses your account's data to split credit across touches, while last click hands everything to the final click. Neither can credit a touch GA4 never recorded, and neither replaces a test of what a channel caused.Can I still use first click or linear attribution in GA4?
No. Google says the first click, linear, time decay and position-based models are no longer available as of November 2023. GA4's Attribution reports now offer data-driven, paid and organic last click, and Google paid channels last click.Which attribution model does Google Ads use by default?
Data-driven, for most conversion actions, says Google's help. Last click is still supported. The model you set on a conversion action also changes how automated bid strategies optimise, so treat a switch as a test, not a setting.
Go deeper: Incrementality testing, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Keep reading
Terms in this article
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- Attribution ModelAn Attribution Model defines how credit for conversions is assigned to marketing touchpoints. It dictates how marketing channels receive credit for sales.
- Attribution ReportAttribution Report shows which touchpoints or channels receive credit for a conversion. It identifies which campaigns drive desired actions.
- ConversionConversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
- CounterfactualCounterfactual is a hypothetical outcome that would have occurred if a subject had received a different treatment.
- Google AdsGoogle Ads is an online advertising platform where advertisers bid to display ads, service offerings, and product listings.
- IncrementalityIncrementality measures the true causal impact of a marketing campaign. It quantifies the additional conversions or revenue directly from that activity.
- Incrementality TestingIncrementality Testing measures the additional impact of a marketing campaign. It compares exposed and control groups to determine causal effect.